Systems, analyzers, controllers, and associated methods to enhance fluid production of refining operations

Abstract

Embodiments of systems and methods include receiving, by a machine learning model, feedstock data from one or more refinery equipment providing feedstock to a hydrotreater. The machine learning model is trained on historical feedstock data and historical data of operation of the hydrotreater. The machine learning model has a target product specification for a first product from the hydrotreater and predicts a property of the feedstock based on the feedstock data. The machine learning model generates an output indicative of an adjustment to a parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification based on the feedstock data, the historical feedstock data, and historical data of operation of the hydrotreater.

Description

SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to, and the benefit of U.S. Provisional Application No. 63 / 786,014, filed April 9, 2025, titled SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONS, U.S. Provisional Application No. 63 / 778,798, filed March 27, 2025, titled SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONS, U.S. Patent Application No. 18 / 948,759, filed November 15, 2024, titled “SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID SEPARATION FOR DISTILLATION OPERATIONS,” U.S. Provisional Application No. 63 / 660,196, filed June 14, 2024, titled “SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONS,” U.S. Provisional Application No. 63 / 658,825, filed June 11, 2024, titled “SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONS,” and U.S. Provisional Application No. 63 / 655,589, filed June 3, 2024, titled “SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS TO ENHANCE FLUID PRODUCTION OF REFINING OPERATIONS,” the disclosures of which are incorporated herein by reference in their entireties.FIELD OF DISCLOSURE

[0002] The disclosure herein relates to systems, analyzers, controllers, and associated methods to enhance fluid production for refining operations and, more particularly, to systems, analyzers, controllers, and associated methods to enhance fluid production of refining operations and sub-operations using machine learning models during the refining operations and sub-operations.BACKGROUND

[0003] Many and varied operations are executed continuously and simultaneously at a refinery. Each operation affects each subsequent operation or sub-operation. For example,if the product from a first operation is produced based on maximizing a first factor, for example, the research octane number of gasoline, the production of that particular product affects further downstream operations. Additionally, further upstream operations may not be suited for production of that particular product or may, at least, cause other operations to produce that particular product in an inefficient manner. Additionally, a variety of starting feedstock are utilized at a refinery. Even further, different batches or portions of one feedstock may vary over time, for example, different portions of a feedstock may include different properties and / or composition over time. Optimization (in other words, efficient and accurate production of targeted products) of such operations and feedstock poses a significant challenge when attempting to meet a target product objective, particularly over a period time, as equipment and materials used in the operation change over time. Such problems pose further difficulties since updating one operation affects every other operation at the refinery.

[0004] Controllers and monitoring devices may be utilized at a refinery to attempt to optimize (in other words, efficiently and accurately produce targeted products) those operations. However, those controllers and monitoring devices utilize algorithms that require expert personnel and that take extended amounts of time to execute. For example, first-principle models require expert personnel to ensure that the first-principle model is accurately calculating some formula, in other words, expert personnel are required to maintain the first-principle model. Further still, the equipment utilized at one refinery may experience a different service or maintenance cycle than equipment at another refinery. Such factors further complicate any attempt at uniform optimization at a plurality of refineries.SUMMARY

[0005] Thus, in view of the foregoing, Applicant has recognized these problems and others in the art, and has recognized a need for systems, analyzers, controllers, and associated methods for enhancing fluid production for refinery operations. Particularly, the present disclosure relates to systems, analyzers, controllers, and associated methods to enhance fluid production of refining operations and sub-operations using machine learning models during the refining operations and sub-operations. Such fluids may include hydrocarbons and / or renewable hydrocarbons and fluid production may include, for example, production of transportation fuel, among other products.

[0006] The disclosure herein provides embodiments of systems, analyzers, controllers, and associated methods for enhancing fluid production of ongoing and / or continuous refining operations, as well as refining sub-operations. Such systems, analyzers, controllers, and associated methods may include obtaining data corresponding to a refinery operation from one or more sources, such as sensors, analyzers, refining equipment, devices, and / or other sources. The data, along with, in some embodiments, a target product, may then be applied to a machine learning model to produce an output indicative of or including parameters that indicate settings for the devices and / or refining equipment to be set to, to accurately achieve or produce the targeted product.

[0007] Accordingly, an embodiment of the disclosure is directed to a method including receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater and receiving current operating parameters of the hydrotreater. The machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater. The machine learning model is trained to meet a sulfur target, while maximizing aromatic saturation constrained by available hydrogen gas feed rates. The method further includes generating, by the machine learning model, a target aromatic saturation based on the sulfur target, the feedstock data, and available hydrogen gas feed rates and generating an adjustment to the current operating parameters of the hydrotreater based on the target aromatic saturation of the feedstocks within the hydrotreater reactor, the sulfur target, the feedstock data, and available hydrogen gas feed rates.

[0008] Yet another embodiment of the disclosure is directed to a method including receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater and receiving current operating parameters of the hydrotreater. The machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater. The method further includes generating, by the machine learning model, a prediction of a state of catalysts within a hydrotreater reactor based on the current operating parameters of the hydrotreater and generating an adjustment to the current operating parameters of the hydrotreater based on the predicted state of the catalysts within the hydrotreater reactor and the feedstock data. In some applications, a machine learning model optimizes sulfur removal or maximizes aromatic saturation based on the predicted state of the catalysts.

[0009] In another embodiment, a method includes receiving, by a machine learning model, a first feedstock data from one or more refinery equipment providing a first feedstock to be hydrotreated, the first feedstock data indicative of one or more properties or composition of the first feedstock. The machine learning model has a target product specification for the first feedstock and receives current data of operational parameters of a first hydrotreater and a second hydrotreater. The machine learning model is trained on historical first feedstock data, historical data of operational parameters of a first hydrotreater, and historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater and the second hydrotreater. The machine learning model generates a recommendation of one of the first hydrotreater or the second hydrotreater to send the first feedstock to be hydrotreated and generates an output indicative of an adjustment to a parameter of operation of the recommended one of the first hydrotreater or the second hydrotreater to hydrotreat the first feedstock to accurately produce the product to the target product specification for the first feedstock based on the feedstock data, the historical feedstock data, historical data of operational parameters of a first hydrotreater, and historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater and the second hydrotreater. In some applications, a machine learning model optimizes sulfur removal or maximizes aromatic saturation as the target product specification based on the historical data and current data of the operational parameters of the first hydrotreater and the second hydrotreater.

[0010] An embodiment according to this disclosure is a method including a machine learning model generating a value for products of a hydrocracker, a value for hydrogen gas fed to the hydrocracker, a value for feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker. A product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker. The machine learning model was trained with historical data including feedstock data, operational data, and product data of the hydrocracker and the fluid catalytic cracker and generates an adjustment to an operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to increase a combined value of the hydrocracker and the fluid catalytic cracker, based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrocracker, and the value for the feedstocks fed to the hydrocracker and the fluid catalyticcracker. In some applications, the value may include the operating costs of the hydrocracker and the fluid catalytic cracker.

[0011] In another embodiment, a method includes a machine learning model that generates a value for products of a hydrotreater, a value for hydrogen gas fed to the hydrotreater, and a value for feedstocks fed to the hydrotreater, a value for the products of a hydrocracker, a value for the hydrogen gas fed to the hydrocracker, a value for the feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker. A product of the hydrotreater includes hydrotreated gas oil fed to the fluid catalytic cracker. A product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker. The machine learning model was trained with historical data including feedstock data, operational data, and product data of the hydrotreater, the hydrocracker, and the fluid catalytic cracker and generates an adjustment to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to increase a combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrotreater and the hydrocracker, and the value for the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker. In some applications, the value may include the operating costs of the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

[0012] Another embodiment of the disclosure is directed to a system for enhancing fluid production for a hydrotreatment. The system may include a hydrotreater to receive a feedstock and refine the feedstock into a refined feedstock. The system may include a plurality of sensors to measure a parameter associated with the hydrotreater and each positioned at one of (a) proximate the hydrotreater or (b) within the hydrotreater. The system may include a plurality of refining operation control devices each positioned proximate and downstream or upstream of the hydrotreater and to control an aspect and / or property of fluid flowing to or from the hydrotreater. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the hydrotreater. The system may include one or more sample analysis assemblies to analyze each collected sample to provide properties of collected samples. The system may include a hydrotreater controller in signal communication with the hydrotreater, the plurality of sensors, the plurality of flow control devices, and the one or more sample analysis assemblies, and storing a trained machine learning model. The hydrotreater controller maybe configured to determine an output including predicted properties of the feedstock and parameter settings of the plurality of refining operation control devices and the hydrotreater based on application of one or more of (i) data measured by the plurality of sensors, (ii) data corresponding to analysis from the one or more sample analysis assemblies, or (iii) a target product and corresponding target properties to the trained machine learning model, and adjust one or more of the amount of feedstock or type of feedstock and parameters associated with the plurality of refining operation control devices and hydrotreater based on the output to enhance production of the refined feedstock.

[0013] Another embodiment of the disclosure is directed to a method for enhancing control of a hydrotreatment operation associated with a petroleum refining operation. The method may include supplying feedstock to a hydrotreater associated with the petroleum refining operation, the feedstock having one or more feedstock properties. The method may include analyzing a feedstock sample via a first analyzer to provide feedstock sample properties. The method may include predicting one or more feedstock sample properties associated with the feedstock sample based on (A) the feedstock sample properties and (B) a first output from application of the feedstock sample properties to a first trained machine learning model. The method may include operating the hydrotreater to produce one or more unit materials, the one or more unit materials having one or more unit materials properties, and the one or more unit materials comprising one or more of gas, gasoline, diesel, or i- butanes. The method may include analyzing the unit material sample via a second analyzer to provide unit material sample properties. The method may include predicting one or more unit material sample properties associated with the unit material sample based on (C) the unit material sample properties and (D) a second output from application of the unit material sample properties to a second trained machine learning model. The method may include controlling, during the hydrotreatment operation, based on the feedstock sample properties and the one or more unit material sample properties, one or more of (a) one or more feedstock properties associated with the feedstock supplied to the hydrotreater; (b) one or more unit product materials properties associated with the unit product materials; (c) operation of the hydrotreater; or (d) operation of one or more upstream equipment or downstream equipment, so that the controlling, during the hydrotreatment operation, causes the hydrotreatment operation to produce one or more of (i) one or more intermediate materials each having one or more properties within a range of one or more target properties of the one or more intermediate materials, (ii) one or more unit product materials each having one or more properties within a range of one or more target properties of the one ormore unit product materials, or (iii) one or more downstream materials each having one or more properties within a range of one or more target properties of the one or more downstream materials, thereby to cause the hydrotreatment operation to achieve material outputs that more accurately and responsively converge on one or more of the target properties.

[0014] Still other aspects and advantages of these embodiments and other embodiments, are discussed in detail herein. Moreover, it is to be understood that both the foregoing information and the following detailed description provide merely illustrative examples of various aspects and embodiments, and are intended to provide an overview or framework for understanding the nature and character of the claimed aspects and embodiments. Accordingly, these and other objects, along with advantages and features of the present disclosure herein disclosed, will become apparent through reference to the following description and the accompanying drawings. Furthermore, it is to be understood that the features of the various embodiments described herein are not mutually exclusive and may exist in various combinations and permutations.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] These and other features, aspects, and advantages of the disclosure will become better understood with regard to the following descriptions, claims, and accompanying drawings. It is to be noted, however, that the drawings illustrate only several embodiments of the disclosure and, therefore, are not to be considered limiting of the scope of the disclosure.

[0016] FIG. 1A and FIG. IB are simplified diagrams of a refining control system to enhance fluid production at a refinery, according to an embodiment of the disclosure.

[0017] FIG. 2 is a simplified diagram that illustrates an apparatus for enhance fluid production at a refinery, according to an embodiment of the disclosure.

[0018] FIG. 3 is another simplified diagram of a refining control system to enhance fluid production at a refinery, according to an embodiment of the disclosure.

[0019] FIG. 4 is a simplified diagram that illustrates training of a machine learning model for enhanced fluid production at refinery, according to an embodiment of the disclosure.

[0020] FIG. 5 is a schematic diagram of a FCC control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0021] FIG. 6 is a schematic diagram of an enhanced hydrotreater control system and a distillation control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0022] FIG. 7 is a schematic diagram of a steam control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0023] FIG. 8 is a schematic diagram of hydrogen control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0024] FIG. 9 is a schematic diagram of a feed control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0025] FIG. 10 is a schematic diagram of a gasoline pool control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0026] FIG. 11 is a schematic diagram of a diesel pool control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0027] FIG. 12A and FIG. 12B are simplified diagrams of control systems to enhance to enhance fluid production at refinery, according to an embodiment of the disclosure.

[0028] FIG. 13 is a flow chart illustrating enhanced fluid production at a refinery, according to an embodiment of the disclosure.

[0029] FIG. 14 is a schematic diagram of processes within a refinery, according to an embodiment of the disclosure.

[0030] FIG. 15 is a schematic diagram of a machine learning model that may be used with the various processes within a refinery, according to an embodiment of the disclosure.

[0031] FIG. 16 is a schematic diagram of a hydrotreater to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0032] FIG. 17 is a schematic diagram of a hydrocracker control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure, according to an embodiment of the disclosure.

[0033] FIG. 18 is a flow diagram of a method, according to an embodiment of the disclosure.

[0034] FIG. 19 is a flow diagram of another method, according to an embodiment of the disclosure.

[0035] FIG. 20 is a flow diagram of yet another method, according to an embodiment of the disclosure.

[0036] FIG. 21 is a flow diagram of a method, according to an embodiment of the disclosure.

[0037] FIG. 22 is a flow diagram of another method, according to an embodiment of the disclosure.

[0038] FIG. 23 is a flow diagram of yet another method, according to an embodiment of the disclosure.

[0039] FIG. 24 is a flow diagram of a method, according to an embodiment of the disclosure.

[0040] FIG. 25 is a flow diagram of another method, according to an embodiment of the disclosure.

[0041] FIG. 26 is a flow diagram of a method, according to an embodiment of the disclosure.DETAILED DESCRIPTION

[0042] So that the manner in which the features and advantages of the embodiments of the systems and methods disclosed herein, as well as others that will become apparent, may be understood in more detail, a more particular description of embodiments of systems and methods briefly summarized above may be had by reference to the following detailed description of embodiments thereof, in which one or more are further illustrated in the appended drawings, which form a part of this specification. It is to be noted, however, that the drawings illustrate only various embodiments of the systems and methods disclosed herein and are therefore not to be considered limiting of the scope of the systems and methods disclosed herein as it may include other effective embodiments as well.

[0043] The disclosure herein provides embodiments of systems, analyzers, controllers, and associated methods for enhancing fluid production of ongoing and / or continuous refining operations, as well as refining sub-operations. Such systems, analyzers, controllers, and associated methods may include obtaining data corresponding to a refinery operation from one or more sources, such as sensors, analyzers, refining equipment, refining operation control devices, other devices, and / or other sources. The data, along with, in some embodiments, a target product, may then be applied to a machine learning model to produce an output indicative of or including parameters that indicate settings for the refining operation control devices and / or refining equipment to be set to, to accurately achieve or produce the targeted product. Such an application of data to a trained machine learning model may occur at one or more different layers in the control system of the refinery. Further, a plurality of controllers positioned throughout the refinery may each include aplurality of trained machine learning models that are trained to enhance fluid production of one or more refinery operations or sub-operations.

[0044] In embodiments, the refining operations and / or sub-operations may include the processing, converting, refining, enhancing and / or otherwise altering a fluid via the refining operation or sub-operation. The fluid may be a liquid, vapor, and / or gas and may include a hydrocarbon, and a final product may include a transportation fuel. “Hydrocarbons” or “hydrocarbon fluids” as used herein, may refer to petroleum fluids, renewable fluids, and other hydrocarbon based fluids. “Petroleum fluids” as used herein, may refer to fluid products containing crude oil, petroleum products, natural gas, renewable liquids and / or gasses, and / or distillates or refinery intermediates. For example, crude oil contains a combination of hydrocarbons having different boiling points that exists as a viscous liquid in underground geological formations and at the surface. Petroleum products, for example, may be produced by processing crude oil and other liquids at petroleum refineries, by extracting liquid hydrocarbons at natural gas processing plants, and by producing finished petroleum products at industrial facilities. For example, a petroleum product may include a transportation fuel, among other products. Refinery intermediates, for example, may refer to any refinery hydrocarbon that is not crude oil or a finished petroleum product (such as gasoline), including all refinery output from distillation (for example, distillates or distillation fractions) or from other conversion units. In some non-limiting embodiments of systems and methods, petroleum fluids may include heavy blend crude oil used at a pipeline origination station, natural gas, and / or other types of crude oil, as will be understood by one skilled in the art. Heavy blend crude oil is typically characterized as having an American Petroleum Institute (API) gravity of about 30 degrees or below. In other embodiments, the petroleum fluids may include lighter blend crude oils, for example, having an API gravity of greater than 30 degrees. “Renewable fluids” as used herein, may refer to fluid products containing plant and / or animal derived feedstock. Further, the renewable fluids may be hydrocarbon based. For example, a renewable fluid may be a pyrolysis oil, oleaginous feedstock, biomass derived feedstock, renewable natural gas or other liquids or gasses, as will be understood by those skilled in the art. The API gravity of renewable liquids may vary depending on the type of renewable liquid.

[0045] As used herein “profit” is the potential financial gain from a potential sale of products from a refinery operation less the costs of feedstocks and operational costs of the refinery operation. The potential value of a sale may be based on current market prices for similar products or current prices paid by contracted customers. Costs of feedstocks mayinclude market prices for similar feedstocks as well as the operational costs of upstream processes preparing the feedstocks for use in the refinery operation. Operational costs may include equipment depreciation, employee wages and benefits, and consumables used to process the feedstocks, such as hydrogen, water, catalysts, steam, natural gas, cooling, electricity, and other utilities. For example, a machine learning model may generate adjustments to one or more operational parameters of a refinery process based on the potential profit from the products it produces. A potential profit may be determined from a generated algorithm or generated simulations based on current market pricing and potential ranges of production rates and the corresponding variable production costs. To continue the example, product A may sell for $2 a barrel and may be produced at a rate ranging from 50 barrels a day to 100 barrels a day. Product B may sell for $4 a barrel and be produced at a rate ranging from 10 barrels a day to 20 barrels a day. However, the more product B produced results in less product A being produced, increased utility costs, and accelerated deactivation of the catalysts of the refinery process. All of these variables and more may be managed by the machine learning model to increase the profitability of the refinery process.

[0046] In an embodiment, the systems and methods may include a computing device, apparatus, and / or controller (referred to hereafter as a controller) to obtain various data points and / or parameters to train a machine learning model. Such data may include a historical data set and / or a currently generated data set including an outcome. As used in this application, “current” means happening now or recently enough to still be relevant to an ongoing process. For example, a current sample data may have been generated from a sample obtained several hours ago from a feedstock that is still being processed by a refinery process. In another embodiment, the data set may include a simulated and / or filled- in data set. For example, a refinery may be modeled based on a first-principle model and synthetic or pseudo-data may be generated for a selected time interval (for example, 1 month, 2 months, 6 months, 1 year, or even longer). For such a data set, random perturbations and / or anomalies are used to simulate a data set. In another embodiment, the data may include a partial data set. In such an embodiment, the partial data set, may be filled in via a first principle model and / or a machine learning model. Each data set may include a series of parameters, properties, spectra, and / or other data points associated with a refining operation or process or sub-operation or sub-process. Each data set may also include target parameters, target properties, and / or an outcome and / or target product. Further, in an embodiment where a supervised machine learning model is utilized, eachoutcome may be classified as a positive or negative outcome or marked in a manner to indicate desirability of the outcome. In other embodiments, the function generated by a set of data may indicate a desired outcome, based on a maximum or minimum point in that function, thus enabling a machine learning model to determine desired parameters based on that maximum or minimum, or based on some other factor in other embodiments. In yet another embodiment, a trained machine learning model may learn or be trained based on trends included in the data (in other words, the trained machine learning model may comprise a deep learning model). In another embodiment, any of the trained machine learning models described herein may predict and / or optimize target parameters and / or fluids used within a refinery, refinery operation, or refinery sub-operation.

[0047] Once these data sets have been received by the controller, the controller may pre- process the data. For example, the controller may normalize the data (in other words, remove data points that appear to be outliers), remove data corresponding to abnormal events (for example, data generated during start-up, shut-down, turn-arounds, and / or upsets), remove undesired data, remove invalid measurements, and / or segregate the data set into sequences of contiguous data based on selected time intervals (for example, time intervals of 30 minutes, 1 hour, 2 hours, and / or 3 hours, or more or less than the time intervals listed).

[0048] Once the data has been pre-processed, the controller may begin training a model based on a portion of the data set. For example, the controller may utilize an 80 / 20 training and testing process. Other percentages may be utilized in training, testing, and / or validation. As the model is fed data, the model may compare data received to the outcome (in other words, whether the outcome was desired based on some factor, such as an indicated positive / negative flag or classification, based on some maximum or minimum of a function generated based on the data, or based on a trend within the data). Once the training portion of data has been utilized, the controller may test and / or validate the model using the remaining portion of the data set. If such testing or validation does not achieve a selected error rate or reach some other error and / or accuracy based threshold, then the controller may re-train or refine the model using a different and / or randomized portion of the data set and a remaining portion of the data set for testing. Once a model has reached that threshold, then the controller may output the trained machine learning model for further use.

[0049] Further, a trained machine learning model may be further refined using new data, as such data is generated. Such a refinement may occur while the trained machine learningmodel is in use. In another embodiment, the trained machine learning model may be refined in an offline environment. In another embodiment, two instances of a trained machine learning model may exist, one stored as an offline copy, while the other is utilized during refining operations. In such an embodiment, the offline copy may be refined and, if testing and / or error rating meets a selected threshold, in addition to other factors, then a controller may replace the version currently utilized to the refined version.

[0050] It will be understood that such systems and methods described herein may utilize a number of trained machine learning models. For example, a model may be trained for each specific operation or process, as well as each particular piece of equipment, at a refinery, such as fluid catalytic cracking (FCC) operations or processes, hydrocracking operations or processes, reforming operations or processes, alkylation operations or processes, isomerization operations or processes, hydrotreating operations or processes, distillation operations or processes, blending operations or processes, hydrodeoxygenation operations or processes, steam management, hydrogen coordination or management, absorption, propylene splitting operations or processes, aromatic recovery, sulfur recovery, coker unit operations, feed optimization, IMO blending, hydrodeoxygenation, hydrocracker operations, other blending operations, Residuum Oil Supercritical Extraction operation, solvent deasphalting (SDA) operation, operations or processes for formation of specific fuels, and / or the refining operation or process overall (which may, in an embodiment, utilize outputs from models associated with each sub-operation or sub-process). The use of terms operation and process refers to the steps taken to produce a particular product from a selected feedstock (and, in some embodiments, other inputs). As such, when referring to a particular refining operation or process, the terms “operation” and “process” may be used interchangeably. Further, such models may be trained specifically for equipment at a particular plant or refinery. For example, a FCC unit at a first plant may exhibit different characteristics than that of a FCC unit at a second plant. Thus, a model trained for one may not work for the other and training a model for either FCC unit may include utilization of historical data corresponding to that FCC unit. Various aspects of one model may be utilized to train other models for other similar equipment though.

[0051] Once a model is available, the controller, one or more sub-operation controllers or sub-controllers, and / or one or more operation controllers including a local enhancement or optimization module or circuitry, predictive controls, and / or equipment and device controls may begin optimizing, enhancing, and / or adjusting an operation and / or parameters associated with that operation, in real-time, near real-time, and / or continuously orsubstantially continuously, at a refinery. In such embodiments, the controller may obtain data from a plurality of sensors, a plurality of refining operation control devices (such as flow control devices, temperature control devices, pressure control devices, and / or other device configured to control an aspect of a refining operation), equipment at the refinery (in other words, refining equipment), and / or one or more sample analyzers and / or, in some embodiments, one or more sub-operation controllers or sub-controllers. In another embodiment, one or more operation controllers may obtain such data, as well as target products and / or other factors or parameters from a refinery controller or platform.

[0052] As noted, one input to any of the models described herein may include spectra or properties of feedstock, intermediaries, products or outputs, and / or other fluids or materials utilized in a refinery, determined via one or more of a spectrographic analyzer or a chromatographic analyzer. The controller or controllers may work in conjunction with such an analyzer to further enhance fluid production (for example, transportation fuel, hydrocarbon based fluid products, and / or other fluids produced during a refining operation) of the refining operation or sub-operation. As such, spectrographic analyzers may be calibrated or standardized and results may be obtained in a faster than typical timeframe, thus enabling prompt acquisition of fluid properties. For example, for any operation described herein, the controller may obtain spectrographic or chromatographic analysis of any feedstock utilized, any intermediaries produced, and / or any products produced by first initiating sample collection. Once a sample has been obtained, the controller and / or an analyzer may initiate analysis of the sample.

[0053] Once the controller or controllers has / have obtained data related to each operation and / or analysis of one or more fluids associated with the operation, then the controller may apply such data and analysis to a corresponding machine learning model. The output of the model may indicate adjustment of one or more devices or refining operation control devices and / or refining equipment and / or adjustment of a feedstock or intermediary used in the operation or sub-operation. In some embodiments, the output may include targets and / or properties for a feedstock and / or blend of feedstock. In another embodiment, the output may be in the form of a vector, each component of the vector corresponding to a value associated with a parameter of equipment or a device or refining operation control device. In an embodiment, a refining operation control device may comprise or include a temperature control device (such as a furnace, heat exchanger, condenser, boiler, induction coil, fans, a cooling device, and / or other device capable of adjusting the temperature of a fluid and / or the temperature within refining equipment), a flow control device (such as apump, valve, control valve, and / or other device capable of adjusting the flow rate of a fluid), a pressure control device (such as a compressor, a pump, a let-down station or valve, and / or another device configured to adjust the pressure of a fluid), and / or other devices configured to adjust some aspect of a fluid and / or aspect of refining equipment.

[0054] Once the controller has the output of the model, the controller may adjust the relevant aspects of the refining operation. For example, the controller may adjust components of a blend utilized in a feedstock, settings for various refining operation control devices (such as temperature, pressure, flow rate, and / or another aspect associated with a fluid and / or device), use of hydrogen, recovery of selected fluids or materials, and / or use of other fluids or materials (for example, a catalyst), among other adjustments.

[0055] In another embodiment, the controller may optimize an operation based on the current demand for selected products. For example, for a particular targeted product, selected amounts of feed and / or intermediaries may be utilized, increasing the demand for that feed and / or intermediaries. In other embodiments, demand may be a factor utilized in training a model. For example, a selected product may experience increased demand at varying times or a particular feedstock, used to produce a particular product, may be in high demand. Data indicating such demand may be utilized in the described trained learning models.

[0056] In yet another embodiment, the controller may compare the output of the model to the current properties for a selected operation. Based on that difference of such a comparison, the controller may adjust various aspects of that operation.

[0057] By utilizing the trained machine learning models, the systems and methods described herein may determine specific adjustments to a plurality of operations and parameters specific to equipment at a refinery to accurately and more frequently (as compared to typical adjustment times) reach a target product. Further, such adjustments may increase efficiency of the refinery equipment and / or reduce energy utilized by the refinery equipment, thus reducing cost of the refinery operation. The target product may be based on a number of factors, such as demand and / or price or cost for the product, cost of the product and / or feedstock, and / or based on a target product provided by a refinery controller or platform. Such adjustments may be determined in real-time or near real-time using data from continuous and / or ongoing refinery operations.

[0058] Thus, rather than attempting to adjust operations at a significant delay, a refinery’s operations may be adjusted in-real time, near real-time, or at time intervals shorter than in typical optimization operations (such typical optimization operations including operationsby operating personnel to efficiently and accurately produce a target product). Further, such adjustments may be determined faster than typical adjustments to operations, leading to relevant and timely adjustments. Further, such analysis and adjustment utilizes complex non-linear equations which typically take longer to analyze, however with the use of machine learning, such analysis occurs significantly faster and with comparable accuracy.

[0059] FIG. 1 A and FIG. IB simplified diagrams of a refining control system to enhance fluid production at a refinery, according to an embodiment of the disclosure. As illustrated in FIG. 1A, a refinery 100 may include various refining control operation devices and refining equipment. While selected equipment are illustrated in FIG. 1A, it will be understood by those skilled in the art that additional and / or different equipment may be included in or at a refinery 100, particularly based on the type of feedstock processed at the refinery. For example, the refinery 100 may include a desalter, blending tanks, storage tanks, and / or wastewater treatment units, among other equipment. Further, each unit or equipment at the refinery 100 may be optimized using the machine learning models disclosed herein, using data specific to the equipment from the refinery 100. In other words, the equipment may be operated such that the corresponding refinery operations produce an accurate and / or on-specification target product.

[0060] As illustrated in FIG. 1A, a refinery 100 may include a refinery controller 101 and / or a plurality of operation controllers 102. As will be illustrated in subsequent drawings, additional components may be included, such as a refining enhancer, other circuitry, various other controllers, and / or other computing devices. In an embodiment, the refinery controller 101 and / or the plurality of operation controllers 102 may include, for example, a trained machine learning model, as well as other instructions to adjust various devices and / or operations or processes within the refinery 100. The refinery controller 101 and / or the plurality of operation controllers 102 may connect to or be in signal communication with (a) one or more sensors, meters, transducers, and / or other measurement devices positioned throughout the refinery 100 and / or (b) to the equipment (for example, connected to some control aspect or device associated with the equipment) positioned at the refinery 100. The refinery controller 101 may be configured to receive data via such a connection. Further, the refinery controller 101 may receive such data in real-time or near real-time. In an embodiment, the refinery controller 101 may determine a target product and / or other parameters for a selected period of time. The refinery controller 101 may provide such data to each of the operation controllers 102. In other embodiments, the refinery controller 101 may utilize outputs from each of the operation controllers 102to determine parameters for a target product. In other embodiments, the refinery controller 101 may apply those outputs, as well as other data, to a machine learning model.

[0061] In another embodiment, each of the operation controllers 102 may include a local enhancement circuitry 184, predictive controls circuitry 194, 191, and / or 195, and / or equipment and device controls 199. In embodiments, the circuitry may be a module or instructions. In embodiments, the operation controller 102 may include one or more varying or different predictive controls. For example, as illustrated, one or more of the predictive controls circuitry 191 may include a trained machine learning model 193. The trained machine learning model 193 may be trained for a specific operation and / or piece of equipment and, in some embodiments, may be trained to recognize an adjustment, maximization, and / or optimization for specified factors of the specific operation and / or piece of equipment. As such, the operation controller 102 may include a plurality of predictive control circuitry 191. The operation controls may also include predictive control circuitry 194, which includes a trained machine learning model 196 and / or a first-principle model (and / or, in some embodiments, another type of model). The trained machine learning model 196 may be trained to fill missing data for the first-principle model 198. The operation controller may also include predictive control circuitry 195, which may include a first-principle model 197. The first-principle model 197 may be a model using a known, physics based equation or formulation.

[0062] In an embodiment, the operation controller 102 may include a local enhancement circuitry 184. The local enhancement circuitry 184 may include a trained machine learning model 190 and target setpoint instructions 192. The trained machine learning model 190 may utilize data associated with a specific refining operation and / or the output from each predictive controls circuitry to produce an output. The target setpoint instructions may utilize the output of the trained machine learning model 190 to determine a set of parameters that equipment and / or devices associated with a specific refining operation should be set to, to reach a target product. The operation controller may also include the equipment and device controls 199. The equipment and device controls 199 may cause equipment and / or devices to adjust to the target setpoints.

[0063] As illustrated in Fig. IB, the operation controllers 102 may further be connected to or in signal communication with a sample collection assembly 186 and / or a sample analysis assembly or sample analyzer 188. The sample analyzers 188 may include spectrographic analyzers, standardized spectrographic analyzers, and / or chromatographic analyzers. The type of spectrographic analyzers utilized may include one or more of near-infraredspectroscopic analyzer, a mid-infrared spectroscopic analyzer, a combination of a nearinfrared spectroscopic analyzer and a mid-infrared spectroscopic analyzer, a Raman spectroscopic analyzer, or a nuclear magnetic resonance spectroscopic analyzer. The sample analyzer 188 may analyze received samples and provide corresponding spectra indicating properties or other analysis indicating components and / or properties of the sample. The operation controllers 102 may also be connected to one or more sub-controllers or sub-operation controllers that are positioned or configured to manage selected aspects or operations of the refinery 100.

[0064] The refinery controller 101 and / or operation controllers 102 may include a processor and a memory or non-transitory machine-readable storage medium storing instructions executable by the processor (as illustrated in subsequent drawings). In some examples, the refinery controller 101 and / or the operation controller 102 may be a computing device. The term “computing device” is used herein to refer to any one or all of programmable logic controllers (PLCs), distributed control systems (DCSs), a proportional integral derivative (PID) controller, a DCS-PID controller, programmable automation controllers (PACs), industrial computers, servers, virtual computing device or environment, desktop computers, personal data assistants (PDAs), laptop computers, tablet computers, smart books, palm-top computers, personal computers, smartphones, virtual computing devices, cloud based computing devices, and similar electronic devices equipped with at least a processor and any other physical components necessarily to perform the various operations described herein. Devices such as smartphones, laptop computers, and tablet computers are generally collectively referred to as mobile devices.

[0065] The term “server” or “server device” is used to refer to any computing device capable of functioning as a server, such as a master exchange server, web server, mail server, document server, or any other type of server. A server may be a dedicated computing device or a server module (e.g., an application) hosted by a computing device that causes the computing device to operate as a server. A server module (e.g., server application) may be a full function server module, or a light or secondary server module (e.g., light or secondary server application) that is configured to provide synchronization services among the dynamic databases on computing devices. A light server or secondary server may be a slimmed-down version of server type functionality that can be implemented on a computing device, such as a smart phone, thereby enabling it to function as an Internet server (e.g., an enterprise e-mail server) only to the extent necessary to provide the functionality described herein.

[0066] As used herein, a “non-transitory machine-readable storage medium” or “memory” may be any electronic, magnetic, optical, or other physical storage apparatus to contain or store information such as executable instructions, data, and the like. For example, any machine-readable storage medium described herein may be any of random access memory (RAM), volatile memory, non-volatile memory, flash memory, a storage drive (e.g., hard drive), a solid state drive, any type of storage disc, and the like, or a combination thereof. The memory may store or include instructions executable by the processor.

[0067] As used herein, a “processor” or “processing circuitry” may include, for example one processor or multiple processors included in a single device or distributed across multiple computing devices. The processor (such as, processing circuitry 202 shown in FIG. 2 and / or a processor included in, for example, refinery controller 101 and / or the operation controllers 102 (not illustrated)) may be at least one of a central processing unit (CPU), a semiconductor-based microprocessor, a graphics processing unit (GPU), a field- programmable gate array (FPGA) to retrieve and execute instructions, a real time processor (RTP), other electronic circuitry suitable for the retrieval and execution instructions stored on a machine-readable storage medium, or a combination thereof.

[0068] Turning to the equipment positioned at the refinery 100, the refinery 100 may include a reactor 104 and, in some embodiments, a regenerator 120. The reactor 104 may be a catalytic reactor and / or a fluid catalytic cracking unit or reactor. The reactor 104 may include one or more sensors or meters positioned within the reactor 104 (such as sensor or meter 108) and / or proximate the reactor (such as sensors or meters 112, 114, 110, and 144). These sensor or meters may measure some aspect of fluid or material where the sensor or meter is positioned. Further, the reactor 104 may receive feedstock and / or an amount of water / steam at one or more locations of the reactor 104. The reactor 104 may be positioned or configured to convert heavy gas oil, residua, and / or other gas oil blends to an effluent or cracked fluid including smaller molecules, the effluent, in some embodiments, being further separated downstream into different products via distillation or fractionation. The reactor 104 may be operated at or it may be beneficial to operate the reactor 104 at a selected temperature range and / or pressure range to reduce over-cracking, which may cause a loss in valuable products, as well as to reduce over-all energy usage. Further, the amount of catalyst within and / or being fed to (for example, from the regenerator 120 and / or as fresh catalyst) the reactor 104 may impact the value of product produced by the reactor 104. Further, as a catalyst is regenerated within the regenerator 120, degradation may occur and / or coke deposited on the catalyst may not be completely burned off, particularly aftermultiple uses, thus further impacting product from the reactor 104. The feedstock fed to the reactor 104 may also affect parameters, such as temperature and residence time, among other parameters. Thus, several factors and / or parameters may impact the product produced by a reactor 104, those factors and / or parameters including temperature, pressure, type of catalyst, quality of catalyst, amount of catalyst, flow rate of feed or feed stock and / or catalyst, amount and temperature of steam injected, and / or properties of feed or feedstock, in addition to the product desired or targeted. The refinery controller 101 and / or operation controllers 102 may gather data related to such factors over time and apply that data, along with, in some embodiments, spectra provided from the sample analyzer 188, to trained machine learning models to produce parameters that enable production of a target product via a minimal amount of energy and / or lowest cost. Such application of data to a model may occur within various modules or circuits of one or more of the operation controllers 102 and / or, in addition to other data generated within the refinery 100, within the refinery controller 101. For example, one trained machine learning model within the predictive controls module 191 may be trained to maximize or be utilized for maximizing operating temperature in relation feedstock and a threshold temperature that may cause overcracking. In another example, another trained machine learning model may be trained to adjust or be utilized for adjusting heater temperature and / or temperature within the reactor in relation to feedstock and process parameters, composition, and / or other aspects to maximize yield or economically optimize yield from the reactor 104. In another embodiment, the trained machine learning model may be trained to adjust or be utilized for adjusting one or more refinery operation control devices to produce a selected yield of a product that also maximizes profit. Such an application of data to such a model may generate a vector that includes parameters or parameter settings corresponding to devices and / or equipment associated with the reactor 104 and, in some embodiments, the regenerator 120. That vector may be utilized by the local enhancement model to further determine, based on the outputs from other models as well as gathered data, parameters or parameter settings that enhance production of effluent from the reactor 104, such parameters or parameter settings being applied to actual equipment and / or devices via the equipment and device controller. Other models may be trained to determine parameters based on other relationships associated with the reactor 104 and / or other equipment.

[0069] In an embodiment, a trained machine learning model may be utilized by predictive controls (which may also be referred to as a prediction model) to optimize targets and / or properties and / or the predictive controls may be utilized by a an online optimizationalgorithm (which may also be referred to as the local enhancement module) to generate or determine targets.

[0070] The refinery 100 may include a regenerator 120. While a reactor 104 with a side- by-side configuration is illustrated in FIG. 1A, it will be understood that other configurations may be utilized, such as a stacked configuration. In an embodiment, in addition to reactor data and corresponding fluid properties, refinery controller 101 and / or operation controllers 102 may obtain data and fluid properties corresponding to the regenerator 120. For example, the refinery controller 101 and / or operation controllers 102 may obtain data from sensors or meters 116, 124, 126, 128, 132, 134, and 138, flow control devices associated with the regenerator 120 (such as valves 118, 130, 136, and 142), and / or the regenerator 120, as well as properties or spectra associated with spent catalyst, regenerated catalyst, a feed or feedstock (for example, to aid in catalyst regeneration), and / or air (which may include pure oxygen or some combination of oxygen and other elements). Data may be obtained from other devices, such as flow control devices (such as, valves and / or pumps, among other devices configured to control flow of a fluid) and / or temperature control devices (such as boilers, heat exchangers, heating coils, condensers, and / or other heating or cooling devices). In an embodiment, the regenerator 120 may be positioned or configured to burn coke off of spent catalyst, the coke being deposited onto the catalyst in the reactor 104. The regenerator 120 may then provide the regenerated catalyst back to the reactor 104. In embodiments, the refinery controller 101 and / or operation controllers 102 may apply the data from the regenerator 120 to produce parameters or parameter settings to adjust corresponding equipment or devices to. The trained machine learning models may be trained or utilized to determine parameters to maximize the amount of carbon build up burned from catalyst, to increase temperature within the reactor 104 (for example, via heat from regenerated catalyst), and / or to minimize the amount of resources utilized by the regenerator 120.

[0071] Other equipment may be positioned throughout the refinery 100 and the refinery controller 101 and / or operation controllers 102 may connect to such equipment. The refinery controller 101 and / or operation controllers 102 may obtain or gather data related to that equipment during the refining operation. For example, the refinery controller 101 and / or operation controllers 102 may obtain data from and / or related to a fractionation column 148 or distillation column. Further, the refinery 100 may include and the refinery controller 101 and / or operation controllers 102 and / or one or more of the sub-controllers 190 may obtain data from a hydrotreater (such as hydrotreater 166 and hydrotreater 174)and / or an alkylation unit 158. The refinery controller 101 and / or operation controllers 102 may obtain data from the valve 146, sensors or meters 150, 154, 160, 164, 168, 172, 178, and 180, as well as the properties associated with the products from the fractionation column 148 (for example, off gas 152, LPG 156, alkylate 162, gasoline 170, diesel 176, slurry 182, and / or other products), the hydrotreater 166 (for example, gasoline or high- octane gasoline), the hydrotreater 174 (for example, diesel, low-sulfur diesel, and / or higher purity diesel), and / or the alkylation unit 158 (for example, alkylate).

[0072] In an embodiment, the refinery controller 101 and / or operation controllers 102 may obtain data in real time, near real-time, and / or continuously or substantially continuously. In another embodiment, the refinery controller 101 and / or operation controllers 102 may obtain data periodically. In another embodiment, the refinery controller 101 and / or operation controllers 102 may apply data to a trained machine learning model at a selected time interval. Such an interval may be based on the time for samples from various positions within the refinery 100 to be collected and then analyzed. In yet another embodiment, each of the operation controllers 102 may obtain data related to a selected section of the refinery 100. Each of the operation controllers 102 may also obtain properties and / or spectra from the sample analyzer 188. After the operation controllers 102 obtain the properties and / or spectra and data, then the operation controllers 102 may apply the properties and / or spectra and data to a corresponding predictive controls module to produce parameters to produce a target product. The local enhancement module of the operation controller may then apply, to a trained machine learning model of the local enhancement module, the output of each of the predictive controls module, the data obtained throughout the refinery 100, and / or the properties and / or each spectra associated with a collected sample. Such an application may produce an enhanced or optimized set of parameters, which may then be applied to equipment or devices via the equipment and devices controls.

[0073] In yet another embodiment, the refinery controller 101 may first obtain data and the properties and / or spectra and then apply the data and the properties and / or spectra to a trained machine learning model. The refinery controller 101 may transmit the output of the trained machine learning model to each operation controller 102. In another embodiment, the refinery controller 101 may provide target products and corresponding parameters to each of the operation controllers 102, based on user input, previously utilized parameters, current cost of a target product and / or feedstock, and / or other factors. In another embodiment, the refinery controller 102 may facilitate communication between each of the operation controllers 102, facilitate data acquisition for the operation controllers 102,and / or facilitate parameter prediction and / or adjustment among the plurality of operation controllers 102. For example, if one operation controller adjusts a process to meet a selected target, that adjustment may affect upstream and / or downstream processes. The refinery controller 101 may facilitate communication and / or perform additional predictions to ensure that such parameter adjustments enable the upstream and / or downstream processes to continue to produce target products.

[0074] FIG. 2 is a simplified diagram that illustrates an apparatus for enhance fluid production at a refinery, according to an embodiment of the disclosure. Such an apparatus 200 may be comprised of a processing circuitry 202, a memory 204, a communications circuitry 206, a modeling circuitry 208, a fluid adjustment circuitry 210, and an equipment and device adjustment circuitry 212, each of which will be described in greater detail below. While the various components are illustrated in FIG. 2 as being connected with processing circuitry 202, it will be understood that the apparatus 200 may further comprise a bus (not expressly shown in FIG. 2) for passing information amongst any combination of the various components of the apparatus 200. The apparatus 200 may be configured to execute various operations described herein, such as those described above in connection with FIGS. 1 A-1B and below in connection with FIGS. 3-26.

[0075] The processing circuitry 202 (and / or co-processor or any other processor assisting or otherwise associated with the processor) may be in communication with the memory 204 via a bus for passing information amongst components of the apparatus. The processing circuitry 202 may be embodied in a number of unusual ways and may, for example, include one or more processing devices configured to perform independently. Furthermore, the processor may include one or more processors configured in tandem via a bus to enable independent execution of software instructions, pipelining, and / or multithreading.

[0076] The processing circuitry 202 may be configured to execute software instructions stored in the memory 204 or otherwise accessible to the processing circuitry 202 (e.g., software instructions stored on a separate storage device). In some cases, the processing circuitry 202 may be configured to execute hard-coded functionality. As such, whether configured by hardware or software methods, or by a combination of hardware with software, the processing circuitry 202 represents an entity (for example, physically embodied in circuitry) capable of performing operations according to various embodiments of the present disclosure while configured accordingly. Alternatively, as another example, when the processing circuitry 202 is embodied as an executor of software instructions, thesoftware instructions may specifically configure the processing circuitry 202 to perform the algorithms and / or operations described herein when the software instructions are executed.

[0077] Memory 204 is non-transitory and may include, for example, one or more volatile and / or non-volatile memories. In other words, for example, the memory 204 may be an electronic storage device (for example, a computer readable storage medium). The memory 204 may be configured to store information, data, content, applications, software instructions, or the like, for enabling the apparatus 200 to carry out various functions in accordance with example embodiments contemplated herein.

[0078] The communications circuitry 206 may be any means such as a device or circuitry embodied in either hardware or a combination of hardware and software that is configured to receive and / or transmit data from / to a network and / or any other device, circuitry, or module in communication with the apparatus 200. In this regard, the communications circuitry 206 may include, for example, a network interface for enabling communications with a wired or wireless communication network. For example, the communications circuitry 206 may include one or more network interface cards, antennas, buses, switches, routers, modems, and supporting hardware and / or software, or any other device suitable for enabling communications via a network. Furthermore, the communications circuitry 206 may include the processing circuitry for causing transmission of such signals to a network or for handling receipt of signals received from a network. The communications circuitry 206, in an embodiment, may enable reception of parameters from various components, devices, and / or sensors (for example, flow control devices, analyzers, sensors, equipment, and / or other components), as well as communication of instructions and / or signals indicative of adjustment to those components and / or devices.

[0079] The apparatus 200 may include a modeling circuitry 208 configured to obtain parameters from one or more components, equipment, devices, sensors, and / or analyzers and / or apply those parameters to a trained machine learning model to obtain parameters that enable equipment to produce a target product. In other embodiments, the modeling circuitry 208 may apply, in addition to the parameters described herein, the output of other similar circuitry (in other words, an additional plurality of modeling circuitry that each correspond to one of a plurality of sub-operations). Obtaining the parameters from the one or more components, equipment, devices, sensors, and / or analyzers may occur periodically, at selected times, continuously, or substantially continuously. In an example, the modeling circuitry 208 may obtain parameters for sub-operations first, generating an output for each sub-operation. Upon generation of an output for each sub-operation, the modeling circuitry208 may obtain each output and a current data set. The modeling circuitry 208 may poll the components, devices, sensors, and / or analyzers to obtain such parameters or, in an embodiment, receive the parameters without polling. The modeling circuitry 208 may obtain the parameters via the communications circuitry 206. Application of the parameters to the trained machine learning model may determine, generate, or cause generation of an output. The output may be indicative of an adjustment to equipment, fluids, devices, and / or operations to meet or accurately meet a target product and / or to operate the equipment at higher than typical efficiency, for example, utilizing less power or resources such as in a heater or boiler or utilizing a heat exchanger to reduce power usage.

[0080] In another embodiment, the modeling circuitry 208 may train the trained machine learning model prior to use. In such embodiments, the modeling circuitry 208 may obtain historical data, preprocess the historical data, and then train and test the machine learning model. In yet another embodiment, after a refining operation (in other words, after a selected product has been generated via the refining operation), the modeling circuitry 208 may re-train or refine the trained machine learning model, based on the results of the refining operation (in other words, the accuracy of the parameters in reaching the target product’s properties).

[0081] In another, the modeling circuitry 208 may train and / or include a plurality of machine learning models. Each of the plurality of machine learning models may correspond to a selected refinery operation and / or sub-operation.

[0082] The modeling circuitry 208 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described above in connection with FIGS. 1A-1B and below in connection with FIGS. 3-26. The output of the modeling circuitry 208 may be transmitted to other circuitry of the apparatus 200 (such as the fluid adjustment circuitry 210 and / or equipment and device adjustment circuitry 212).

[0083] In addition, the apparatus 200 further comprises the fluid adjustment circuitry 210 that may cause adjustment of feedstock and / or other fluids utilized in a refining operation. In an embodiment the output from the modeling circuitry 208 may be a matrix, a series of parameters, and / or some indicator. In an embodiment, the fluid adjustment circuitry 210 may utilize that output to adjust a blend of feedstock and / or the fluid used in other inputs (for example, an amount of hydrogen, butane, other alkanes, and / or other fluids). The fluid adjustment circuitry 210 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, asdescribed above in connection with FIGS. 1A-1B and below in connection with FIGS. 3- 26. The fluid adjustment circuitry 210 may further utilize communications circuitry 206 to transmit signals to adjust the type and / or amount of feedstock to utilize.

[0084] In addition, the apparatus 200 further comprises the equipment and device adjustment circuitry 212 that may cause adjustment of equipment and / or devices utilized in a refining operation. In an embodiment, the equipment and device adjustment circuitry 212 may utilize that output to adjust temperature, pressure, flow rate, and / or other parameters corresponding to the equipment and / or devices positioned within the refinery (e.g., by setting valve positions, pump speeds, etc.). The equipment and device adjustment circuitry 212 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described above in connection with FIGS. 1A and IB and below in connection with FIGS. 3-26. The equipment and device adjustment circuitry 212 may further utilize communications circuitry 206 to transmit signals to adjust equipment and / or devices utilized.

[0085] Although components 202-212 are described in part using functional language, it will be understood that the particular implementations necessarily include the use of particular hardware. It should also be understood that certain of these components 202-212 may include similar or common hardware. For example, the modeling circuitry 208, the fluid adjustment circuitry 210, and the equipment and device adjustment circuitry 212 may, in some embodiments, each at times utilize the processing circuitry 202, memory 204, or communications circuitry 206, such that duplicate hardware is not required to facilitate operation of these physical elements of the apparatus 200 (although dedicated hardware elements may be used for any of these components in some embodiments, such as those in which enhanced parallelism may be desired). Use of the terms “circuitry,” with respect to elements of the apparatus therefore shall be interpreted as necessarily including the particular hardware configured to perform the functions associated with the particular element being described. Of course, while the terms “circuitry” should be understood broadly to include hardware, in some embodiments, the terms “circuitry” may in addition refer to software instructions that configure the hardware components of the apparatus 200 to perform the various functions described herein.

[0086] Although the modeling circuitry 208, the fluid adjustment circuitry 210, and the equipment and device adjustment circuitry 212 may utilize processing circuitry 202, memory 204, or communications circuitry 206 as described above, it will be understood that any of these elements of apparatus 200 may include one or more dedicated processors,specially configured field programmable gate arrays (FPGA), or application specific interface circuits (ASIC) to perform its corresponding functions, and may accordingly utilize processing circuitry 202 executing software stored in a memory or memory 204, communications circuitry 206 for enabling any functions not performed by special-purpose hardware elements. In all embodiments, however, it will be understood that the modeling circuitry 208, the fluid adjustment circuitry 210, and the equipment and device adjustment circuitry 212 are implemented via particular machinery designed for performing the functions described herein in connection with such elements of apparatus 200.

[0087] In some embodiments, various components of the apparatus 200 may be hosted remotely (e.g., by one or more cloud servers) and thus need not physically reside on the corresponding apparatus 200. Thus, some or all of the functionality described herein may be provided by third party circuitry. For example, a given apparatus 200 may access one or more third party circuitries via any sort of networked connection that facilitates transmission of data and electronic information between the apparatus 200 and the third party circuitries. In turn, that apparatus 200 may be in remote communication with one or more of the other components describe above as comprising the apparatus 200.

[0088] As will be appreciated based on this disclosure, example embodiments contemplated herein may be implemented by an apparatus 200 (or by a refinery controller). Furthermore, some example embodiments (such as the embodiments described for FIGS. 1A-1B and 3-26) may be a computer program product comprising software instructions stored on at least one non-transitory computer-readable storage medium (such as memory 204). Any suitable non-transitory computer-readable storage medium may be utilized in such embodiments, some examples of which are non-transitory hard disks, CD-ROMs, flash memory, optical storage devices, and magnetic storage devices. It should be appreciated, with respect to certain devices embodied by apparatus 200 as described in FIG. 2, that loading the software instructions onto a computing device or apparatus produces a special-purpose machine comprising the means for implementing various functions described herein.

[0089] FIG. 3 a simplified diagram that illustrates example refining controllers and an example refining enhancer to enhance control of a refining process at a refinery, according to an embodiment of the disclosure. As illustrated in FIG. 3, a refinery may include one or more operation controllers 302. The operation controllers 302 may connect to, for example, a number of feeds and / or processing units (refinery equipment configured to process a feedstock or other input). As illustrated, the operation controllers 302 may connect to andreceive data from feed A 303 A, feed B 303B, and up to feed 303N, sensors or other devices associated with each feed (such as sensor 304A, 304B, and up to 304N), and various flow control devices (such as valve 306A, valve 306B, and up to valve 306N). In such embodiments, each of feed A 303 A, feed B 303B, and up to feed 303N may flow to a first processing unit 308. A processing unit, for example, in this case, the first processing unit 308, may include one or more refining devices or equipment positioned at a refinery (for example, a FCC unit, a distillation column, and other equipment as described herein). The first processing unit 308 may convert, process, and / or transform a feed into a unit material (such as unit material A 310A, unit material B 310B, and up to unit material N 310 N). Additional processing units may be positioned throughout the refinery. As illustrated though, the unit materials may flow to a “Nth” processing unit 314. Sensors (such as sensor 312A, 312B, and up to 312N) and valves (such as valve 313 A, valve 313B, and up to valve 313N) may be positioned between the feed and “Nth” processing unit 314. The final processing device (in other words, the “Nth” processing unit 314) may produce one or more end materials (such as end materials A 316A, end materials 316B, and up to end materials 316N). Sensors (such as sensor 318A, 318B, and up to 318N) and valves (such as valve 320A, valve 320B, and up to valve 320N) may be positioned between the end materials and the material destination 322.

[0090] In an embodiment, as each feed is fed to the next processing unit, the operation controller 302 may determine various characteristics and / or properties of the feed. For example, the operation controller 302 may determine temperature, pressure, and / or flow rate, in addition to the composition of the feed and the spectra or properties determined via spectrographic analysis of the feed. For example, as illustrated, the operation controller 302 may determine or obtain feed information 324 (including, at least feed composition 326 and / or feed properties 328, among other data), unit material information 330 (including, at least unit material composition 332 and / or unit material properties 334, among other data), and / or end material information 336 (including, at least end material composition 338 and / or end material properties 340, among other data). Thus, the operation controller 302 may obtain data related to each feed / material in real-time or near real-time, during a refinery operation, and / or directly or indirectly (for example, spectra may be obtained via a sample or spectrographic analyzer).

[0091] Once all the data has been obtained, the operation controller 302 may apply, to the machine learning model 360 of a local enhancer 358, the data including processing unit constraints 350, a target product 352 (including a target composition 354 and targetproperties 356), and / or material differences 344 (including composition differences 346 and properties difference 348) as determined via a comparator 342 (the comparator positioned or configured to compare composition and properties of different materials). The machine learning model 360 may produce material targets 362 which may be utilized to produce target feed ratios 364 and target operation unit parameters 366. In another embodiment, these values may be fed to the comparator and then, after obtaining differences related to another material, reapplied to the machine learning model 360. The machine learning model 360 may then produce adjusted targets 370 (including adjusted target feed ratios 372 and adjusted target operation unit parameters 374).

[0092] In another embodiment, the output of the machine learning model (for example, a vector comprising a plurality of components, each component being a parameter setting for a selected or specific device or equipment) may be compared to current parameter settings for the selected or specific devices or equipment in the comparator 342. In such embodiments, if the comparator 342 determines that there is a difference between an output of the machine learning model, then the parameter settings of the equipment or devices at the refinery may be adjusted.

[0093] In embodiments, the controller 302 may drive the materials to the target by adjusting the valves and / or feed (for example, the blend of different feeds or materials used in the subsequent operation) at one or more points in the overall refining operation.

[0094] The machine learning model 360 may include neural networks, supervised learning models, semi -supervised learning models, unsupervised learning models, or some combination thereof, as will be readily understood by one having ordinary skill in the art. In another embodiment, different types of machine learning algorithms may be utilized for different refinery operations. In further embodiments, some refining operations may use, rather than or in addition to a neural network, decision trees, support vector machines, hidden Markov models, Bayesian networks, linear regression, k-means, and / or tabular reinforcement learning. Specific neural networks that may be utilized include a recurrent neural network, such as a long short-term memory network. Such neural networks may utilize a fixed horizon of historical data to predict future behavior. Additionally, such neural networks may utilize standard active functions, for example a rectified linear unit or a hyperbolic tangent. As noted, in embodiments, different models may be utilized for different operations. The determination for which model to use for each operation may be determined based on error rates associated with a selected model, the R2value, SHAP plotsand / or values, gain directions and / or magnitude, and / or gain distributions, among other factors.

[0095] In an embodiment, the operation controller 302, local enhancer 358, and / or comparator 342 may be included in a single controller, a plurality of controllers, one or more computing devices, and / or as one or more modules or as instructions. In other embodiments, a plurality of controllers or computing devices may each include a specific model corresponding to one of the processing units. In another embodiment, the operation controller 302 may include or may be a supervisory controller that considers the predictions of other processing unit specific controllers when generating adjusted targets via the supervisory controller’s machine learning model.

[0096] As noted, data may be obtained in real time or near real-time. In some embodiments, the application of data to a machine learning model may be delayed by the time taken to obtain spectra or properties for a feed or material. Thus, in an embodiment where the operation controller 302 is a supervisory controller, the supervisory controller may generate adjusted targets after each sub-controller generates a target for a specific processing unit. Thus, the overall adjustment targets may be determined at a second time interval, greater than the first time interval, while each sub-adjustment target may be determined at a first time interval.

[0097] In another embodiment, a refinery may include a plurality of operation controllers. Each operation controller 302 may include a plurality of trained machine learning models. Each trained machine learning model may be trained to recognize a specific or selected trend in a set of data. Thus, each operation may be adjusted based on the outputs of a plurality of models, ensuring the operation as a whole produces an accurate target product. Further, each operation controller may interact with each other operation controllers. For example, adjusted parameters from all operation controllers may be provided to each operation controller. Thus, as one operation is adjusted, a downstream and / or upstream operation may be further adjusted based on the adjustment of the one operation.

[0098] FIG. 4 is a simplified diagram that illustrates the training of a machine learning model for enhanced fluid production at refinery, according to an embodiment of the disclosure. Each model described herein may be trained prior to use. Such training may be performed prior to use with a set of historical data specific to a refinery. In a further embodiment, a plurality of machine learning models may be trained, each based on data specific to an operation and selected equipment at the refinery.

[0099] As noted, the machine learning models described herein may be trained using data. As illustrated in FIG. 4, the data may include historical, equipment specific data 402. In other embodiments, the training data may include data related to the entire operation of a refinery, as well as outputs from equipment specific models. In another embodiment, a machine learning model may be re-trained and / or refined via current and marked up equipment specific data 404. The historical equipment specific data 402 and current and marked up equipment specific data 404 may include feed composition, feed properties, material composition, material properties, a target product or products, target composition, target properties, temperatures in equipment, pressure in equipment, flow rates associated with feed and / or materials, and / or equipment parameters. In embodiments, the historical equipment specific data 402 may include or may be utilized to generate a non-linear concave function. In such embodiments, the desired outcome may be determined based on the maximum of such a function. In another embodiment, the desired outcome may be included or added to the data set. In yet another embodiment, training may include the machine learning model learning particular patterns that indicate what the desired outcome may be based on trends within the data. In another embodiment, physics-based data may be provided along with the historical data set to ensure that outputs from a trained machine learning model remain consistent and / or emulate real process / actual possibilities. In yet another embodiment, a plurality of machine learning models may be trained for the same operation. Each of the plurality of machine learning models may utilize different portions of historical data and / or other inputs to cause the model to maximize a specific attribute or parameter. Another model may be trained to utilize the outputs of each of those plurality of models, in addition to data.

[0100] Once the historical data, and any other current data, is available, that data may be pre-processed 406. In such embodiments, the data may be normalized. In other words, outlying data points that are anomalies may be removed from the data set. Further, data corresponding to abnormal events may be removed, such as data generating during startup, shut-down, turn-arounds, maintenance, and / or upsets. Further, undesired data and invalid measurements may be removed. Finally, data may be segregated or separated into sequences based on time. The sequences may comprise data obtained over a consecutive time period, such as time intervals of 30 minutes, 1 hour, 2 hours, and / or 3 hours, or more or less than the time intervals listed. In another embodiment, other factors may be utilized to segregate or separate the data, such as feed used and / or target product being produced.

[0101] Once the data set has been pre-processed, a model may be trained 408. In embodiments, a portion of the data set (for example, 70%, 80%, or 90%) may be fed to the machine learning model. The machine learning model may utilize the inputs versus the known desired outcome (such as target product composition and properties) and / or known undesired outcome to “learn” what parameters can be utilized to reach the known desired outcome and what parameters lead to the known undesired outcome. Once the data has been used to train the machine learning model, then the remaining portion of the data set may be utilized to test 410 the trained machine learning model. If the trained machine learning model does not meet or achieve a selected error rate, then trained machine learning model may be re-trained or refined with a different randomized portion of the data set, and the re-training repeated as necessary, until the selected error rate is met or achieved. In another embodiment, other training schema may be utilized. In another embodiment, readiness of the trained machine learning model may be determined based on how close the trained machine learning model comes to an expected outcome, based on the test data set.

[0102] Once the trained machine learning model 412 meets a selected error rate, then the trained machine learning model may be released for further use. In another embodiment, a separate step may include selection of a type of machine learning model prior to training of the machine learning model. In other embodiments, various types of models may be trained, then tested. The most accurate models, determined by an error rate for each model, may be utilized.

[0103] Another machine learning model may, when trained, determine a predicted feedstock blend for an operation based on value to unit constraints. Another machine learning model may, when trained, determine adjustments to hydrotreater severity to vary aromatic saturation against downstream process constraints (in other words, the amount of feed from a hydrotreater to a downstream process such as a fluid catalytic cracking process, an isomerization process, a catalytic reforming process, or a fuel blending pool, may be adjusted based on, for example, the content and / or composition of the feed, among other factors). Another machine learning model may, when trained, determine adjustments to a hydrotreater to give feed composition that provides optimal downstream process yield against process constraints, regulatory requirements, or customer specifications and, in some embodiments, in terms of profit. In yet another embodiment, another machine learning model may, when trained, determine adjustments to hydrotreater severity to maximize aromatic saturation to hydrotreater constraints.

[0104] FIG. 5 is a schematic diagram of a FCC control system 500 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. As noted, specific sections or portions of a refinery may include a sub-controller to enhance that particular operation. As illustrated in FIG. 5, a FCC operation may be enhanced via a FCC controller 502 and corresponding machine learning models (for example, FCC specific machine learning models utilized in the local enhancement module 504 and / or the predictive controls module 506). The FCC controller 502 may connect to sources of and control various parts of the FCC reactor 532, including, in some embodiments, a regenerator 552. For example, the FCC controller 502 may control amounts and / or type of feed 514, FCC catalyst 510, fluffing steam 522, stripping steam 526, pre-stripping steam 536, and / or spent catalyst 542 flowing from the FCC reactor 532. Further, the FCC controller 502 may connect to sources of and control spent catalyst 542 (for example, via slide valve 548), regenerated catalyst 556 (for example, via slide valve 564), and / or air 560 flowing into the regenerator 552. Further, the FCC controller 502 may obtain data related to each input material or feed, as well as the temperature and / or pressure within the FCC reactor 532 and / or regenerator 552.

[0105] Prior to generating adjusted parameters for operation of the FCC reactor 532, the FCC controller 502 may initiate collection of samples of one or more of the materials or feeds flowing into and out of the FCC reactor 532 and / or regenerator 552, via the sample collection and analysis assembly 508. Once the sample collection and analysis assembly 508 obtains one or more samples, the sample collection and analysis assembly 508 may analyze those samples to produce properties and / or a spectra indicative of various properties, as well as data from sensors disposed throughout the FCC operation. Once the properties and / or spectra are analyzed, the sample collection and analysis assembly 508 may transmit the data, including the sample and sensor data and analysis, to the FCC controller 502. The data may include feedstock data indicative of one or more of the properties or composition of the feedstocks of the FCC, operational data indicative of one or more of the temperatures, pressures, feed rates of feedstocks, catalysts, steam, and other materials throughout the FCC, and other operating parameters of the FCC, and product data indicative of one or more of the properties or composition of the products of the FCC. Upon reception of the data, the FCC controller 502 may apply the data, along with target product composition and / or properties, to one or more machine learning models within the predictive controls module 506 (or, in other embodiments, a plurality of predictive controls modules may be included in the FCC controller 502 and each may include a machinelearning model. Alternatively, a single machine learning model may incorporate the functionality of the predictive control modules and the local enhancement module. In other embodiments, the output may indicate that a new or fresh catalyst should be utilized.

[0106] The output of each of the machine learning models in the predictive controls module 506 may then be utilized by the local enhancement module 504 to determine (for example, as a vector) new parameters and / or feed blend or composition to supply to the FCC reactor 532. The FCC controller 502 may then adjust the parameters of and / or feeds and / or materials for the FCC reactor 532 and / or regenerator 552.

[0107] The FCC reactor 532 may exhibit non-linear operation in relation to temperature, as well as in relation to other factors. If the temperature is below or above a threshold range, then the output from the FCC reactor 532 may not include the highest possible product yield. Thus, at least one machine learning model for a FCC operation may, when trained, determine a desired outcome based on the maximum of the non-linear objective function relating to temperature and the output or yield (for example, an increase in amount of one or more products of a plurality of products) of the FCC reactor 532. In another embodiment such a non-linear objective function may maximize profit (for example, based on an increase in yield of a particular product or one or more products of a plurality of products). In another embodiment, a non-linear objective function relating to temperature and product composition may be used by a machine learning model to generate target operational parameters to achieve a target point of the non-linear objective function. The resulting model may be utilized to drive, in part, the FCC reactor 532 operation to that maximum non-linear function or non-linear profit function. Such a function is also dependent on the properties of the feed, which, as noted, change over time, among other factors. For example, another machine learning model may, when trained, determine a desired outcome based on the maximum or minimum of another non-linear function relating to temperature of other areas, pressure within the FCC reactor 532, temperature within the regenerator 552, amount of catalyst, type of catalyst, age or deactivation state of catalyst, and / or flow rates to and / or from the FCC reactor 532. Another machine learning model may, when trained, determine a predicted feedstock blend for a FCC operation based on value to unit constraints. Another machine learning model may, when trained, determine adjustments to hydrotreater severity to vary aromatic saturation against FCC constraints (in other words, the amount of feed from an upstream hydrotreater to a FCC reactor 532 may be adjusted based on, for example, the content and / or composition of the feed, among other factors). Another machine learning model may, when trained, determine adjustments to a hydrocracker conversion to feedunconverted gas oil to the FCC that provides optimal FCC yield against FCC constraints and, in some embodiments, in terms of profit. The FCC constraints may include minimum regeneration temperature, wet gas compressor capacity, FCC fractionation limits, surge drum capacity, charge pump capacity, heater capacity, reactor capacity, regenerator capacity, main air blower capacity, and gas concentration section limits. In yet another embodiment, another machine learning model may, when trained, determine adjustments to hydrotreater severity to maximize aromatic saturation to hydrotreater constraints.

[0108] In another embodiment, the FCC controller 502 may first cause sampling of various fluids (e.g., liquids, vapors, and / or gases) used and / or produced in the FCC operation. For example, as FCC operation occurs (for example, as a continuous and / or ongoing operation) various fluids and / or materials may be utilized and / or produced therein. Prior to application of data to any of the machine learning models described herein, the FCC controller 502 may initiate capture of one or more of those fluids via the sample collection and analysis assembly 508. Once samples are analyzed, the FCC controller 502 may predict properties of the corresponding feedstocks and operational parameters that may achieve an accurate output of a target product. In other words, the FCC controller 502 may determine the composition of feedstocks and the blends that may be used to achieve a target product property and / or composition. The feedstocks may then be adjusted, blended, and / or supplemented to achieve a target feedstock composition determined by one or more of the machine learning models of the FCC controller 502. Stated another way, the FCC controller 502 may control the properties, composition, and / or feed ratios associated with a feedstock or hydrocarbon feedstock and / or an intermediate fluid or intermediate product.

[0109] In an embodiment related to FIG. 1, the machine learning model may be trained or configured to determine an amount of giveaway of light cycle oil (LCO) in the slurry 182 against an exchanger fouling rate in a slurry circuit from the bottoms of the FCC main fractionator 148. The light cycle oil (LCO) is shown as becoming diesel 176 after hydrotreating in the hydrotreater 174. The slurry contains contaminants released in the FCC process, as well as polynuclear aromatics that were not cracked during the FCC process. The slurry 182 is generally a thick fluid and may not flow well through the conduits from the FCC main fractionator. The addition of light cycle oil (LCO) assists in improving the flow properties of the slurry 182 preventing clogs and fouling downstream. The machine learning model may be trained or configured to determine a slurry cut point to prevent, reduce, or significantly reduce fouling in downstream slurry exchanger sections by permitting some light cycle oil (LCO) to be added to the bottoms and the slurry 182.

[0110] FIG. 6 is a schematic diagram of an enhanced hydrotreater control system and a distillation control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. As illustrated in FIG. 6, a distillation / fractionation portion of the refinery may include a distillation controller 602 and / or a hydrotreater controller 608. The distillation controller 602 may obtain data related to a fractionation / distillation column 616. For example, the distillation controller 602 may obtain temperature and / or pressure within the fractionation / distillation column 616. Further, the distillation controller 602 may initiate collection of samples of various fluids associated with the fractionation / distillation column 616 via the sample collection and analysis assembly 614. For example, the sample collection and analysis assembly 614 may obtain samples of off-gas 624, LPG 628, naphtha / gasoline 630, diesel 640, and / or slurry 620, among other fluids associated with the fractionation / distillation column 616. Sensor packages 618, 622, 626, 630, 636 may be disposed to measure the temperature, pressure, flow rate, and composition of the products from the fractionation / distillation column 616 including off gas 624, LPG 628, gasoline 634, diesel 640, and slurry 620.[OHl] Once the sample collection and analysis assembly 614 obtains the samples, the sample collection and analysis assembly 614 may analyze the samples to produce properties or spectra indicative of the properties of each collected sample. The sample collection and analysis assembly 614 may provide the properties and / or spectra to the distillation controller 602. The distillation controller 602 may then apply the data, properties, and / or spectra to the to one or more training machine learning models associated with one or more of a local enhancement module 604 and / or predictive controls module 606. Based on the output of the trained machine learning models, which may indicate parameter and / or feed adjustment of the fractionation / distillation column 616, the distillation controller 602 may adjust the parameters and / or the feed via the local enhancement module 604 and, in some embodiments, an equipment and device control module. The equipment and device control module may comprise a PLC or DCS. In some embodiments, the equipment and device control module may comprise a DCS-PID module, controller, or circuitry.

[0112] In an embodiment, the distillation machine learning model 604 may be trained to maximize lift within the distillation column. Such a model may utilize temperature and / or feed flow rate to maximize such a parameter (in other words, to maximize lift or pressure). Further, such a model may predict parameters based on, in part, the concave objective function for profit to determine a feed and / or temperature input that drive the distillationcolumn to include a maximum lift in a pressure limited column. The parameters for such an application of data to a trained machine learning model may include providing more energy and / or material to the column or increasing material or energy to increase pressure if all pressure control handles are exhausted. For example, the output from one of the trained machine learning models may indicate an increase in temperature and / or feed to maximize the output of a selected one or more products from the fractionation / distillation column 616. Such a maximization of the output may optimize profit for the fractionation / distillation column 616.

[0113] The fractionation / distillation column 616 may include a vacuum column. Such issues described above may also occur for a vacuum column when the vacuum tower may no longer be cooled (for example, due to cooling water limitations) or when a vacuum ejector system is limited and may no longer decrease pressure. The concave objective function in this case can be used to determine what feed and temperature in the column with uncontrolled pressure give the most profitable light vacuum gas oil (LVGO), middle vacuum gas oil (MVGO), heavy gas oil (HVGO) lift from resid.

[0114] In another example, the fractionation / distillation column 616 may comprise a vacuum column and / or crude unit. In such examples, one trained machine learning model may be trained or configured to utilize a resid viscosity, asphalt m-value, and / or other asphalt property as an input and / or, in other embodiments, sample analysis of the outputs and inputs of the vacuum column. Further, the trained machine learning model may utilize as an input, current valve settings, heater outlet temperature (for example a vacuum heater outlet temperature), and / or vacuum operating parameters. The output of such a model may include updated settings or set points for the valve settings, heater outlet temperature (for example a vacuum heater outlet temperature), and / or vacuum operating parameters. Such a trained machine learning model may minimize the amount of gas oil in the resid, when the resid is being utilized to make asphalt. The distillation controller 602 may utilize such an output to set equipment to parameters or settings output from the trained machine learning model.

[0115] The issues described above may also occur for an atmospheric vacuum column, where the feed and temperature in the column with uncontrolled pressure may be determined with a neural network model and controller to give higher lift of naphtha, distillate, and atmospheric gas oil from a crude column feed. For an atmospheric vacuum column, several constraints may be considered, such as product quality, hydraulic constraints, operating limits, and column differential pressure limits. Alternatively,problems associated with operation of the atmospheric vacuum column may be formulated to manipulate other variables such as pump-around return flow, pump-around return temperatures, and stripping steam that optimize the lift in the column when the column pressure is not being controlled.

[0116] In an embodiment, properties of feedstock utilized in a distillation operation may and / or other operations described herein may include boiling point, viscosity, composition, API gravity, distillation points, Coker gas oil content, carbon residue content, nitrogen content, sulfur content, saturates content, thiophene content, single-ring aromatics content, dual-ring aromatics content, triple-ring aromatics content, or quad-ring aromatics content. In another embodiment, fluids, target products, material, and / or unit materials produced by a fractionation / distillation column 616 may include one or more of an amount of butane- free gasoline, an amount of total butane, an amount of dry gas, an amount of coke, an amount of gasoline, octane rating, an amount of light fuel oil, an amount of heavy fuel oil, an amount of hydrogen sulfide, an amount of sulfur in light fuel oil, or an aniline point of light fuel oil. In another embodiment, properties of the fluids, target products, material, and / or unit materials produced by the fractionation / distillation column 616 may include one or more of pentane content, raw crude water content, desalted crude water content, heavy atmospheric gas oil (HAGO) content, light atmospheric gas oil (LAGO) flash, or kerosene flash point. In yet another embodiment, the distillation controller 602 may control the pentane content, the raw crude water content, the desalted crude water content, the heavy atmospheric gas oil (HAGO) content, the light atmospheric gas oil (LAGO) flash, or the kerosene flash point, one or more of: crude blend, make-up water, desalter severity, HAGO wash rate, stripping, LAGO draw rate, stripping steam, or kerosene draw of the one or more of the first processing units.

[0117] In yet another embodiment, the properties of the fluids, target products, material, and / or unit materials produced by the fractionation / distillation and absorber column 616 may include one or more of ethane content, propane content, propene content, isobutane content, or n-butane content. In such embodiments, the distillation controller 602 may control one or more of the ethane content, the propane content, the propene content, the isobutane content, or the n-butane content, one or more of: absorber pressure, lean oil flow rate, lean oil temperature, high-pressure separator temperature, reactor conversion, or stripper reboiler duty.

[0118] In yet another embodiment, the properties of the fluids, target products, material, and / or unit materials produced by the fractionation / distillation column 616 may includeone or more of high-pressure separator water content or stripper bottoms water content. In such embodiments, the distillation controller 602 may control a temperature of a high- pressure separator.

[0119] In another embodiment, the fractionation / distillation column 616 may comprise a vacuum tower (in other words, a distillation tower operating under reduced pressure). In such embodiments, one of the trained machine learning models may infer or predict a micro-carbon residue (MCR) or HVGO MCR based on various parameters and / or feeds (such as, for example, wash rate, c-factor, bed distribution, lift drive-up entrainment, and / or feed properties, among others), enabling a controller (for example, the distillation controller 602) to determine adjustments to a vacuum distillation operation based on the inferred or predicted MCR. In another embodiment, rather than or in addition to utilizing HVGO MCR, the trained machine learning models may infer or predict nickel and / or vanadium (and / or other metals) content in HVGO. In yet another embodiment, rather than or in addition to utilizing HVGO MCR, the trained machine learning models may infer or predict a distillation cutpoint, a gas oil lift from resid, and / or HVGO wash bed lifecycle for HVGO production and / or to prevent premature shutdown.

[0120] In another embodiment, the trained machine learning model may also optimize deisopentanizer fractionation, such an optimization increasing octane on, in some embodiments, an isomerization unit. Such a trained machine learning model may be utilized to optimize DIP feed limit based on various factors, such factors being applied to the trained machine learning model. Further, such factors may include amount of steam utilized, temperature from a reboiler, a reflux rate, feed properties, and / or output properties.

[0121] In another embodiment, the distillation controller 602 (and / or, in embodiments, the local enhancement module 604 and / or predictive controls module 606) may include a trained machine learning model trained and / or configured to determine a salt point temperature for the fractionation / distillation column 616 (or, in some embodiments, a crude atmospheric distillation column). In such embodiments, the distillation controller 602 may input fractionation / distillation column 616 setpoints, valve setpoints, overhead temperature, and / or overhead reflux, among other factors, to the model. The trained machine learning model may output updates to the setpoints for the fractionation / distillation column 616 and / or valves, as well as temperature and / or overhead reflux setpoints. Such a trained machine learning model may prevent salt deposition in overhead piping and / or other downstream mechanical equipment. These salt depositionsmay lead to a loss of containment, premature damage of equipment, and / or premature equipment or plant shutdown.

[0122] Further, FIG. 6 illustrates a hydrotreater controller 608. The hydrotreater controller 608 may obtain data related to each hydrotreater 632 and 638. Data obtained from each hydrotreater may be analyzed separately, as each hydrotreater performs a different function (for example, increase gasoline octane or remove sulfur and / or impurities from diesel). In yet another embodiment, the hydrotreater controller 608 may include a machine learning model specific for and trained for each specific hydrotreater, due to the changes each piece of equipment may experience over time.

[0123] The hydrotreater controller 608 may obtain data related to each hydrotreater and / or initiate capture of samples associated with each hydrotreater. The sample collection and analysis assembly 614 may then analyze the samples to produce properties or spectra indicative of properties of the fluids associated with the hydrotreater. Once the hydrotreater controller 608 obtains data, properties, and / or spectra related to a hydrotreater, the hydrotreater controller 608 may apply the data, properties, and / or spectra to a machine learning model within one or more of the local enhancement module 610 or the predictive controls module 612. The output of the hydrotreater machine learning model may indicate adjustments to parameters and / or feed associated with the hydrotreater. Using such an output, the local enhancement module 610 may adjust the parameters and / or feed associated with the hydrotreater, for example via a PID controller, DCS controller, PLC controller, or a DCS-PID controller.

[0124] In some embodiments, the local enhancement module 610 may include programming and an algorithm configured to facilitate optimization of the hydrotreaters 632, 638 to achieve the target parameters based on the inputs, the target parameters, unit constraints, and the outputs of the hydrotreaters 632, 638 from one or more machine learning models (e.g., of the predictive controls modules 612), and inputs. The algorithm may be modeled by a machine learning model and configured to facilitate optimization of the hydrotreaters 632, 638 based on a machine learning model. In some embodiments, the local enhancement module 610 includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g. maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules 612), the target parameters, unit constraints, and inputs.

[0125] In an embodiment, the machine learning model utilized in the hydrotreater controller 608 may be trained to maximize sulfur removal at a lowest possible temperature.In an example, as a targeted product’s sulfur level or amount becomes lower, the hydrotreater temperature setting increase becomes even higher to achieve the same amount of sulfur removal. Thus, the output of the machine learning model of the hydrotreater controller 608 may indicate adjustment of hydrotreater severity to achieve maximum aromatic saturation of hydrotreater feed against hydrotreater operation constraints. Such an output may comprise a vector or other list of values indicative of equipment and / or device parameter settings.

[0126] In an embodiment, one of the hydrotreaters may be a naphtha hydrotreater. The naphtha hydrotreater may be utilized, in conjunction with a distillate hydrotreater, to treat wild naphtha. The treated wild naphtha may be utilized by one or more different refinery equipment, such as, but not limited to, a hydrocracker, a crude unit, and / or a gasoline desulfurization unit. In such embodiments, one trained machine learning model may be trained or configured to maximize the amount of treated naphtha produced in relation to the amount of naphtha used in the refinery equipment. The trained machine learning model may adjust the kerosene or flash target in a crude column to increase or adjust distillate transported to the naphtha or distillate hydrotreater. Further, the trained machine learning model may also determine an amount of heavy coker naphtha to transport to distillate hydrotreaters to thereby produce a minimum amount of wild naphtha via the naphtha hydrotreater.

[0127] FIG. 7 is a schematic diagram of a steam control system 1700 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. Many sections or portions of the refinery may utilize steam. For example, steam may be used in a FCC unit for stripping, while, in another unit, steam may be utilized as a heat source. As such, the sections or portions of a refinery that utilize steam may include a steam controller 1702. Similar to previously described controllers, the steam controller 1702 may obtain data associated with the equipment of the refinery, such as from one or more steam drums 1708, one or more boilers 1710, one or more waste heat sources 1712, one or more heat exchangers 1714, one or more condensers 1716, and one or more refinery equipment 1718 (in particular, refinery equipment 1718 that utilizes steam to some degree). The steam controller 1702 may apply the data and / or properties to one or more machine learning models of the local enhancement module 1704 and / or predictive controls module 1706 to produce an output indicative of adjustment to parameters. The steam controller 1702 may then utilize the output to adjust various parameters associated with refinery equipment that utilizes steam via the local enhancement module 1706.

[0128] The steam controller 1702 may prioritize steam to units or portions of the refinery that are generating high value products and / or products or intermediaries that are utilized to produce high value products. As such, at least one machine learning model for steam management may maximize steam production for one or more different refinery operations, while providing sufficient steam to other refinery operations that utilize steam.

[0129] FIG. 8 is a schematic diagram of hydrogen control system 1800 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. Similar to steam, several refinery operations may utilize hydrogen for various purposes. Further, a refinery may also produce some amount of hydrogen, either as a main product and / or as a by-product. Thus, hydrogen management may occur at the refinery level, rather than at a sub-level. Similar to previously described controllers, the hydrogen controller 1802 may obtain data associated with the equipment that produces hydrogen and / or utilizes hydrogen, such as from a steam methane reformer 1808, a catalytic reformer 1809, a hydrolysis unit 1810, an external hydrogen source 1812 (such data including an amount of hydrogen available at a selected times), and / or refinery equipment 1814 that utilizes hydrogen. The hydrogen controller 1802 may also initiate capture of samples of fluids associated with the devices and / or equipment that produce and / or utilize hydrogen. The sample collection and analysis assembly (not illustrated) may then analyze the samples and produce properties and / or a spectra for each sample. The hydrogen controller 1802 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 1804 and / or predictive controls module 1806 to produce an output indicative of adjustment to parameters and / or feed. The hydrogen controller 1802 may then utilize the output to adjust various parameters and / or feed associated with the hydrogen production units and / or equipment or devices utilizing hydrogen via the local enhancement module 1806.

[0130] The hydrogen controller 1802 may prioritize hydrogen to units or portions of the refinery that are generating high value products and / or products or intermediaries that are utilized to produce high value products, as well as maximize purity of hydrogen, such as via methane reforming and / or hydrolysis. As such, at least one machine learning model for hydrogen management may maximize hydrogen production or acquisition for one or more different refinery operations, while providing sufficient hydrogen to other refinery operations that utilize hydrogen.

[0131] FIG. 9 is a schematic diagram of a feed control system 1900 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. Eachrefinery operation utilizes a feed stream or feedstock. Such a feed stream or feedstock may, in some examples, be adjusted, to optimize or enhance fluid production of a corresponding refinery operation. Similar to previously described controllers, the feed controller 1902 may obtain data associated with the equipment that produces a feed or intermediary, blends a feed or intermediary, and / or utilizes a feed and / or intermediary, such as from one or more blend tanks 1908, one or more in-line blend pipes 1910, one or more feedstock sources 1912, one or more intermediary sources 1914, one or more sample collection and analysis assemblies 1916, and / or refinery equipment 1918. The feed controller 1902 may also initiate capture of samples of fluids associated with the devices and / or equipment that produce, blend, and / or utilize feed. The sample collection and analysis assembly (not illustrated) may then analyze the samples and produce properties and / or a spectra for each sample. The feed controller 1902 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 1904 and / or predictive controls module 1906 to produce an output indicative of adjustment to parameters and / or feed. The feed controller 1902 may then utilize the output to adjust various parameters and / or blends of a feed or intermediary associated with the refinery equipment 1918 via the local enhancement module 1906. The trained machine learning model utilized for the feed controller 1902 may be trained to or utilized to predict feed, blend, and / or intermediary properties and / or components that produce a maximum or greater than typical yield for a particular refinery operation.

[0132] FIG. 10 is a schematic diagram of a gasoline pool control system 2000 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. After processing various feeds via one or more different operations or sub-operations, various products from those operations or sub-operations may be combined to produce a selected gasoline or to form a selected gasoline pool. Such a gasoline pool may include a specification, the specification including one or more of an octane number, a research octane number, a motor octane number, a Reid vapor pressure, an amount of benzene, a density, a color, a flash point, an amount and / or type of lubricant, an amount and / or type of detergents, an amount and / or type of anti-rust agents, amount and / or type of anti-icing agents, and / or amount of sulfur or other contaminants, among other properties. To achieve the properties specified, a refinery controller and / or a gasoline pool controller 2002, for example may obtain analysis of and / or other data associated with various products within a refinery and select a percentage, e.g., the percentage in relation to the whole final gasoline product, of each product for combination or blending. As such, the refinery controllerand / or gasoline pool controller 2002 may obtain data from a variety of sources. For example, the gasoline pool controller 2002 may obtain data from a FCC unit 2008, a fractionation / distillation column 2010, an alkylation unit 2012, a gasoline desulfurization unit 2014, an isomerization unit 2015, a reformer 2016 or catalytic reformer, and / or an external gasoline source 2017 (for example, the external gasoline source 2017 may include a source that a refinery purchases and / or obtains gasoline from). Further, the gasoline pool controller 2002 may obtain data from the sample collection and analysis assembly 2018 and / or other refinery equipment 2020. In other embodiments, the gasoline pool controller 2002 may connect to a refinery controller and / or other controllers or sub-operation controllers within the refinery to obtain data related to that operation or sub-operation. Further, the gasoline pool controller 2004 may obtain data and / or properties from the collection and analysis assembly 2018 related to other fluids produced within the refinery. Once such data and / or properties has been collected, the gasoline pool controller may apply the data to one or more trained machine learning models stored in the local enhancement module 2004 and / or the predictive controls module 2006 to produce amounts of each varying fluid and / or properties associated with a fluid that causes production of the fluid to acheive selected properties. In an embodiment, the local enhancement module 2004 may apply the output of a plurality of trained machine learning models each stored in one or more of a plurality of predictive controls modules 2006 to a trained machine learning model of the local enhancement module 2004.

[0133] Once a prediction is determined by the gasoline pool controller 2004, the gasoline pool controller 2004 may adjust one or more devices within the refinery to cause those one or more devices to provide fluids for and / or adjust parameters to produce the selected or target gasoline. The selected components of the selected or targeted gasoline may then be blended.

[0134] In an embodiment, the trained machine learning models within the gasoline pool controller 2002 may be trained to or utilized for predicting parameters and / or properties for one or more of the units described in relation to the gasoline pool controller 2002. For example, if a specific property of a target gasoline pool is known or input into the trained machine learning model, along with other related data, then the gasoline pool controller 2002 may determine, based on application of that specific property and / or data to a trained machine learning model, process parameters and / or feed and / or intermediary properties for that particular process. In an embodiment, another controller specific for that operation may utilize the output to drive that operation to produce a fluid that accurately exhibits thatspecific parameter. In further embodiments, the gasoline pool controller 2002 may determine such parameters and / or properties for a plurality of other operations. In other embodiments, the gasoline pool controller 2002 may work in conjunction with other controllers to produce selected fluids. Thus, the gasoline pool controller 2002, in embodiments, may coordinate targeted outputs, by adjusting or causing adjustment of one or more refinery operations or sub-operations and / or by blending selected components from the refinery operations or sub-operations, from a plurality of operations to meet a selected gasoline pool specification.

[0135] In another embodiment, the gasoline pool controller 2002 may select a source of gasoline based on an output from the trained machine learning models. The trained machine learning models may output a blend percentage of gasoline from the sources to reach a target octane and / or volatility limit. In such embodiments, the volatility limit may comprise a RVP or vapor liquid ratio. In another embodiment, such a blend percentage of gasoline may comprise a higher percentage of gasoline from the external gasoline source 2017 and smaller percentages of gasoline and / or other fluids from other sources to reach the octane and / or volatility limit.

[0136] FIG. 11 is a schematic diagram of a diesel pool control system 2100 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. After processing various feeds via one or more different operations or sub-operations, various products from those operations or sub-operations may be combined to produce a selected diesel fuel or to form a selected diesel pool. Such a diesel pool may include a specification, the specification including one or more of a cetane number, a density, a color, a flash point, a pour point, an amount and / or type of lubricant, an amount and / or type of detergents, an amount and / or type of anti-rust agents, an amount and / or type of anti-icing agents, and / or an amount of sulfur or other contaminants, among other properties. To reach the properties specified, a refinery controller and / or a diesel pool controller 2102, for example may obtain analysis of and / or other data associated with various products within a refinery and select a percentage, the percentage in relation to the whole final diesel product, of each product for combination or blending. As such, the refinery controller and / or diesel pool controller 2102 may obtain data from a variety of sources. For example, the diesel pool controller 2102 may obtain data from a FCC unit 2108, a fractionation / distillation column 2110, a hydrocracker 2112, a hydrotreater 2114, a hydrodeoxygenation (HDO) unit 2115, and / or a coker unit 2116. Further, the diesel pool controller 2102 may obtain data from the sample collection and analysis assembly 2118and / or other refinery equipment 2120. In other embodiments, the diesel pool controller 2102 may connect to a refinery controller and / or other controllers or sub-operation controllers within the refinery to obtain data related to that operation or sub-operation. Further, the diesel pool controller 2102 may obtain data and / or properties from the collection and analysis assembly 2118 related to other fluids produced within the refinery. Once such data and / or properties has been collected, the diesel pool controller 2102 may apply the data to one or more trained machine learning models stored in the local enhancement module 2104 and / or the predictive controls module 2106 to produce amounts of each varying fluid and / or properties associated with a fluid that causes production of the fluid to reach selected properties. In an embodiment, the local enhancement module 2104 may apply the output of a plurality of trained machine learning models each stored in one or more of a plurality of predictive controls modules 2106 to a trained machine learning model of the local enhancement module 2104.

[0137] Once a prediction is determined by the diesel pool controller 2102, the diesel pool controller 2102 may adjust one or more devices (for example, refinery operation control devices) within the refinery to cause those one or more devices to provide fluids for and / or adjust parameters to produce the selected or target diesel. The selected components of the selected or targeted diesel may then be blended.

[0138] In an embodiment, the trained machine learning models within the diesel pool controller 2102 may be trained to or utilized for predicting parameters and / or properties for one or more of the units described in relation to the diesel pool controller 2102. For example, if a specific property of a diesel pool controller 2102 is known or input into the trained machine learning model, along with other related data, then the diesel pool controller 2102 may determine, based on application of that specific property and / or data to the trained machine learning model, process parameters and / or feed and / or intermediary properties for that particular process. In an embodiment, another controller specific for that operation may utilize the output to drive that operation to produce a fluid that accurately exhibits that specific parameter. In further embodiments, the diesel pool controller 2102 may determine such parameters and / or properties for a plurality of other operations. In other embodiments, the diesel pool controller 2102 may work in conjunction with other controllers to produce selected fluids. Thus, the diesel pool controller 2102, in embodiments, may coordinate targeted outputs, by adjusting or causing adjustment of one or more refinery operations or sub-operations and / or by blending selected components fromthe refinery operations or sub-operations, from a plurality of operations to meet a selected diesel pool specification

[0139] FIG. 12A and FIG. 12B are simplified diagrams of control systems 2700 to enhance to enhance fluid production at refinery, according to an embodiment of the disclosure. As noted, control system 2700 may include an operation controller 2701. Further, the operation controller 2701 may connect to one or more sensors 2716A, 2716B, and up to 2716N, one or more devices 2718 A, 2718B, and up to 2718N (such as flow control devices and / or temperature control devices), one or more equipment 2720A, 2720B, and up to 2720N, one or more analyzers 2722A, 2722B, and up to 2722N, and one or more predictive controls 2714A, 2714B, and up to 2714N. The operation controller 2701 may include memory 2704 and one or more processors 2702. The memory 2704 may store instructions executable by one or more processors 2702. In an example, the memory 2704 may be a non-transitory machine-readable storage medium. As noted, the memory 2704 may store or include instructions executable by the processor 2702.

[0140] As used herein, “signal communication” refers to electric communication such as hardwiring two components together or wireless communication, as understood by those skilled in the art. For example, wireless communication may be Wi-Fi®, Bluetooth®, ZigBee, or other near-field communications. In addition, signal communication may include one or more intermediate controllers or relays disposed between elements in signal communication.

[0141] The memory 2704 may include or store sample and data collection and instructions 2706. Upon execution of such instructions, the operation controller 2701 may obtain samples associated with each equipment 2720A, 2720B, and up to 2720N. Further, the operation controller 2701 may obtain data from the one or more sensors 2716A, 2716B, and up to 2716N and / or one or more flow control devices 2718 A, 2718B, and up to 2718N. Upon collection of the samples, the operation controller 2701 may send the sample toone of the one or more analyzers 2722A, 2722B, and up to 2722N. The one of the one or more analyzers 2722A, 2722B, and up to 2722N may then analyze the sample and generate properties and / or a spectra.

[0142] The operation controller 2701 may connect to and receive data from the one or more predictive controls 2714A, 2714B, and up to 2714N. In an embodiment, the operation controller 2701 may receive the output from each trained machine learning model of each the predictive controls 2714A, 2714B, and up to 2714N. In an embodiment, the output may comprise a vector or, in other embodiments, a value indicative of a parameter adjustment.

[0143] The memory 2704 may include or store trained machine learning models 2708. The trained machine learning models 2708 may include at least one trained machine learning model to generate an output indicative of parameter and / or feed adjustment. The operation controller 2701 may apply the data, properties, spectra, and / or the output of each trained machine learning model from one or more predictive controls 2714A, 2714B, and up to 2714N to the trained machine learning models 2708 to generate an output indicative of parameter adjustments and / or feed adjustment.

[0144] The memory 2704 may include or store parameter adjustment instructions 2710. Upon generation of the output, the operation controller 2701 may adjust parameters associated with equipment at the refinery. Further the memory 2704 may include or store feed adjustment instructions 2712 to adjust feed based on the output.

[0145] In FIG. 12B, predictive controls 2714 may connect to subsets of each of the components described in FIG. 12A. For example, the predictive controls 2714 may connect to a subset of the sensors 2736A, 2736B, and up to 2736N, a subset of the devices 2738A, 2738B, and up to 2738N, a subset of the equipment 2740A, 2740B, and up to 2740N, and / or a subset of the analyzers 2742A, 2742B, and up to 2742N. The predictive controls 2714 may include a trained machine learning model 2728 and instructions stored in a memory 2726 and executable by a processor 2724.

[0146] The instructions may include sample and data collection instructions 2730, which when executed cause the predictive controls 2714 to collect various data points and / or properties. Based on application of the data received to the trained machine learning model 2728 and, in some embodiments, an output from the operation controller 2701, the predictive controls 2714 may supply or provide the output to the operation controller 2701.

[0147] FIG. 13 is a flow chart illustrating enhanced fluid production at a refinery, according to an embodiment of the disclosure. Unless otherwise specified, the actions of method 2800 may be completed within an operation controller and / or predictive controls. Specifically, method 2800 may be included in one or more programs, protocols, or instructions loaded into the memory 2704 of operation controller 2701 and executed on the processor 2702 or one or more processors of the operation controller 2701 of FIG. 12A. In other embodiments, method 2800 may be implemented in or included in components of FIGS. 1 A-17. The order in which the operations are described is not intended to be construed as a limitation, and any number of the described blocks may be combined in any order and / or in parallel to implement the methods.

[0148] At block 2802, the one or more predictive controls may each obtain data from corresponding sources. In such an example, each of the predictive controls may poll corresponding sensors, flow control devices, equipment, temperature control devices, and / or other data generating sources related to a corresponding refinery unit to obtain data therefrom. Further, the predictive controls may initiate sample collection for inputs or feedstock, as well as intermediaries and / or products produced by the corresponding source. Such a sample may be analyzed by one or more spectroscopic analyzers, which may subsequently produce properties and / or spectra of the samples.

[0149] At block 2804, once each sub-controller has obtained data, each sub-controller may apply that data, which may also include one or more different properties and / or spectra, to a corresponding machine leaning model stored therein. In another embodiment, the targeted product and / or targeted properties may be applied, in addition to the data described above, to the machine learning model of the predictive controls. The machine learning model may produce vectors, indicators, and / or other values indicative of one or more parameters that may cause the corresponding source to produce a targeted product. Each of the parameters output from the trained machine learning model may correspond to some aspect of the source. For example, the parameters may include temperature, amounts of other fluids used in the operation (for example, hydrogen or alkanes, among others), pressure, flow rate, residence time, feedstock used, and / or intermediaries used.

[0150] At block 2806, each of the predictive controls may determine whether the subparameters are different than currently set parameters. If the parameters are different, then, in some embodiments and at block 2808, each of the predictive controls may adjust one or more of the equipment, devices, or fluids. In such embodiments, the predictive controls may provide the adjustments to an equipment and device controller. In another embodiment, the predictive controls may provide the adjustments to the operation controller 2701 (FIG. 12A) to perform such adjustments. In another embodiment, rather than or in addition to adjusting the equipment or devices, the predictive controls may provide the trained machine learning model output to the operation controller 2701.

[0151] At block 2810, the operation controller 2701 (FIG. 12A) may determine updated parameters and / or fluid contents and / or ratios based on application of the obtained data, as well as the outputs from each of the predictive controls, to a trained machine learning model. The operation controller 2701, in some embodiments, may first obtain data from the all, substantially all, a portion of the refinery, or from a corresponding operation. Data, as noted above, may include data from sensors, meters, equipment, and / or other devices, aswell as properties and / or spectra obtained from samples taken from each unit within the refinery. Further, the operation controller 2701 may obtain the output of each model of each predictive controls. Once the operation controller 2701 obtains all relevant data, the operation controller 2701 may determine the updated parameters based on application of that data to a machine learning model.

[0152] At block 2812, the operation controller 2701 (FIG. 12A) may determine whether the output of the model indicates updates to the parameters or whether the determined parameters are different than the current parameters. For example, the operation controller 2701 may compare the values of the updated parameters to the currently set parameters. If the parameters are different, then at block 2814, the operation controller 2701 may adjust the devices, equipment, or fluid within the refinery to the updated parameters.

[0153] In an embodiment, refinery operations may occur continuously or substantially continuously. As such, method 2800 may be an iterative and continuous process that occurs in real-time or near real-time. As target products change and / or other aspects of the refinery change, parameters may continue to be adjusted via method 2800.

[0154] FIG. 14 is a schematic diagram of a refinery 3500. Crude oil 3502 is initially processed by an atmospheric distillation tower 3504 into its constituent parts. Some of the lightest parts from the atmospheric distillation tower 3504 are a gas 3506 that is sent to gas processing 3508. Gas processing 3508 separates sour gas 3509 from the feeds thereto, which include gas 3506 and other gas 3507 resulting from different processes throughout the refinery. Gas processing 3508 diverts the sour gas 3509 to amine treating 3516. The remainder of the gas 3506 and other gas 3507 that is not sour gas is passed to the mercaptan treater 3510 that separates mercaptans from the fuel gas. The fuel gas may be finished as liquid petroleum gas (“LPG”) 3512. The separated butane 3514 may be kept as a finished product, sent to the gasoline blending pool, or sent to C4 isomerization 3591 to be processed into isobutane 3593 as a finished product or further sent to an alkylation unit 3594.

[0155] Amine treating 3516 separates the hydrogen sulfide gas 3520 from the refinery fuel 3518. The hydrogen sulfide gas 3520 and the hydrogen sulfide gas collected from processes throughout the refinery 3500 is sent to the sulfur plant 3522. The sulfur plant 3522 converts the hydrogen sulfide gas 3520 into sulfur 3530 as a finished product.

[0156] Additionally, sour water 3532 collected from processes throughout the refinery 3500 is sent to the sour water steam stripper 3528. The sour water steam stripper 3528 uses steam 3534 to remove hydrogen sulfide gas 3526 from the sour water 3532. The hydrogen sulfide gas 3526 is also sent to the sulfur plant 3522.

[0157] The atmospheric distillation tower 3504 separates light naphtha 3536 from the crude oil 3502. The light naphtha 3536 is sent to a hydrotreater 3538 that removes sulfur from the light naphtha 3536. The hydrotreated naphtha is then sent to isomerization 3540 to be processed into isomerate 3542. Isomerate 3542 are isomers of the light naphtha that have higher octane values. The isomerate 3542 may then be sent to the gasoline blending pool.

[0158] The atmospheric distillation tower 3504 separates heavy naphtha 3544 from the crude oil 3502. Optionally, the heavy naphtha 3544 may be sent to a splitter 3545. The splitter 3545 may include one or more splitter columns that separate heavier C7+ naphtha molecules from lighter components of the heavy naphtha 3544. The heavier C7+ naphtha may be sent to the hydrotreater 3554 to be desulfurized and produced as jet fuel 3558. The remaining lighter components of the heavy naphtha may be sent to a hydrotreater 3546 to remove sulfur from the heavy naphtha 3544 prior to being sent to the catalytic reformer 3548 to be converted into reformate 3550.

[0159] Alternatively, the heavy naphtha 3544 may be sent directly to the hydrotreater 3546 to remove sulfur from the heavy naphtha 3544 prior to being sent to the catalytic reformer 3548. From the hydrotreater 3546, the desulfurized heavy naphtha is sent to a catalytic reformer 3548 to be processed into reformate 3550. Reformate 3550 includes high-octane branched and cyclic hydrocarbons, such as benzene, toluene, xylene, and ethylbenzene. The reformate 3550 may then be sent to the gasoline blending pool.

[0160] Another product from the atmospheric distillation tower 3504 is jet fuel 3552. The jet fuel 3552 may include kerosene and other equivalent hydrocarbons. The jet fuel 3552 may be sent from the atmospheric distillation tower 3504 to a hydrotreater 3554 to remove contaminants from the jet fuel 3552, such as sulfur and mercaptans. Once desulfurized, the finished jet fuel 3558 may be sold or sent to the jet fuel blending pool to be blended with other products and additives and then made available for sale and distribution. The jet fuel 3552 may also be separated into kerosene (not shown).

[0161] Further, diesel 3560 is another product separated from crude oil 3502 by the atmospheric distillation tower 3504. The diesel 3560 is sent from the atmospheric distillation tower 3504 to a hydrotreater 3562 to remove sulfur. The desulfurized diesel 3564 may then be sold or sent to the diesel blending pool.

[0162] The atmospheric distillation tower 3504 also separates atmospheric gas oil 3566 and atmospheric bottoms 3568 from the crude oil 3502. The atmospheric bottoms 3568 may be sent to vacuum distillation 3570 where the atmospheric bottoms 3568 may befurther separated into light vacuum gas oil 3572 (“LVGO”), medium vacuum gas oil 3599 (“MVGO), heavy vacuum gas oil 3584 (“HVGO”), and vacuum residuum 3521. Vacuum distillation 3570 may also be configured to separate the atmospheric bottoms 3568 into more or less components.

[0163] The atmospheric gas oil 3566, the LVGO 3572, the MVGO 3599, and deasphalted oil 3501 from solvent deasphalting 3598 (“SDA”) may be sent to a hydrotreater 3574 to remove sulfur, then fed into a fluid catalytic cracker 3576. The fluid catalytic cracker 3576 processes the atmospheric gas oil 3566 and the light vacuum gas oil 3572 into naphtha 3578, jet fuel 3581, diesel 3582, fuel oil 3583, and butenes and pentenes 3592.

[0164] The butenes and pentenes 3592 from the fluid catalytic cracker 3576, as well as the isobutane 3593 from C4 isomerization 3591, may be sent to an alkylation unit 3594 where the feedstocks are processed into alkylate 3596. The alkylate 3596 may then be sold or sent to the gasoline blending pool. Alkylate 3596 has a high octane rating and low RVP.

[0165] The naphtha 3578 may be sent to a hydrotreater 3580 to further remove sulfur from the naphtha 3578. The desulfurized naphtha may then be sold or sent to the gasoline blending pool. The jet fuel 3581 may also be passed through a hydrotreater 3585 to remove sulfur and other contaminants and then may be sent to the jet fuel blending pool. The diesel 3582 may also be passed through a hydrotreater 3587 to remove sulfur and other contaminants and then sent to the diesel blending pool, and the fuel oil 3583 may be sold or stored for later sale and distribution. Lastly, the fuel oil 3583 may also be passed through a hydrotreater 3585 to remove sulfur and other contaminants and then sent to the fuel oil blending pool. In some embodiments, the fuel oil blending pool may be used to blend formulations of low sulfur fuel oil or ultra-low sulfur fuel oil.

[0166] The MVGO 3599, HVGO 3584, and the deasphalted oil 3501 may be sent to the hydrocracker 3586 to be processed into gasoline 3588, diesel 3590, and jet fuel 3595. The hydrocracker 3586 may also produce smaller hydrocarbons that may be sent to gas processing 3508. The hydrocracker 3586 process integrates desulfurization and other contaminant removal, so further hydrotreating is not necessary for its products. The gasoline 3588 may be sold or sent to the gasoline blending pool. The diesel 3590 may be sold or sent to the diesel blending pool. The jet fuel 3595 may be sold or sent to the jet fuel blending pool.

[0167] The vacuum residuum 3521 may be sent to solvent deasphalting 3598 (“SDA”) or used directly in asphalt 3519. In some applications, the vacuum residuum 3521 may be sent to the asphalt blending pool. The SDA 3598 may be used to extract lighter componentsfrom the vacuum residuum 3521 using solvents such as propane, butane, pentane, or a combination of these hydrocarbons to extract deasphalted oil 3501 from the vacuum residuum 3521. As discussed above, the deasphalted oil 3501 may be sent to the hydrocracker 3586 or the fluid catalytic cracker 3576 for refining into hydrocarbon products including naphtha, jet fuel, diesel, and fuel oil. Once the lighter components are removed, the remaining components may be referred to as pitch 3579. The pitch 3579 may also be used in the asphalt blending pool.

[0168] The vacuum residuum 3521 and pitch 3579 may also be sent to the coker 3597 to be processed into naphtha 3503 and processed through a gasoline desulfurization unit 3556 (“GDU”). The GDU 3556 may include one or more processes useful to remove contaminants from the naphtha 3503 in addition to gasoline desulfurization, including hydrotreating processes, catalysts, particulate catches, clay treaters, salt driers, and mercaptan treaters, separators, and steam strippers. The naphtha 3503 may then be sent to the gasoline blending pool. The coker 3597 also processes the vacuum residuum 3521 and pitch 3579 into jet fuel 3511, diesel 3513, fuel oil 3515, and petroleum coke 3517. The jet fuel 3511 may be passed through a hydrotreater 3523 and then sent to the jet fuel blending pool. The diesel 3513 may be sent to a hydrotreater 3525 to remove sulfur and then sent to the diesel blending pool. The fuel oil 3515 may also be processed through hydrotreater 3527 and then sent to the fuel oil blending pool.

[0169] The hydrotreaters 3538, 3546, 3554, 3562, 3574, 3580, 3585, 3587, 3589, 3523, 3525, and 3527 refer broadly to desulfurization and contaminant removal processes generally, including hydrotreating processes, clay treaters, salt driers, mercaptan treaters, gasoline desulfurization units, separators, steam strippers, filters, and particulate catches.

[0170] Many of the connections, feedstocks, and outputs are not shown, and those of skill in the art recognize that different configurations are possible and processes may be added or replaced by other processes known in the art. For example, atmospheric distillation tower 3504 and vacuum distillation 3570 may each represent multiple units set up in parallel or series, and may separate their feedstocks into more or fewer crude oil components. Additional equipment may be added to each process to further refine and separate the products of each process. For example, products and components of the products may be passed through isomerization, reformation, and alkylation processes not shown in FIG. 14 to process the products and components to meet regulatory requirements and customer specifications.

[0171] FIG. 15 is a schematic diagram of a machine learning model 3800 according to one embodiment. As used herein, a “machine learning model” refers to a computer algorithm or model, such as a classification model, a regression model, a language model, an object detection model, a multi-modal model, or an artificial intelligence, that can be trained and tuned based on training input to approximate unknown functions. For example, a machine learning model may refer to a neural network or other machine learning algorithm or architecture that learns and approximates complex functions and generates outputs based on a plurality of inputs provided to the machine learning model. Further, a “machine learning model” may refer to one or multiple machine learning models that cooperatively generate one or more outputs based on corresponding inputs. For example, a machine learning model may refer to any system architecture having multiple discrete machine learning components that consider different kinds of information or inputs.

[0172] The machine learning model 3800 is connected to a historical database 3802. The historical database 3802 may include multiple separate databases of information including process data, incident information and analysis, business information, demand planning, historical forecast information, historical pricing and purchasing data, and process training information. In some embodiments, the historical database 3802 may be a copy of historical databases that are maintained and updated by the machine learning model 3800. The machine learning model 3800 is also connected to a training database 3803 that includes the process data and historical information that is used for training of the machine learning model 3800.

[0173] As discussed in relation to other figures of this disclosure, the machine learning model 3800 may be used to analyze and improve specific processes, process controllers, and process interactions. The machine learning model 3800 may also be used as a refinery controller.

[0174] In this embodiment, the machine learning model 3800 may receive feedstock sensor, sample, and process data 3804 from a feedstock process controller 3814 or directly from the sensors, sampling and testing systems, and labs obtaining data from the feedstock process. A feedstock process is any process preceding the targeted process under review by the machine learning model 3800. For example, the feedstock process may include one or more of atmospheric distillation, vacuum distillation, filtering, hydrotreater, mercaptan treater, splitter, stripper, and separator.

[0175] Further, the machine learning model 3800 may receive targeted process sensor, sample, and process data 3806 from a targeted process controller 3816 or directly from thesensors, sampling and testing systems, and labs obtaining data from the targeted process. The machine learning model 3800 may also receive subsequent process sensor, sample, and process data 3808 from a targeted process controller 3816 or directly from the sensors, sampling and testing systems, and labs obtaining data from the targeted process. The feedstock sensor, sample, and process data 3804, the targeted process sensor, sample, and process data 3806, and the subsequent process sensor, sample, and process data 3808 may include operating pressure and temperature data for the components of each process and sub-process. The feedstock sensor, sample, and process data 3804, the targeted process sensor, sample, and process data 3806, and the subsequent process sensor, sample, and process data 3808 may include material composition data describing the material moving through the process, such as feed rates, and quantities and percentages of contaminants, reactants, and the hydrocarbons moving through the process.

[0176] The machine learning model 3800 may also access business data 3810. Business data 3810 may include demand planning information, forecasting information for various products, current market pricing and internal company pricing information, product distribution information, and information regarding the current and anticipated regulatory landscape. Business data 3810 may also include information regarding the costs, availability, and location of storage, disposal, and in some cases carbon capture options of waste products from each process. Business data 3810 may also include historical and current costs of each process and the refinery products, as well as current and historical market pricing for each product. Business data 3810 may also include the inventory levels of various products that may be stored in the storage tanks 3575 of the various blending pools.

[0177] The machine learning model 3800 may also access lab data 3811 that is obtained from laboratory analysis of samples collected from processes within a refinery and products produced by those processes. Lab data 3811 may include chemical composition and component distribution data about each sample, as well as information about when and where the sample was taken, tests performed on the sample, and sensor data from the process at the time the sample was taken.

[0178] The machine learning model 3800 may also access weather data 3813 as the ambient temperature, pressure, and solar radiation may also affect the temperature and pressure of feedstocks, products, processes, and storage tanks within a refinery. The machine learning model may build algorithms to predict the effect of weather on processoperating parameters, feedstocks, the resulting products from those processes, and the effect of weather on the storage of products within the blending pools.

[0179] The machine learning model 3800 may publish to and receive instructions from an engineering gateway 3812. The engineering gateway 3812 may act as a user interface for engineers to review process and machine learning model 3800 data, recommendations, warnings, and requests. A user may use the engineering gateway 3812 to assist the machine learning model 3800 in refining and tuning its algorithms to better predict and adjust each process in response to changes in ambient weather, demand seasonality, and composition and quality of feedstocks including the crude oil received for processing into refined hydrocarbons. The engineering gateway may also be used to facilitate active learning by the machine learning model 3800.

[0180] The machine learning model 3800 may access, receive data from, and provide instructions to a feedstock process controller 3814, a targeted process controller 3816, and a subsequent process controller 3818. In some embodiments, the machine learning model 3800 may adjust an algorithm used by the targeted process controller 3816 based on changes identified from the adjusted algorithm. Once an adjusted algorithm has been selected by the machine learning model 3800, the machine learning model 3800 requests approval to implement the adjusted algorithm through the engineering gateway 3812. Upon receipt of approval, the machine learning model 3800 saves a copy of the approval information in the historical data 3802 and sends the approved adjusted algorithm to the targeted process controller 3816.

[0181] The machine learning model 3800 includes a data analysis module 3822. The data analysis module 3822 may be used by the machine learning model 3800 to review data and identify data that may be considered outliers and disregarded. Once the data set has been reviewed and amended to remove outlier data, the machine learning model 3800 may save the revised data set in either the training data 3803 or the historical data 3802 for retraining of the machine learning model 3800, for analysis and adjustment of a targeted process, or for later use.

[0182] The machine learning model 3800 includes an interpolation module 3824. The interpolation module 3824 may be used to analyze a data set and interpolate missing data from the data set. The interpolation module 3824 may compare the historical data 3802 with feedstock sensor, sample, and process data 3804, the targeted process sensor, sample, and process data 3806, and the subsequent process sensor, sample, and process data 3808 to identify process changes. The interpolation module 3824 may be able to identify and flagsmall deviations for further investigation. The interpolation module 3824 works with the communication module 3828 to publish the data regarding the flagged deviation for user review and guidance. For example, the interpolation module 3824 may be used to identify miscalibrated sensors or test data that may be inaccurate. The machine learning model 3800 may accomplish this by identifying and labeling data as outliers. Labeled data may be associated with a sensor or data and then reported to a user through the engineering gateway 3812. The machine learning model 3800 may use the interpolation module 3824 to identify components that may be wearing out and catalysts that may be deactivated or poisoned. The interpolation module 3824 may also communicate through the communication module 3828 with the engineering gateway 3812 to identify these potential process concerns to a user for further investigation.

[0183] The machine learning model 3800 includes a prediction module 3826. The prediction module 3826 may analyze the feedstock sensor, sample, and process data 3804, the targeted process sensor, sample, and process data 3806, and the subsequent process sensor, sample, and process data 3808 to predict changes in a process. For example, temperature excursions may be predicted in a process and the prediction module 3826 may communicate through the communication module 3828 with the engineering gateway 3812 to notify a user of a predicted temperature excursion and recommending process changes to avoid the temperature excursion. A user may accept the recommended process changes through the engineering gateway 3812 or provide different instructions for the machine learning model 3800 to implement. In predicting changes in a process, prediction module 3826 may recommend regeneration or replacement of a catalyst or repair or replacement of a component of a process as part of a maintenance procedure. The prediction module 3826 may provide a recommended maintenance window for the maintenance procedure to occur.

[0184] The machine learning model 3800 includes a communication module 3828. The communication module 3828 may translate recommendations, instructions, and data from other modules from machine code into a human language or convert data into graphs and other visual communication elements. The communication module 3828 may send or publish communications to the engineering gateway 3812. The communication module 3828 may also communicate with the various process controllers and other models used by the refinery.

[0185] In some embodiments, the machine learning model 3800 includes a confidence module 3830. The confidence module 3830 may review the analysis and recommendations of the different modules of the machine learning model 3800 and assign a confidence levelto the analysis and recommendations. For example, the confidence module 3830 may produce SHAP values, or use LIME or anchors to analyze each algorithm that is adjusted or created by the machine learning model 3800. The results may be published by the communication module 3828 to the engineering gateway 3812 to assist a user in reviewing each algorithm and the changes recommended by the machine learning model 3800.

[0186] In some embodiments, the machine learning model 3800 includes a regulatory module 3832. The regulatory module 3832 may access the business data 3810 to assist in regulatory compliance. For example, the regulatory module 3832 may flag a process that may be predicted by the prediction module 3826 to move out of compliance. The regulatory module 3832 may issue a warning through the communication module 3828 to the engineering gateway 3812. The regulatory module 3832 may also identify windows of time when regulations are lessened and recommend changes to process parameters to reduce process costs. The regulatory module 3832 may make recommendations to gasoline blending pool controllers, diesel blending pool controllers, and jet fuel blending pool controllers to adjust blending parameters in line with upcoming regulatory changes. Further, the regulatory module 3832 may make recommendations for various process controllers to adjust their parameters to promote the production of blend components inline with demand that may accompany regulatory changes.

[0187] In some embodiments, the machine learning model 3800 includes a forecasting module 3834. The forecasting module 3834 may analyze business data 3810 and recommend that various process controllers adjust their parameters to promote the production of components that may be in greater demand. The forecasting module 3834 may also recommend maintenance windows for equipment producing products that may be in low demand during specific time frames. The forecasting module 3834 may forecast pricing for various components based on the historical data 3802 and the business data 3810 and publish the forecasted pricing via the communication module 3828 to the engineering gateway 3812.

[0188] In some embodiments, the machine learning model 3800 may include an authority module 3836 and additional modules 3838, such as a display module for converting data into graphical representations of a data set or a translation module for converting data between different languages, formats, or units. The authority module 3836 may track user instructions and approvals for various instructions and changes to be made by the machine learning model 3800 to the various controllers throughout the refinery. In some applications, the authority module 3836 may consider a recommendation from a module ofthe machine learning model 3800 and automatically authorize the machine learning model 3800 to issue an instruction to the targeted process controller 3816 to make a change to an associated process. In other applications, the authority module 3836 may direct the communication module 3828 to request approval through the engineering gateway 3812 before a recommendation may be implemented and instructions sent to the targeted process controller 3816.

[0189] The machine learning model 3800 provides better process control and faster changes to processes within a refinery. The machine learning model 3800 also provides better data and clarity to changes and predicted changes within the nonlinear processes found within a refinery.

[0190] FIG. 16 is a schematic diagram of a hydrotreater 3900. The hydrotreater 3900 may be used to remove sulfur, nitrogen, and oxygen from various feedstock 3902 and to saturate olefins and aromatics of the feedstock 3902. The hydrotreater 3900 may also be used to remove metals, including silicon, arsenic, and sodium from organo-metallic molecules of the feedstock 3902. The hydrotreater 3900 may also be used to improve properties of feedstock 3902, including the stability, smoke point, and corrosiveness of jet fuel, the cetane number of diesel and distillates, and the lower freeze, cloud, and pour points of waxes that may be found in diesel and heavier hydrocarbon feedstocks. The hydrotreater 3900 may also remove potential catalyst poisons and process inhibitors from feedstocks 3902. Depending on the feedstock 3902, the hydrotreater 3900 is used to remove contaminants from feedstocks used in other refinery operations, sent to a blending pool to be blended into a final product, or produce a final product for use or sale.

[0191] The hydrotreater 3900 may be configured to hydrotreat feedstock 3902, including light and heavy naphtha, kerosene and jet fuel, diesel fuel, distillates, atmospheric gas oil, light vacuum gas oil, medium vacuum gas oil, and deasphalted oil, as well as products from various processes including fluid catalytic cracker products and coker products. For example, feedstock 3902 may be obtained from multiple sources within a refinery including atmospheric distillation 3504, vacuum distillation 3570, coker 3597, hydrocracker 3586, and fluid catalytic cracker 3576 as shown in FIG. 14. The hydrotreater 3900 may include a reactor that may be used as a hydrocracker. Feedstock 3902 may include streams of feedstock from multiple sources.

[0192] As shown, the hydrotreater 3900 may include a feed drum 3904 that receives feedstocks 3902 from one or more sources with each feedstock 3902 from a different source having different properties and compositions of hydrocarbons and contaminants. The feeddrum 3904 may mix the feedstocks 3902 and may be used to provide a more uniform flow rate of feedstock 3902 to the hydrotreater 3900. The feed drum 3904 may include a pump to pressurize and pump the feedstock 3902 to a charge heater 3906. The charge heater 3906 heats and mixes the feedstock 3902 with hydrogen gas 3908 in preparation to be reacted in an optional reactor 3910 in the presence of catalyst 3912 and then in reactor 3914 in the presence of catalyst 3916. An optional reactor 3952 may also be used to hydrotreat or hydrocrack the feedstock 3902. In some configurations, the optional reactor may represent a plurality of hydrocracker reactors for selectively cracking large polyaromatic molecules. If configured as a hydrocracker, different catalysts, such as amorphous silicon aluminum catalysts or zeolite containing catalysts, may be selected to facilitate cracking of the larger hydrocarbons. Additionally, the optional reactor 3952 may be operated at higher temperatures and pressures to enhance cracking of hydrocarbons in the feedstocks 3902.

[0193] Properties, composition, and feed rates of feedstock 3902 as feedstock 3902 moves through the hydrotreater 3900 may be measured by sensor packages 3920, 3922, 3924, 3926, 3928, 3930, 3932, 3960. The sensor package 3920 may be disposed to receive the operational parameters of and measured data from the upstream processes providing the feedstock 3902 to the hydrotreater 3900. The sensor package 3920 may also be disposed to measure, analyze, and sample the feedstock 3902 as it is fed into the feed drum 3904. The sensor package 3922 is disposed to measure the operating parameters of the feed drum 3904 and the feedstock 3902 within it. In particular, the sensor package 3922 may include a fluid level sensor to monitor the fill rate and fill status of the feed drum 3904 to prevent over filling and emptying the feed drum 3904. The sensor package 3924 may be disposed to measure the properties and composition of feedstock 3902 as it exits the feed drum 3904 and enters the charge heater 3906. The sensor package 3926 is disposed to measure the operating parameters of the charge heater 3906. The sensor package 3928 is disposed to measure the properties and composition of feedstock 3902 as the feedstock 3902 enters the optional reactor 3910. The sensor package 3930 may be disposed to measure the properties and composition of feedstock 3902 entering the reactor 3914 and sensor package 3932 may be disposed to measure the properties and composition of the hydrotreated feedstock 3902 exiting the reactor 3914. The sensor package 3960 may be disposed to measure the properties and composition of feedstock 3902 exiting the optional reactor 3952.

[0194] The sensor packages 3920, 3922, 3924, 3926, 3928, 3930, 3932, 3960 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensors and differential pressuresensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 shown in FIG. 6.

[0195] All of the data from sensor packages 3920, 3922, 3924, 3926, 3928 may be referred to as feedstock data that may be sent to the hydrotreater controller 608 and a machine learning model installed in the controller as part of local enhancement 610, predictive controls 612, or a machine learning model in communication with the hydrotreater controller 608 as shown in FIG. 6. The feedstock data may be used by the machine learning model to determine or predict the properties and composition of the feedstock 3902. In some applications, the feedstock data may be used by the machine learning model to predict the properties and composition of the feedstock 3902 in real time or near real-time, with lab and analyzer data from samples taken from the feedstock 3902 being used periodically by the machine learning model to calibrate or adjust the predicted properties and composition of the feedstock 3902.

[0196] The data from sensor packages 3930, 3932, 3944 may be referred to as hydrotreating data that may be sent to the hydrotreater controller 608 and a machine learning model installed in the controller as part of local enhancement 610, predictive controls 612, or a machine learning model in communication with the hydrotreater controller 608 as shown in FIG. 6. The hydrotreating data may be used to predict the effectiveness of the hydrotreating process and determine when and what adjustments to the hydrotreating process should be made to achieve target product properties. The hydrotreating data may also be used to calibrate the machine learning model and to refine the operational parameters of the reactors 3910, 3914, 3952 within a hydrotreater 3900.

[0197] Hydrogen gas 3908 may be added to the feedstock 3902 at various points throughout the hydrotreater 3900. For example, hydrogen gas 3908 may be mixed with the feedstock 3902 as the feedstock is moving between the feed drum 3904 and passed into the charge heater 3906. Hydrogen gas 3908 may also be added to the feedstock 3902 as the feedstock 3902 moves from the charge heater 3906 to the optional reactor 3910. Hydrogen gas 3908 may also be added to the feedstock 3902 as the feedstock 3902 moves to and enters the reactor 3914 and the optional reactor 3952. Hydrogen gas 3908 may be fed directly into the optional reactor 3910, the reactor 3914, and the optional reactor 3952 tocompensate for hydrogen gas 3908 consumed by the hydrotreating process in the optional reactor 3910 and the reactor 3914. Hydrogen gas 3908 may also be used to quench and assist in temperature control of the exothermic processes within the optional reactor 3910 and the reactor 3914.

[0198] Sensor packages 3934, 3936, 3938, 3940, 3942, 3962, 3964 may be disposed to measure, analyze, and sample the hydrogen gas 3908 being fed into the hydrotreater 3900. The sensor packages 3934, 3936, 3938, 3940, 3942, 3962, 3964 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensors and differential pressure sensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 shown in FIG. 6.

[0199] As the feedstock 3902 flows through the optional reactor 3910, the reactor 3914, and the optional reactor 3952, the hydrogen gas 3908 is reacted with the feedstock 3902 in the presence of catalysts 3912, 3916, 3954 that may provide hydrodesulfurization, hydrodenitrogenation, hydrodeoxygenation, olefin and aromatic saturation, hydrodemetallization, and potentially, hydrocracking of larger hydrocarbons. In some configurations, the hydrotreater 3900 may include a single reactor or two or more reactors in series or parallel that may be reactors designed to emphasize one or more different hydrotreating processes within the hydrotreater 3900.

[0200] Hydrodesulfurization is the process that separates sulfur in the form of hydrogen sulfide from the hydrocarbons of the feedstock 3902. The larger the aromatic molecule is that contains sulfur, the higher the temperature is required to remove the sulfur. However, the higher the temperature, the more likely hydrocracking occurs, which may increase the formation of lighter hydrocarbons. Further, hydrodesulfurization may be inhibited by the presence of nitrogen, hydrogen sulfide, carbon monoxide, and PNA condensation. PNA condensation may be controlled through sharper distillation cuts upstream to reduce PNA in the feedstock 3902.

[0201] Hydrodenitrogenation is the process that separates nitrogen in the form ammonia that begins by saturating the nitrogen containing ring. The partial pressure of hydrogenshould be high to enhance the saturation and the hydrotreater reactor temperature should be high enough to facilitate the separation of nitrogen from the hydrocarbon molecule.

[0202] Hydrodeoxygenation is the process that separates oxygen in the form water from the hydrocarbons of the feedstock 3902. The partial pressure of hydrogen should be high to enhance the separation of oxygen from the hydrocarbon to form water.

[0203] Olefin and aromatic saturation is highly exothermic and may facilitate hydrodesulfurization, hydrodenitrogenation, and hydrodemetallization processes. Additionally, failure to saturate olefins in the feedstock 3902 may result in the formation of polymerized gums and reformation of mercaptans.

[0204] Hydrodemetallization is the process that separates metals, including vanadium, nickel, silicon, and arsenic, from hydrocarbons in the feedstock 3902. When separated, the metals may deposit on the catalysts 3912, 3916 resulting in deactivation of the catalysts 3912, 3916 over time.

[0205] Hydrocracking occurs at higher temperatures and may generate light hydrocarbons. In some applications, paraffins may be hydrocracked to form branched isoparaffins. Aromatics may also be cracked to form paraffins or to permit removal of sulfur, nitrogen, oxygen, and metals from large organo-metallic molecules.

[0206] The catalysts 3912, 3916 may be made of different layers of different catalyst types and functionality, including nickel molybdenum catalysts, cobalt molybdenum catalysts, combinations of different catalysts, and other known catalysts. The type of catalysts used in a reactor and the composition of feedstock to be hydrotreated may be used to determine operating parameters of temperatures, pressures, and flow rates. In other words, the catalysts determine the types of feedstocks that may be successfully hydrotreated to achieve target product properties as provided by customer specifications or regulatory requirements.

[0207] As the feedstocks 3902 and hydrogen gas 3908 are fed into the optional reactor 3910, the properties of the feedstocks 3902 and hydrogen gas 3908 may be measured, analyzed, and sampled by a sensor package 3928. Sensor package 3944 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to the inlet of the optional reactor 3910 or proximate to the beginning of the catalysts 3912. A second sensor package 3946 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to one or more injection points for quenching hydrogen gas 3908 into the reactor 3910, proximate an end of the catalysts 3912, or the exit of the reactor 3910.

[0208] The optional reactor 3910 may be used to hydrotreat and trap contaminants that may cause poisoning or deactivation of the catalysts 3912, 3916, saturate the olefins and di olefins in the feedstocks 3902 before reaching the reactor 3914, and to reduce pressure drops in the following reactors by concentrating as much of the hydrogenation of the feedstocks 3902. This configurations may permit the catalysts 3916 of the reactor 3914 and the optional reactor 3952 to last longer and operate at lower temperatures to effectively and efficiently remove the contaminants from the hydrocarbons in the feedstocks. The optional reactor 3910 may use regenerated or less expensive catalysts 3912 to reduce operating costs per barrel of hydrotreated feedstock 3902. Further, cooling resources may be used to remove the exothermic heat released by the hydrogenation of the feedstocks 3902 in preparation for further hydrotreating. In some configurations, the reactor 3910 may include a plurality of reactors, arranged in parallel or series, to facilitate continued operation while one may have its catalysts removed and replaced.

[0209] The properties of the feedstocks 3902 and hydrogen gas 3908 fed into the reactor 3914 may be measured, analyzed, and sampled by a sensor package 3948. Sensor package 3948 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to the inlet of the reactor 3914 or proximate to the beginning of the catalysts 3916. A second sensor package 3950 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to one or more injection points for quenching hydrogen gas 3908 into the reactor 3914, proximate an end of the catalysts 3916, or the exit of the reactor 3914.

[0210] The reactor 3914 may be used to remove silicon, arsenic, sulfur, and nitrogen, and to saturate any remaining olefins and diolefins in the feedstocks 3902. Temperatures and pressures across the reactor may be managed through cooling and the use of quenching provided through injection of hydrogen gas 3908. Temperatures and the change of temperature across the catalysts 3916 may be managed to reduce the formation of mercaptans within the reactor 3914.

[0211] The properties of the feedstocks 3902 and hydrogen gas 3908 fed into the optional reactor 3952 may be measured, analyzed, and sampled by a sensor package 3956. Sensor package 3956 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to the inlet of the optional reactor 3952 or proximate to the beginning of the catalysts 3954. A second sensor package 3950 may be disposed to measure the properties of feedstocks 3902 and hydrogen gas 3908 proximate to one or more injectionpoints for quenching hydrogen gas 3908 into the optional reactor 3952, proximate an end of the catalysts 3954, or the exit of the optional reactor 3952.

[0212] The optional reactor 3952 may be used to reduce recombinant sulfur and to complete hydrotreating the feedstocks 3902. The optional reactor 3952 may also use hydrogen gas 3908 to quench the temperatures within the optional reactor 3952 and to prevent the partial pressure of hydrogen gas from falling too low. In some configurations, the

[0213] The sensor packages 3944, 3946, 3948, 3950, 3956, 3958 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensors and differential pressure sensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 shown in FIG. 6. In particular, the sensor packages 3944, 3946, 3948, 3950, 3956, 3958 may be used to determine a weighted average bed temperature of the catalysts 3912, 3916, 3954, the temperature change across the catalysts 3912, 3916, 3954, and a pressure drop across each of the catalysts 3912, 3916, 3954.

[0214] In general, the operating parameters of a hydrotreater that may be adjusted by a machine learning model include hydrogen gas pressure at an outlet of a reactor, hydrogen gas concentration within a reactor, and liquid hourly space velocity or LHSV that is the volumetric flow rate of feedstock per hour divided by the volume of catalyst in the reactor. In other words, LHSV represents the residence time within the reactor of a feedstock relative to the volume utilization of catalysts within the reactor. Operating parameters of a hydrotreater also include the ratio of hydrogen gas to the feedstocks, the composition of gasses within the reactor including but not limited to hydrogen gas, hydrogen sulfide, carbon monoxide, and nitrogen gas, reactor pressure, reactor exotherm, available cooling, the weighted average bed temperature representing the average temperature across the catalysts in the reactor, and the pressure and temperature differentials across the catalysts in the reactor.

[0215] Hydrogen gas availability throughout a reactor may affect catalyst deactivation rates, flow capacity through the reactor or LHSV, and the temperatures needed for theseparation of contaminants from the feedstock 3902. Hydrogen starvation within a reactor may occur if the hydrogen gas feed rates are too low or if consumption of hydrogen gas exceeds the available hydrogen gas within an area of the reactor. Consequently, by accurately predicting or determining the composition of feedstocks 3902, hydrogen feed rates may be predicted by a machine learning model trained with the historical data of the reactor’s operation, historical feedstock data, and historical product data.

[0216] The partial pressure of hydrogen gas may be determinative of the operating pressure within the reactor and the consumption of hydrogen across the catalysts is indicated by the pressure drop across the catalysts. The use of quenching within a reactor by injecting hydrogen gas may reduce the pressure drop across the catalysts and assist in preventing hydrogen starvation within the reactor.

[0217] As hydrogen is consumed through saturation of the feedstock 3902 and removal of contaminants, heat is released by the exothermic reaction. Consequently, feed rates and composition of feedstocks 3902, activity of the catalysts, and cooling of the reactors may determine reactor temperatures. The difference in temperature between the beginning of the catalysts and the end of the catalysts in the reactors is indicative of catalyst activity. In other words, as the activity of the catalyst declines with coking and deactivation, the temperature differential reduces between the inlet and outlet reactor temperatures.

[0218] Operating objectives include sulfur, nitrogen, and other contaminant removal from the feedstocks. The resulting hydrotreated feedstocks may be sent to product pools for sale and blending into other products, or may be sent to other processes, including reforming, isomerization, hydrocracking, or fluid catalytic cracking.

[0219] Depending on the feedstock, a specific product property may be enhanced through hydrotreating. In distillate and gas oil hydrotreaters, saturating the double bonds in aromatics, olefins, and diolefins in the feedstock results in an increase in the volume of the feedstock (also referred to as volumetric swell), thus, increasing product value. Volumetric swell may be balanced with catalyst deactivation on the reactors. A machine learning model may be used to predict a target reactor temperature based on available cooling, feedstock composition, and available hydrogen gas feed rates to achieve a target aromatic saturation in the reactors. This is a nonlinear correlation between multiple potential inputs.

[0220] The machine learning model may be trained with historical data gathered from one or more hydrotreaters and in the case of an existing hydrotreater, the historical data of the existing hydrotreater. The historical data includes feedstock properties and composition and product properties and composition, as well as the operating parameters used tohydrotreat those feedstocks and products. Additionally, the machine learning model may be trained on hydrogen consumption within each reactor and the effect of different processes on reactor temperature.

[0221] Further, a machine learning model may be trained to achieve a target sulfur content in the hydrotreated product, but if the target or maximum aromatic saturation temperature is higher, the machine learning model may operate the reactor at the higher temperature in order to achieve the target or maximum aromatic saturation of the feedstocks. The machine learning model may also be instructed that if the reactor temperature predicted to achieve a target sulfur content is higher than the target or maximum aromatic saturation temperature, then the reactor temperature predicted to achieve a target sulfur content may be implemented by the machine learning model.

[0222] Additionally, the machine learning model may predict a target reactor pressure for one or more locations within the reactor including proximate the inlet, proximate the injection points of hydrogen gas used to quench the reactor, or proximate the outlet of the reactor. Pressure may be increased by the partial pressure of hydrogen gas mixed with the feedstocks 3902 or the partial pressure from vaporization of feedstock. For distillate and gas oil hydrotreaters, a higher pressure throughout the reactor may increase volumetric swell of the hydrotreated feedstocks and may reduce coking on the catalysts resulting in deactivation.

[0223] In contrast for naphtha hydrotreaters, saturation of the naphtha may be undesirable as the resulting octane number of the naphtha will fall. Additionally, lower operating temperatures and temperature differences between the inlet and outlet of a reactor may be used in an effort to maintain higher octane numbers by not over treating or over saturating aromatics and olefins, while still separating nitrogen and sulfur from the feedstocks 3902.

[0224] In some applications, a small number of olefins may be formed through cracking near a reactor outlet. The remaining olefins and the newly formed olefins may form recombination mercaptans. The recombination mercaptans may become significant at high operating temperatures in a reactor or at lower temperatures in the presence of iron sulfide. Additionally, FCC and coker naphtha may contain high levels of nitrogen that may inhibit some catalysts and may require higher operating temperatures to separate the nitrogen from the feedstocks 3902.

[0225] Hydrotreating kerosene and jet fuel may be managed to improve smoke point through saturation of aromatics in the feedstock 3902. Kerosene and jet fuel may include some naphtha that may also be saturated during hydrotreating, which may reduce the flashpoint of the hydrotreated feedstock 3902. A machine learning model may direct upstream processes, especially distillation, to sharpen cuts of feedstock 3902 to reduce the concentrations of difficult to treat sulfur species such as benzothiophenes and dibenzothiophenes, color bodies, and larger aromatic molecules. When difficult to treat sulfur species are present in higher concentrations, accelerated deactivation of the reactor catalysts may occur. Thus, hydrotreater performance may change rapidly when the quality of kerosene and jet fuel feedstocks decrease. If better quality feedstocks are used, catalyst performance may recover from hard use and improve slowly over time. Consequently, a machine learning model may select and generate an order of feedstocks to be processed to increase catalyst life while meeting processing requirements.

[0226] When hydrotreating kerosene and jet fuel feedstocks, the operating temperature should be kept low with a high partial pressure of hydrogen gas to obtain product properties, including color and smoke point. Because operating temperatures may be close to the vaporization temperature of the kerosene and jet fuel feedstocks, some vaporization may result in flow maldistribution and low liquid mass flow while concentrating the difficult to treat sulfur species in the remaining liquid feedstocks resulting in discolored out-of- specification product. Consequently, a machine learning model may raise the operating temperature of the reactor to ensure all of the feedstock is in the vapor phase to facilitate hydrotreating.

[0227] Like hydrotreating kerosene and jet fuel feedstocks, diesel or distillate, and gas oil hydrotreating may also be improved through a sharper distillation cut of the upstream products to reduce concentrations of difficult to treat sulfur species including dibenzothiophenes. If difficult to treat sulfur species are not removed, the operating temperatures of the reactors may be raised resulting in higher hydrogen gas consumption and increased coking on the catalysts resulting in accelerated deactivation of the catalysts. The distribution of hydrogen gas within the feedstocks 3902 and the reactors 3910, 3914, 3952 may be better controlled to avoid regions of low hydrogen gas partial pressure resulting fouling and catalyst deactivation.

[0228] Hydrotreating diesel may be used to increase volumetric output through swelling of diesel volume and to improve the properties of the hydrotreated diesel. For example, cold flow properties may be improved through selective cracking of n-paraffin chain length resulting in decreased freeze, cloud, and pour points. Further, cetane number may also be improved through hydrotreating as the aromatics are saturated. As the saturation occurs,the volume of the feedstock increases as hydrogen is added to the molecular structure of the feedstocks 3902.

[0229] To achieve ultra-low sulfur diesel specifications, higher operating temperatures and lower feedstock feed rates may be used. The higher temperatures and lower feed rates may increase hydrogen gas consumption and may also increase fouling and catalyst deactivation. Eventually, the hydrotreater constraints in operating temperature or hydrogen gas feed rates may be reached, which may indicate the catalysts may deactivated to the point that the catalysts should be replaced or regenerated.

[0230] Hydrotreating cracked feedstocks may shorten the useful life of catalysts within a reactor. Cracked feedstocks may contain more sulfur and nitrogen containing compounds and more olefins and aromatics. Consequently, higher operating temperatures within the reactor may be used to hydrotreat the cracked feedstocks and hydrogen consumption may be increased. Further, more coking and deactivation of the catalysts may occur with hydrotreating cracked feedstocks.

[0231] After the feedstocks 3902 have been hydrotreated in the reactors 3910, 3914, 3952, the hydrotreated feedstocks 3902 are then passed into a separator 3966. The separator 3966 may include a sensor package 3968 including one or more of a temperature sensor and a pressure sensor disposed with a portion of the separator 3966. The separator 3966 separates the feedstocks 3902 into sour water 3978, which may include ammonium salts and sulfides, a lighter portion, which may include excess hydrogen gas and light hydrocarbons that is sent to a recycle gas scrubber 3970, and the remaining hydrotreated product. The sensor package 3968 may include one or more of temperature sensors, pressure sensors, flow rate sensors, composition sensors, such as spectrometers and gas chromatographs, and sampling units to obtain samples of the hydrotreated feedstocks 3902 in the separator 3966.

[0232] The lighter portion is treated with an amine, such as diethanolamine solution in the recycle gas scrubber 3970 to remove hydrogen sulfide. The hydrogen sulfide separated from the hydrotreated feedstocks 3902 may be dissolved in water to form sour water 3978 in the separator 3966. The sour water 3978 may be sent for further processing at a sour water steam stripper. The composition of the sour water 3978 may be sampled, analyzed, measured by a sensor package 3980.

[0233] The hydrogen gas 3972 may be recycled and sent to be mixed into the feedstocks 3902 at one or more locations in the hydrotreater 3900. The composition or purity of the recycled hydrogen gas 3972 may be monitored by a sensor package 3974. Methane maybuild up in the recycled hydrogen gas 3972 and may be purged periodically as purge gas 3976 to improve the purity of the recycled hydrogen gas 3972.

[0234] From the separator 3966, the hydrotreated feedstocks 3902 enters the stabilizer 3982. The stabilizer 3982 further separates the hydrotreated feedstocks 3902 into gases, light hydrocarbons, and hydrotreated products 3990. The gases may be sent to further processing to remove any remaining hydrogen sulfide gas. The light hydrocarbons may be sent to further processing or used to generate steam. The hydrotreated products 3990, which depends on the feedstocks 3902, may include desulfurized naphtha, desulfurized kerosene and jet fuel, desulfurized diesel, and desulfurized gas oil. A sensor package 3988 may include temperature sensors, pressure sensors, flow rate sensors, composition sensors such as spectrometers and gas chromatographs, and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 of FIG. 6. The sensor package 3988 may disposed to measure, sample, and analyze the properties and composition of the hydrotreated products 3990.

[0235] The stabilizer 3982 may include a sensor package 3984 disposed before or as the hydrotreated feedstocks 3902 enter the stabilizer 3982 and a sensor package 3986 disposed within the stabilizer 3982. The sensor packages 3984, 3986 may measure, analyze, or sample the hydrotreated feedstocks 3902. The sensor packages 3984, 3986 may include one or more of temperature sensors, pressure sensors, flow rate sensors, composition sensors such as spectrometers and gas chromatographs, and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 of FIG. 6.

[0236] Optionally, the hydrotreater 3900 may include a parallel line 3931 of one or more hydrotreating reactors. The reactors may include a reactor 3901, a reactor 3903, and a reactor 3905, or other reactors (not shown). Like reactors 3910, 3914, 3952, the reactors3901, 3903, 3905 include catalysts 3907, 3909, 3922 used to hydrotreat the feedstocks3902. The reactors 3901, 3903, 3905 may include a supply of hydrogen gas 3908 that may be mixed with the feedstock 3902 prior to being fed into each reactor 3901, 3903, 3905, as well as used as a temperature quench within each of the reactors 3901, 3903, 3905 and to prevent the partial pressure of hydrogen from dropping too low at the outlets of each of the reactors 3901, 3903, 3905. Further, sensor packages 3913, 3915, 3917, 3919, 3921, 3923, 3925, 3927, 3929 may be disposed throughout the parallel line 3931 to measure, analyzeand sample the feedstocks 3902 and hydrogen gas 3908 as the feedstocks 3902 and hydrogen gas 3908 are processed through the reactors 3901, 3903, 3905 of the parallel line 3931. Sensor packages 3913, 3915, 3917, 3919, 3921, 3923, 3925, 3927, 3929 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensors and differential pressure sensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 of FIG. 6. In particular, the sensor packages 3913, 3915, 3919, 3921, 3925, 3927 may be used to determine the weighted average bed temperature (commonly referred to as “WABT”) for each of the reactors 3901, 3903, 3905 and to determine the pressure drop across the catalysts 3907, 3909, 3911.

[0237] One or more of feedstocks 3902 or feedstocks 3945 may be fed into an optional feed drum 3933 of the parallel line 3931 or into the feed drum 3904. Feedstocks 3902 and Feedstocks 3945 may be processed by different upstream processes. For example, feedstocks 3902 may be gas oil from atmospheric distillation and feedstocks 3945 may be gas oil from vacuum distillation. Consequently, feedstocks 3902 and feedstocks 3945 may have different properties and compositions of hydrocarbons and contaminants. A machine learning model may select feedstocks to be fed into a first reactor based on the condition of the reactor’s catalysts, while the feedstocks selected by the machine learning model to be fed into a second reactor may be from a different upstream process.

[0238] Alternatively, feedstocks 3902 may be fed into an optional charge heater 3935 of the parallel line 3931 from the feed drum 3904. Further, feedstocks 3902 may be fed into the reactors 3901, 3903, 3905 of the parallel line 3931 from the charge heater 3906. Hydrogen gas 3908 may be mixed into the feedstocks 3902 before being fed into the optional charge heater 3935. Further, one or more sensor packages 3937, 3939, 3941, 3943 may be disposed to measure, analyze, and sample the feedstocks 3902 and hydrogen gas 3908 as the feedstocks 3902 and hydrogen gas 3908 are prepared to be fed into the optional reactors 3901, 3903, 3905 of the parallel line 3931. Sensor packages 3937, 3939, 3941, 3943 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensorsand differential pressure sensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 3902 to be analyzed in a lab or by the sample collection and analysis assembly 614 of FIG. 6.

[0239] A machine learning model may control distribution of feedstocks 3902 into the reactors 3910, 3914, 3952, 3901, 3903, 3905 to maximize catalyst life. Further, a machine learning model may control distribution of feedstocks 3902 into the reactors 3910, 3914, 3952, 3901, 3903, 3905 to optimize catalyst usage while managing toward a shutdown of all reactors 3910, 3914, 3952, 3901, 3903, 3905 at the same time. The machine learning model may balance the distribution of feedstocks 3902 between the parallel lines of reactors to facilitate the equal usage and deactivation of the catalysts 3912, 3916, 3954, 3907. 3909, 3911.

[0240] Alternatively, a machine learning model may distribute feedstocks 3902 between the reactors 3910, 3914, 3952, 3901, 3903, 3905 to facilitate continuous processing while allowing specific reactors to be taken offline for one or more of maintenance or catalyst replacement. A machine learning model may manage the distribution of feedstocks so that a first reactor line is shut down because its catalysts have been fully or almost fully deactivated, while the catalysts of a second or parallel line of reactors are near 50 percent deactivation or 50 percent of their useful life.

[0241] A machine learning model may correlate feed composition, reactor temperatures, and sulfur content of the products of a reactor to predict and manage catalyst life in the reactor or line of reactors. Alternatively, a machine learning model may use periodic lab data gathered from periodic samples of the products of a reactor to generate a prediction of or model a predictive algorithm of sulfur content in the hydrotreated products when a sulfur analyzer is not available. In other applications, a machine learning model may correlate feed composition, reactor temperatures, and metal content of the products of a reactor to predict and manage catalyst life in the reactor or line of reactors. A machine learning model may use periodic lab data gathered from periodic samples of the products of a reactor to generate a prediction of or model a predictive algorithm of metal content in the hydrotreated products when an analyzer, such as a spectrometer, is not available.

[0242] The machine learning model may be programmed to optimize catalyst usage while managing the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window. In addition to controlling feedstocks for managing catalyst life, the machine learning model may manage reactor temperatures in support of the catalyst life cycle and managing the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window. In some applications, the first catalyst replacement window overlaps with the second catalyst replacement window. In other applications, the first catalyst replacement window does not overall with the second catalyst replacement window. In yet other applications, the first catalyst replacement window may be predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about twenty-five percent to about seventy- five percent deactivated. Alternatively, the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about fifty percent deactivated. The first catalyst replacement window may be predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about twenty-five percent to about seventy-five percent of its predicted useful life remaining or the first catalyst replacement window may be predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about fifty percent of its predicted useful life remaining. In some applications, the first catalyst replacement window may be predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has less than twenty-five percent of its predicted useful life remaining. In others, the first catalyst replacement window may be predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has more than seventy-five percent of its predicted useful life remaining.

[0243] The machine learning model may generate a predicted state of the catalysts within the first hydrotreating reactor and the second hydrotreating reactor based on the feedstock data, the operational parameters, and product data of the first hydrotreating reactor and the second hydrotreating reactor. The feedstocks may include a first feedstock from a first upstream process and a second feedstock from a second upstream process so that the machine learning model may generate a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor based on the predicted state of the catalysts within the first hydrotreating reactorand the second hydrotreating reactor. The distribution decision may be used to achieve a lower WABT in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property or achieve a lower ratio of hydrogen to feedstock in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property.

[0244] The machine learning model may generate a value for the products of the hydrotreater, a value to costs of operating the hydrotreater, a value to hydrogen gas fed to the hydrotreater, and a value to the feedstocks fed to the hydrotreater. The machine learning model may then generate a recommendation to improve the value of the hydrotreater operation based on the value to the products of the hydrotreater, less the value to the costs of operating the hydrotreater, less the value of the hydrogen gas fed to the hydrotreater, and less the value to the feedstocks fed to the hydrotreater while achieving a target product property. The target property may be a volume swell target, a maximum volume swell, aromatic saturation target, a maximum aromatic saturation, a sulfur concentration target, a cetane number target, a pour point target, a smoke point target, a flash point target, or other property target that may be affected by the hydrotreating process.

[0245] The machine learning model may generate a value for the products of a hydrocracker, a value for the utility costs of operating the hydrocracker, which may include costs of fuel gas, electricity, and steam, a value for the hydrogen gas fed to the hydrocracker, a value for the feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, a value for the costs of operating the fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker. The machine learning model may access current business data to generate these values. The current business data may include current market prices for each of the products of the fluid catalytic cracker and hydrocracker. The predicted product flows for the hydrotreater, hydrocracker, fluid catalytic cracker may be generated to use in the machine learning. The predicted properties for the products for the hydrotreater, hydrocracker, fluid catalytic cracker may be generated by the machine learning model. For combined optimization of multiple units, the machine learning model may provide the predicted feed and properties for a downstream unit. The machine learning model may generate a predicted flow and properties of the unconverted oil from the hydrocracker and then model FCC operation to determine a yield of the unconverted oil in the FCC. The machine learning model may be trained to predict the quantities and qualities of the products of the hydrotreater, the products of the FCC,including naphtha, j et fuel, kerosene, and diesel, as well as the products of the hydrocracker, including naphtha, jet fuel, kerosene, and diesel.

[0246] Values may be assigned to the products of the FCC yield prediction from the unconverted oil from the hydro cracker. The machine learning model may generate a predicted flow and properties of the treated gas oil product from the gas oil hydrotreater and then model FCC operation to determine the yield of the treated gas oil in the FCC. Values may be assigned to the products of the FCC yield prediction. In other words, the machine learning model may manage hydrocracker reactor operation to target a gas oil conversion rate that optimizes combined hydrocracker and FCC operation, or combined hydrotreater, hydrocracker, and FCC operation.

[0247] Alternatively, the machine learning model may generate a predicted flow and properties of a feedstock through the hydrocracker and gas oil hydrotreater and the flow of the unconverted oil from the hydrocracker and hydrotreated gas oil to the FCC. The machine learning model may then model each operation to determine yields of each product of the hydrotreater, hydrocracker, and FCC. Values may be assigned to each of the products of the yield prediction. Using the assigned values, each process may be optimized to produce the highest combined value of products. In other words, the machine learning model may manage hydrocracker reactor operation to target a gas oil conversion rate that optimizes combined hydrotreater, hydrocracker, and FCC operation.

[0248] These machine learning model predictions facilitate combined unit and feed stock stream optimization. A single machine learning model may be used to provide the combined unit and feed stock stream optimization. Alternatively, multiple machine learning models that are disposed to manage and optimize specific portions of each process may cooperate together to provide the combined unit and feed stock stream optimization.

[0249] A product of the hydrotreater may include hydrotreated gas oil fed to the fluid catalytic cracker and a product of the hydrocracker may include unconverted gas oil fed to the fluid catalytic cracker. Consequently, the machine learning model may be trained with historical data including feedstock data, operational data, and product data of the hydrotreater, the hydrocracker, and the fluid catalytic cracker and generate an adjustment to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve a combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker, based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value for the costs of operating the hydrotreater, the hydrocracker, and the fluid catalytic cracker, the value of the hydrogengas fed to the hydrotreater and the hydrocracker, and the value for the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker. Then, the machine learning model or a user may adjust the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker based on the generated adjustment. In some applications, the machine learning model may generate an adjustment to a cutoff of the gas oil produced by atmospheric distillation or vacuum distillation to provide a more desirable feedstock property in the feedstocks of the hydrotreater or hydrocracker.

[0250] The operational parameters of the hydrotreater, the hydrocracker, or the fluid catalytic cracker include operating parameters for the heaters, reactors, and other equipment included with these units. Operating parameters include the feed rates of feedstocks into and operating temperatures and pressures of the heaters, reactors, and other equipment of the hydrotreater, the hydrocracker, or the fluid catalytic cracker.

[0251] The machine learning model may generate an adjustment to the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker is constrained by a target product property of one or more products of the hydrotreater, the hydrocracker, or the fluid catalytic cracker. In some applications, the target property is aromatic saturation, volume swell, or maximum aromatic saturation limited by sulfur content in a product. Alternatively, the target property may be a cetane number or cold temperature property. Each adjustment may be constrained by limitations of the hydrotreater, the hydrocracker, and the fluid catalytic cracker including mechanical, metallurgical, electrical, and hydraulic limitations. For example, a feed rate may be constrained by the diameter of piping between reactors or cooling may be constrained by current weather conditions. Limitations may also include catalyst deactivation states of catalysts of the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

[0252] In some applications, a product of the fluid catalytic cracker is a feedstock to a gasoline desulfurization unit. Thus, the machine learning model may generate adjustments to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker based on product data or operational data from the gasoline desulfurization unit. Further, the machine learning model may generate values for products of the gasoline desulfurization unit and generate adjustments to an operational parameter of one or more of the hydrotreater, the hydrocracker, the fluid catalytic cracker, or the gasoline desulfurization unit based on the value of the products of the gasoline desulfurization unit. The machine learning model for one or more of the hydrotreater,hydrocracker, and fluid catalytic cracker may predict the flow and properties of the naphtha to be processed in the gasoline desulfurization unit. This prediction can be used in a machine learning model of the gasoline desulfurization unit to predict the flow and properties of the gasoline desulfurization unit product.

[0253] An objective function provided by a user may be used by the machine learning model in optimizing a process or series of related processes. The objective function indicates one or more goals a machine learning model should try to obtain through adjustments to the operating parameters of a process of interrelated processes.

[0254] In some embodiments, a controller includes a local enhancement module. The local enhancement module includes an algorithm configured to facilitate optimization of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to achieve the target parameters (e.g., the objective function) based on the outputs from the first machine learning models (e.g., of predictive controls modules of the controller), the target parameters, unit constraints, and inputs. The algorithm may be configured to facilitate optimization of the hydrotreater, the hydrocracker, or the fluid catalytic cracker based on a machine learning model. The local enhancement module may control the refinery operation control devices based on the outputs and constraints of the hydrotreater, the hydrocracker, or the fluid catalytic cracker. In some embodiments, the local enhancement module includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g. maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules), the target parameters, unit constraints, and inputs. The local enhancement module may control the refinery operation control devices based on the outputs and constraints of the one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker.

[0255] In an embodiment, the objective function directs the machine learning model to increase the value of a combined hydrocracker and fluid catalytic cracker operation where unconverted gas oil is fed from the hydrocracker to the fluid catalytic cracker. In some applications, an upstream hydrotreater process providing products to the fluid catalytic cracker and or the hydrotreater may be included in the optimization process, as well as a downstream gasoline desulfurization unit treating the products of the fluid catalytic cracker. The machine learning model may receive business data that may be used by the machine learning model to generate a value of a product of the hydrocracker, a value of a product of the fluid catalytic cracker, a value of a product of a hydrotreater, a value of the costs of operating the hydrotreater, the hydrocracker, and the fluid catalytic cracker, a value of thehydrogen gas fed to the hydrotreater and the hydrocracker, and a value of the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker. To optimize each process, the interdependence of the hydrotreater, the hydrocracker, and the fluid catalytic cracker are accounted for through the values. In particular, the products for one unit may provide the feedstock for another. For example, the unconverted gas oil from the hydrocracker becomes the feedstock for the fluid catalytic cracker. In another example, the hydrotreated gas oil from the hydrotreater becomes the feedstock for the fluid catalytic cracker. In another example, the naphtha produced by the fluid catalytic cracker becomes the feedstock for the gasoline desulfurization unit. The feedstock value, product value, hydrogen gas used, and operating cost for the entire system of units may be used to generate changes to the operational parameters of each of the interdependent processes. An advantage of using a machine learning model in this application is the machine learning model may optimize the amount of unconverted gas oil from the hydrocracker to the fluid catalytic cracker to increase the value of both operations. The unconverted gas oil is a desirable feedstock for the fluid catalytic cracker because of its low concentration of contaminants due to processing in the hydrocracker and may be used as a clean feed to help counter some deactivation of the catalysts of the fluid catalytic cracker.

[0256] For example, the system of units to be optimized may begin with an objective function to be maximized equal to a total value of products less a total cost of feedstocks less operating costs subject to operational limits and engineer specified-constraints, which may be written as:Objective = Sum of (Product*Product Value)- Sum of (Feed*Feed Value)- Operating costThe machine learning model may generate outputs for operational settings (parameters) chosen to be manipulated variables of the controller in order to optimize the system of units according to the objective function. The optimization of the objective function may be subject to operational constraints, including known unique constraints of specific equipment.

[0257] Alternatively, an objective function used by the machine learning model may include a value of a product of the hydrocracker, a value of a product of the fluid catalytic cracker, a value of the hydrogen gas fed to the hydrocracker, and a value of the feedstocks fed to the hydrocracker and the fluid catalytic cracker. In another application, an objectivefunction used by the machine learning model may include a value of a product of the hydrotreater, a value of a product of the hydrotreater and the fluid catalytic cracker, a value of the hydrogen gas fed to the hydrotreater, and a value of the feedstocks fed to the hydrotreater and the fluid catalytic cracker. The objective functions may include a value for the operating costs of each of the hydrotreater, fluid catalytic cracker, hydrocracker, or gasoline desulfurization unit as applicable to each objective function.

[0258] The values may be positive for finished products that may be sold or blended into final products and may be based off of market value or the values provided for under existing contracts. The values for costs of operation, feedstocks, and hydrogen may be ascribed a negative value in the objective function. In some applications, the cost of feedstocks may include the costs to obtain and process the feedstocks into a condition that may be fed into one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker. In some applications, a value assigned to a feedstock that is a byproduct of an upstream process, such as unconverted gas oil from the hydrocracker, may have a reduced value as a feedstock to emphasize the value in further processing of the unconverted gas oil by the fluid catalytic cracker and the resulting valuable products produced from the unconverted gas oil. In other words, the value assigned to a feedstock may be reduced to below the costs of producing the feedstock to account for the higher value of the products processed by an upstream process. For example, a value assigned to the unconverted gas oil processed by the hydrocracker may have reduced value below its actual cost of production to indicate its reduced value offset by the higher values of the naphtha, jet fuel, kerosene, and diesel produced by the hydrocracker.

[0259] The objective function may be written to target a desired property of a product of the hydrotreater, fluid catalytic cracker, hydrocracker, or gasoline desulfurization unit. By assigning values to products having certain product properties, a machine learning model may generate predictions concerning the products of the fluid catalytic cracker, including naphtha, jet fuel, and diesel based on current operating parameters and feedstock data. The machine learning model may generate outputs for operational settings chosen to be manipulated variables of the controller in order to generate recommended operational parameters to produce those products with those targeted properties. In other words, generate outputs for operational settings chosen the be manipulated variables of the controller in order to produce those products with those targeted properties, including but not limited to aromatic saturation, product volume swell, octane number, octane-barrels, cetane number, cetane barrels, flash point, pour point, smoke point, cloud point, sulfurcontent, or nitrogen content to improve the overall value of the of the hydrotreater, fluid catalytic cracker, hydrocracker, or gasoline desulfurization processes. Value may be the market value of each product less the cost to produce the product, or the potential profit margin of each product. Known unique constraints of specific equipment may be used when generating recommended operational parameters to prevent generating adjustments that would not be possible to implement.

[0260] While this schematic illustrates a basic flow of feedstocks through a hydrotreater, a wide variety of configurations exist. Alternatively, the hydrotreater may include multiple reactors, separators, and stabilizers that may be organized in series or in parallel. Further, a wide variety of sensors may be used to monitor and record data about the processes within the naphtha hydrotreater 3900.

[0261] A machine learning model may optimize and protect the hydrotreater 3900 by adjusting the flow rates of feedstocks 3902 and hydrogen gas 3908 from the feed drum 3904. The machine learning model may also measure the composition of the feedstocks 3902 and hydrogen gas 3908 and adjust the temperature, pressure, and flow rate of the mixed feedstocks 3902 and hydrogen gas 3908 within the charge heater 3906 before being fed into the reactors 3910, 3914, 3952. The machine learning model may control the flow rate of mixed feedstocks 3902 and hydrogen gas 3908 into the reactors 3910, 3914, 3952. The machine learning model may control the operating temperatures and pressures of the reactors 3910, 3914, 3952, as well as the quenching flows of hydrogen gas within the reactors 3910, 3914, 3952. The machine learning model may also control the operating parameters of the separator 3966 and the stabilizer 3982. Adjusting these controls and obtaining the sensor data allows the machine learning model to develop improved algorithms. The machine learning model may adjust the naphtha hydrotreater 3900 to improve operations to support downstream process and provide feedstocks that may improve the efficiency of downstream processes.

[0262] FIG. 17 is a schematic diagram of a hydrocracker control system 2300 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. As noted, specific sections or portions of a refinery may include a sub-controller to enhance that particular operation. As illustrated in FIG. 5, a hydrocracking operation may be enhanced via a hydrocracker controller 2302 and corresponding machine learning models (for example, hydrocracker specific machine learning models utilized in the local enhancement module 2304 and / or the predictive controls module 2306).

[0263] Feedstocks 2301 may include gas oils from atmospheric distillation and vacuum distillation. The feedstocks 2301 may be mixed with hydrogen gas 2303 and recycled hydrogen gas 2305 and then passed through a heat exchanger 2310 and sent to a feed heater 2312, where the feedstocks 2301 and hydrogen gas 2303, 2305 are heated in preparation to be fed into reactor 2314. The reactors 2314 may be one or more hydrotreater reactors for removing contaminants, including sulfur, nitrogen, and metals, from the feedstocks 2301 similarly to the reactors 3910, 3914, 3952, 3901, 3903, 3905 of the hydrotreater 3900 of FIG. 16. Once processed by the reactors 2314, the feedstocks 2301 are fed into the reactors 2316.

[0264] The reactors 2316 include one or more hydrocracker reactors that may be arranged in parallel or series. The reactors 2316 operate at high pressures and elevated temperatures. The reactors 2316 use high pressure hydrogen gas 2303, 2305 and high temperatures to saturate diolefins, olefins, and aromatics and to crack the larger molecules of the gas oil feedstocks in the presence of catalysts, such as platinum or palladium to support hydrogenation and amorphous silica-alumina or zeolite to facilitate cracking, into lighter molecules including propane and butane 2328, naphtha 2330, diesel 2332, kerosene and jet fuel 2334, and unconverted gas oil 2336 (often referred to as “UCO”).

[0265] The operational parameters of reactors 2316 include inlet, outlet, and weighted average bed temperatures, hydrogen-to-oil ratio, and hydrogen gas recycle rate. Hydrogen gas prevents coking and helps reduce contaminants including remaining sulfur and nitrogen in the feedstocks 2301. The operating parameters correlate to a conversion percent of the feedstocks 2301 into the resulting products including propane and butane 2328, naphtha 2330, diesel 2332, kerosene and jet fuel 2334 versus unconverted gas oil 2336. The conversion percent is determined by the total volume of products minus the volume of unconverted gas oil 2336, divided by the total volume of products.

[0266] Once passed through the reactors 2316, the feedstocks 2301 may be passed through the heat exchanged 2310 to heat the incoming unprocessed feedstocks 2301 and then fed into a separator 2318 that captures and recycles hydrogen gas 2305. From the separator 2318, the hydrocracked feedstocks 2301 may be passed through another heat exchanger 2320 and passed into a stabilizer 2322 that may be used to remove light ends 2324 such as hydrogen sulfide, methane, and remaining hydrogen gas. The hydrocracked feedstocks 2301 may then be passed back through the heat exchanger 2320 and then to a fractionation column 2326 where the hydrocracked feedstocks 2301 are separated into productsincluding propane and butane 2328, naphtha 2330, diesel 2332, kerosene and jet fuel 2334, and unconverted gas oil 2336.

[0267] The hydrocracker controller 2302 may connect to sources of and control various parts of the equipment, such as reactors 2314, 2316, heat exchangers 2310, 2320, a separator 2318, and / or a stabilizer 2322. Further, the hydrocracker controller 2302 may obtain data from sample collection and analysis assembly 2308, which may be connected to sensor packages 2340, 2342, 2344, 2346, 2348, 2350, 2352, 2354, 2356, 2358. 2360, 2362 distributed throughout the hydrocracker control system 2300 and disposed to sample, measure, or analyze one or more properties of the feedstocks 2301, hydrogen gas 2303, 2305 as they move through the hydrotreater, the light ends 2324, and the products of the hydrotreater including propane and butane 2328, naphtha 2330, diesel 2332, kerosene and jet fuel 2334, and unconverted gas oil 2336. Sensor packages 2340, 2342, 2344, 2346, 2348, 2350, 2352, 2354, 2356, 2358. 2360, 2362 may include temperature sensors; such as resistance temperature detectors, thermocouples, and infrared sensors; pressure sensors, such as hydrostatic pressure sensors and differential pressure sensors; flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow; density and composition analyzers, such as sulfur analyzers, refractometers, spectrometers, including near infrared spectrometry, and gas chromatographs; and sampling units for analyzing the composition of a sample through testing, distillation, or obtaining physical samples from the feedstocks 2301 to be analyzed in a lab or by the sample collection and analysis assembly 2308.

[0268] Prior to generating adjusted parameters for operation of the hydrocracker operation, the hydrocracker controller 2302, which may include a machine learning model or be in communication with a machine learning model, may initiate collection of samples of one or more of the feedstocks 2301 and hydrogen gas 2303 utilized in the hydrocracking operations, via the sample collection and analysis assembly 2308. Once the sample collection and analysis assembly 2308 obtains one or more samples, the sample collection and analysis assembly 2308 may analyze those samples to produce properties and / or a spectra indicative of various properties. Once the properties and / or spectra are generated, the sample collection and analysis assembly 2308 may transmit the properties and / or spectra to the hydrocracker controller 2302. Upon reception of the properties and / or spectra, the hydrocracker controller 2302 may apply the properties and / or spectra and other data, along with target product composition and / or properties, to a plurality of machine learning models within the predictive controls module 2306 (or, in other embodiments, aplurality of predictive controls modules may be included in the hydrocracker controller 2302 and each may include a machine learning model). The output of each of the machine learning models in the predictive controls module 2306 may then be utilized by the local enhancement module 2304 to determine (for example, as a vector) new parameters and / or feed blend or composition to supply to the equipment utilized in the hydrocracking operation. In other embodiments, the output may indicate that a new or fresh catalyst should be utilized. The hydrocracker controller 2302 may then adjust the parameters of and / or feeds and / or materials for the equipment associated with the hydrocracker operation 3900.

[0269] In an embodiment, a machine learning model may, when trained, determine adjustments to a hydrocracker controller to give FCC feed composition that provides optimal FCC yield against FCC constraints. In yet another embodiment, another machine learning model may, when trained, determine adjustments to hydrotreater severity to maximize aromatic saturation to hydrotreater constraints. The hydrocracker may be managed by a machine learning model to maximize the conversion percent of the feedstocks 2301.

[0270] In another embodiment, the hydrocracker controller 2302 may first cause sampling of various fluids used and / or produced in the hydrocracker operation. For example, as hydrocracking occurs (for example, as a continuous and / or ongoing operation) various fluids and / or materials may be utilized and / or produced therein. Prior to application of data to any of the machine learning models described herein, the hydrocracker controller 2302 may initiate capture of one or more of those fluids via the sample collection and analysis assembly 2308. Once analyzed, the hydrocracker controller 2302 may predict properties of the corresponding fluids that may achieve an accurate output of a target product, as well as various parameters as described herein. In other words, the hydrocracker controller 2302 may determine composition of a fluid to be utilized. The fluids used may then be adjusted, blended, and / or supplemented to meet the composition determined by one or more of the machine learning models of the hydrocracker controller 2302. Stated another way, the hydrocracker controller 2302 may control the properties, composition, and / or feed ratios associated with a feedstock or hydrocarbon feedstock and / or an intermediate fluid or intermediate product.

[0271] In another embodiment, the machine learning model may access business information and assign a value to the products of the hydrocracker and a value to the costs of operating the hydrocracker, the feedstocks 2301, and the hydrogen gas 2303 to generate an overall value of the hydrocracker. Business information may include the current marketprice for each of the products, employee salaries, benefits, and wages, current costs and usage of electricity, the current market or supplier price of hydrogen gas, the maintenance costs of the equipment, the depreciated cost of the equipment, and any other budgeted line items assigned as a cost of operating the hydrocracker. Business information and costs of operation may also include the life cycle of each reactor’s 2314, 2316 catalysts and the costs of replacement and regeneration predicted by the machine learning model based on the historical, current, and forecast costs. Costs of operation may include utilities and the costs of producing steam and cooling that may be used in generating a potential profit of the hydrocracker. The value of each product may be determined as a potential profit from each product. The machine learning model may then generate one or more adjustments to the operating parameters hydrocracker, the feed rates of feedstocks 2301, the feed rate of hydrogen gas 2303, the operating parameters of the upstream processes, and the operating parameters of the downstream processes. For example, the machine learning model may adjust the cut off of the gas oil produced by atmospheric distillation or vacuum distillation to provide a more desirable feedstock property in the feedstocks 2301.

[0272] In addition to the values and costs assigned to the feedstocks, products and operations of the hydrocracker, the machine learning model may assign values to the operating costs of the FCC shown in FIG. 5 downstream of the hydrocracker and the products produced from the unconverted gas oil by the FCC to generate a more complete overall objective function of the combined production of the hydrocracker and FCC. This overall objective function of the combined production of the hydrocracker and FCC may be constrained by the operational constraints of each of the processes and used by the machine learning model to improve the efficient operation of both the FCC and the hydrocracker.

[0273] This objective function may further include the values and costs assigned to other feedstocks of the FCC including the hydrotreaters used to process other FCC feedstocks. Consequently, the objection function may be used to improve the efficient operation of the FCC, the hydrocracker, and the one or more hydrotreaters upstream of the FCC.

[0274] Alternatively, the machine learning model may generate a predicted flow and properties of a feedstock through the hydrocracker and gas oil hydrotreater and the flow of the unconverted oil from the hydrocracker and hydrotreated gas oil to the FCC. The machine learning model may then model each operation to determine yields of each product of the hydrotreater, hydrocracker, and FCC. Values may be assigned to each of the products of the yield prediction. Using the assigned values, each process may be optimized toproduce the highest combined value or potential profit of products. In other words, the machine learning model may manage hydrocracker reactor operation to target a gas oil conversion rate that optimizes combined hydrotreater, hydrocracker, and FCC operation to produce a combination of products with the most potential profit.

[0275] In an embodiment, each controller may utilize various data points and properties to predict parameters and fluids to reach a target product. Those controllers may each connect to a supervisory controller or refinery controller. In embodiments, the supervisory controller or refinery controller may utilize the output of each machine learning model of each of the controllers. In yet another embodiment, each of the controllers may utilize some output from the machine learning model of the supervisory controller or refinery controller. Further, each of the controllers and / or the supervisory controller may adjust a refining operation control device, and thus adjust a process, in real-time, near real-time, and / or continuously.

[0276] FIG. 18 is a flow diagram of a method, according to an embodiment of the disclosure. The method includes receiving, by a machine learning model, feedstock data from one or more refinery equipment providing feedstock to a hydrotreater at 4010. The feedstock data is indicative of one or more operating parameters of the one or more refinery equipment. The machine learning model is trained on historical feedstock data and historical data of operation of the hydrotreater. The machine learning model has a target product specification for a first product from the hydrotreater. The method further includes predicting, by the machine learning model, a property of the feedstock based on the feedstock data at 4020, and generating, by the machine learning model, an output indicative of an adjustment to a parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification based on the feedstock data, the historical feedstock data, and historical data of operation of the hydrotreater at 4030.

[0277] FIG. 19 is a flow diagram of another method, according to an embodiment of the disclosure. In some aspects, the techniques described herein relate to a method including receiving, by a machine learning model, a first feedstock data from one or more refinery equipment providing a first feedstock to be hydrotreated at 4110. The first feedstock data is indicative of one or more properties or composition of the first feedstock. The machine learning model is trained on historical first feedstock data, historical data of operational parameters of a first hydrotreater, historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater andthe second hydrotreater. The machine learning model has a target product specification for the first feedstock. The method further including receiving, by the machine learning model, current data of operational parameters of the first hydrotreater and the second hydrotreater at 4120 and generating, by the machine learning model, a recommendation of one of the first hydrotreater or the second hydrotreater to send the first feedstock to be hydrotreated at 4130. The method further comprising generating, by the machine learning model, an output indicative of an adjustment to a parameter of operation of the recommended one of the first hydrotreater or the second hydrotreater to hydrotreat the first feedstock to accurately produce the product to the target product specification for the first feedstock based on the feedstock data, the historical feedstock data, historical data of operational parameters of a first hydrotreater, historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater and the second hydrotreater at 4140.

[0278] FIG. 20 is a flow diagram of another method, according to an embodiment of the disclosure. In some aspects, the techniques described herein relate to a method including receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater at 4210 and receiving, by the machine learning model, current operating parameters of the hydrotreater at 4220. The machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater. The method further includes generating, by the machine learning model, a prediction of a state of catalysts within a hydrotreater reactor based on the current operating parameters of the hydrotreater at 4220 and generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater based on the predicted state of the catalysts within the hydrotreater reactor and the feedstock data at 4230.

[0279] FIG. 21 is a flow diagram of another method, according to an embodiment of the disclosure.In some aspects, the techniques described herein relate to a method including receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater at 4310 and receiving, by the machine learning model, current operating parameters of the hydrotreater at 4320. The machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater. The machine learning model is trained to meet a sulfur target, while maximizing aromatic saturation constrainedby available hydrogen gas feed rates. The method further including generating, by the machine learning model, a target aromatic saturation based on the sulfur target, the feedstock data, and available hydrogen gas feed rates at 4330 and generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater based on the target aromatic saturation of the feedstocks within the hydrotreater, the sulfur target, the feedstock data, and available hydrogen gas feed rates at 4340.

[0280] FIG. 22 is a flow diagram of another method, according to an embodiment of the disclosure. In some aspects, the techniques described herein relate to a method including generating, by a machine learning model, a value for products of a hydrotreater, a value for hydrogen gas fed to the hydrotreater, and a value for feedstocks fed to the hydrotreater, a value for the products of a hydrocracker, a value for the hydrogen gas fed to the hydrocracker, a value for the feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker at 4410. A product of the hydrotreater includes hydrotreated gas oil fed to the fluid catalytic cracker. A product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker. The machine learning model was trained with historical data including feedstock data, operational data, and product data of the hydrotreater, the hydrocracker, and the fluid catalytic cracker. The method further including generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve a combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrotreater and the hydrocracker, and the value for the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker at 4420.

[0281] FIG. 23 is a flow diagram of another method, according to an embodiment of the disclosure.In some aspects, the techniques described herein relate to a method including generating, by a machine learning model, a value for products of a hydrocracker, a value for hydrogen gas fed to the hydrocracker, a value for feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker at 4520. A product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker. The machine learning model was trained with historical data includingfeedstock data, operational data, and product data of the hydrocracker and the fluid catalytic cracker. The method further including generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to improve a combined value of the hydrocracker and the fluid catalytic cracker, based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrocracker, and the value for the feedstocks fed to the hydrocracker and the fluid catalytic cracker at 4520.

[0282] FIG. 24 is a flow diagram of another method, according to an embodiment of the disclosure. A method includes supplying feedstock to a hydrotreater associated with the petroleum refining operation, the feedstock having one or more feedstock properties at 4610 and analyzing a feedstock sample via a first analyzer to provide feedstock sample properties at 4620. The method further includes predicting one or more feedstock sample properties associated with the feedstock sample based on (A) the feedstock sample properties and (B) a first output from application of the feedstock sample properties to a first trained machine learning model at 4630 and operating the hydrotreater to produce one or more unit materials, the one or more unit materials having one or more unit materials properties, and the one or more unit materials including one or more of gas, gasoline, diesel, or isobutanes at 4640. The method further includes analyzing the unit material sample via a second analyzer to provide unit material sample properties at 4650 and predicting one or more unit material sample properties associated with the unit material sample based on (C) the unit material sample properties and (D) a second output from application of the unit material sample properties to a second trained machine learning model at 4660. The method including controlling, during the hydrotreatment operation, based on the feedstock sample properties and the one or more unit material sample properties, one or more of: (a) one or more feedstock properties associated with the feedstock supplied to the hydrotreater; (b) one or more unit product materials properties associated with the unit product materials; (c) operation of the hydrotreater; or (d) operation of one or more upstream equipment or downstream equipment at 4670. The method further includes controlling, during the hydrotreatment operation, causes the hydrotreatment operation to produce one or more of: (i) one or more intermediate materials each having one or more properties within a range of one or more target properties of the one or more intermediate materials, (ii) one or more unit product materials each having one or more properties within a range of one or more target properties of the one or more unit product materials, or (iii) one or more downstream materials each having one or more properties within a range of one or more target propertiesof the one or more downstream materials at 4680. The method thereby causes the hydrotreatment operation to achieve material outputs that more accurately and responsively converge on one or more of the target properties at 4690.

[0283] FIG. 25 is a flow diagram of another method, according to an embodiment of the disclosure. The method includes receiving business data to a machine learning model at 4710. The machine learning model is trained on data indicative of one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, products of the fluid catalytic cracker. The business data is related to the one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, products of the fluid catalytic cracker. The method further including generating, by the machine learning model, values for the one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, and products of the fluid catalytic cracker at 4720. The method further includes generating, by the machine learning model, one or more adjustments to parameters of the hydrocracker and the fluid catalytic cracker at 4730. The hydrocracker provides unconverted gas oil to the fluid catalytic cracker based on an objective function to improve the value of the hydrocracker and fluid catalytic cracker operation. The method optionally includes implementing the one or more adjustments to parameters of the hydrocracker and fluid catalytic cracker at 4740.

[0284] FIG. 26 is a flow diagram of a method, according to an embodiment of the disclosure. The method includes receiving a fluid catalytic cracker feedstock data indicative of a feedstock being fed into a fluid catalytic cracker at a machine learning model at 4810 and receiving operational parameters of the fluid catalytic cracker at the machine learning model at 4820. The method further includes receiving a hydrocracker feedstock data indicative of a feedstock being fed into a hydrocracker at the machine learning model at 4830 and receiving operational parameters of the hydrocracker at the machine learning model at 4840. The machine learning model generates an adjustment to an operational parameter of the hydrocracker to produce an amount of unconverted gas oil fed into the fluid catalytic cracker at 4850 and generates a prediction of the hydrocracker product yield from hydrocracker based on the adjustment to the operational parameters of the hydrocracker, wherein the hydrocracker product yield includes the amount of unconvertedoil from hydrocracker at 4860. The machine learning model generates a prediction of an effect on the fluid catalytic cracker product yield based on the generated adjustment in the amount of unconverted oil from the hydrocracker fed into the fluid catalytic cracker at 4870. The method may further include an optional step of determining whether the adjustment to the operational parameter of the hydrocracker improves a combined product yield of the hydrocracker and the fluid catalytic cracker at 4880.

[0285] In the drawings and specification, several embodiments of systems and methods to provide in-line mixing of hydrocarbon liquids have been disclosed, and although specific terms are employed, the terms are used in a descriptive sense only and not for purposes of limitation. Embodiments of systems and methods have been described in considerable detail with specific reference to the illustrated embodiments. However, it will be apparent that various modifications and changes may be made within the spirit and scope of the embodiments of systems and methods as described in the foregoing specification, and such modifications and changes are to be considered equivalents and part of this disclosure.

Claims

1. CLAIMSWHAT IS CLAIMED IS:

1. A method compri sing : receiving a fluid catalytic cracker feedstock data indicative of a feedstock being fed into a fluid catalytic cracker at a machine learning model; receiving operational parameters of the fluid catalytic cracker at the machine learning model; receiving a hydrocracker feedstock data indicative of a feedstock being fed into a hydrocracker at the machine learning model; receiving operational parameters of the hydrocracker at the machine learning model; generating, by the machine learning model, an adjustment to an operational parameter of the hydrocracker to produce an amount of unconverted gas oil fed into the fluid catalytic cracker; generating, by the machine learning model, a prediction of hydrocracker product yield from hydrocracker based on the adjustment to the operational parameters of the hydrocracker, wherein the hydrocracker product yield includes the amount of unconverted oil from hydrocracker; and generating, by the machine learning model, a prediction of an effect on the fluid catalytic cracker product yield based on the generated adjustment in the amount of unconverted oil from the hydrocracker fed into the fluid catalytic cracker.

2. The method of claim 1, further comprising determining whether the adjustment to the operational parameter of the hydrocracker improves a combined product yield of the hydrocracker and the fluid catalytic cracker.

3. The method of claims 1 or 2, further comprising determining whether the adjustment to the operational parameter of the hydrocracker improves a combined product potential profit of products from the hydrocracker and the fluid catalytic cracker.

4. The method of any of claims 1 to 3, wherein the machine learning has been trained with historical data of fluid catalytic cracker feedstock data, operational parameters of the fluid catalytic cracker, fluid catalytic cracker product yield, fluid catalytic cracker productproperties, hydrocracker feedstock data, operational parameters of the hydrocracker, hydrocracker product yield, and hydrocracker product properties.

5. The method of claim 4, wherein the machine learning has been trained with process flow data indicative of the processing of a hydrocracker feedstock upstream of the hydrocracker and a flow of hydrocracker products to processes downstream of the hydrocracker, including the fluid catalytic cracker.

6. The method of claim 5, wherein the machine learning has been trained with process flow data indicative of the processing of a fluid catalytic cracker feedstock upstream of the fluid catalytic cracker and the flow of fluid catalytic cracker products to processes downstream of the fluid catalytic cracker, including a gasoline blending pool.

7. The method of any of claims 1 to 5, further comprising generating, by the machine learning model, an adjustment to an operational parameter of the fluid catalytic cracker to achieve a target property of a product of the fluid catalytic cracker based on the predicted amount of unconverted oil from hydrocracker fed into the fluid catalytic cracker.

8. The method of claim 7, further comprising determining whether the adjustment to the operational parameter of the hydrocracker improves a combined product potential profit of products from the hydrocracker and the fluid catalytic cracker.

9. The method of any of claims 1 to 8, further comprising implementing the adjustment to the operational parameter of the fluid catalytic cracker.

10. The method of claim 9, wherein implementing the adjustment to the operational parameter of the fluid catalytic cracker, includes adjusting the operational parameter of the fluid catalytic cracker.

11. The method of any of claims 7 to 10, wherein the target property of a product of the fluid catalytic cracker includes a target yield of FCC gasoline.

12. The method of any of claims 7 to 11, wherein the target property of a product of the fluid catalytic cracker includes a target yield of diesel.

13. The method of any of claims 7 to 12, wherein the target property of a product of the fluid catalytic cracker includes a target yield of kerosene.

14. The method of any of claims 7 to 13, wherein the target property of a product of the fluid catalytic cracker includes a target yield of jet fuel.

15. The method of any of claims 7 to 14, wherein the target property of a product of the fluid catalytic cracker includes a target yield of jet fuel.

16. The method of any of claims 7 to 15, wherein the target property of a product of the fluid catalytic cracker includes a target octane-barrels.

17. The method of any of claims 7 to 16, wherein the target property of a product of the fluid catalytic cracker includes a target octane number.

18. The method of any of claims 7 to 17, wherein the target property of a product of the fluid catalytic cracker includes a target Reid vapor pressure.

19. The method of any of claims 7 to 18, wherein the target property of a product of the fluid catalytic cracker includes a target cetane number.

20. The method of any of claims 7 to 19, wherein the target property of a product of the fluid catalytic cracker includes a target cloud point.

21. The method of any of claims 7 to 20, wherein the target property of a product of the fluid catalytic cracker includes a target pour point.

22. The method of any of claims 7 to 21, wherein the target property of a product of the fluid catalytic cracker includes a target flash point.

23. The method of any of claims 7 to 22, wherein the target property of a product of the fluid catalytic cracker includes a target freezing point.

24. The method of any of claims 7 to 23, wherein the target property of a product of the fluid catalytic cracker includes a target smoke point.

25. The method of any of claims 1 to 24, wherein generating, by the machine learning model, the adjustment to the operational parameter of the hydrocracker is constrained by the hydrocracker achieving a target property of a product of the hydrocracker.

26. The method of claim 25, wherein the target property of a product of the hydrocracker includes a target octane-barrels.

27. The method of any of claims 25 to 26, wherein the target property of a product of the hydrocracker includes a target octane number.

28. The method of any of claims 25 to 27, wherein the target property of a product of the hydrocracker includes a target Reid vapor pressure.

29. The method of any of claims 25 to 28, wherein the target property of a product of the hydrocracker includes a target sulfur content.

30. The method of any of claims 25 to 29 , wherein the target property of a product of the hydrocracker includes a target nitrogen content.

31. The method of any of claims 25 to 30, wherein the target property of a product of the hydrocracker includes a target cetane number.

32. The method of any of claims 25 to 31, wherein the target property of a product of the hydrocracker includes a target cloud point.

33. The method of any of claims 25 to 32, wherein the target property of a product of the hydrocracker includes a target pour point.

34. The method of any of claims 25 to 33, wherein the target property of a product of the hydrocracker includes a target flash point.

35. The method of any of claims 25 to 34, wherein the target property of a product of the hydrocracker includes a target freezing point.

36. The method of any of claims 25 to 34, wherein the target property of a product of the hydrocracker includes a target smoke point.

37. The method of any of claims 1 to 25, further comprising implementing the adjustment to the operational parameter of the hydrocracker.

38. The method of claim 37, wherein implementing the adjustment to the operational parameter of the hydrocracker, includes adjusting the operational parameter of the hydrocracker.

39. The method of any of claims 37 to 38, further comprising: receiving hydrocracker product data indicative of hydrocracker product yield at the machine learning model; generating a comparison of the hydrocracker product data with the predicted hydrocracker product yield; and training the machine learning model with the comparison of the hydrocracker product data with the predicted hydrocracker product yield.

40. The method of any of claims 1 to 39, further comprising: receiving a hydrotreater feedstock data indicative of a feedstock being fed into a hydrotreater at the machine learning model; receiving operational parameters of the hydrotreater at the machine learning model; and generating, by the machine learning model, an adjustment to an operational parameter of the hydrotreater to achieve a target property of a product of the hydrotreater based on the predicted amount of unconverted oil from hydrocracker fed into the fluid catalytic cracker, wherein a product of the hydrotreater is fed into the fluid catalytic cracker.

41. The method of claim 40, wherein the machine learning has been trained with historical data of hydrotreater feedstock data, operational parameters of the hydrotreater, hydrotreater product yield, and hydrotreater product properties.

42. The method of claim 41, wherein the machine learning has been trained with process flow data indicative of the processing of a hydrotreater feedstock upstream of the hydrotreater and a flow of hydrotreater products to processes downstream of the hydrotreater, including the fluid catalytic cracker.

43. The method of any of claims 40 to 42, further comprising generating, by the machine learning model, a prediction of an effect on the fluid catalytic cracker product yield based on the generated adjustment to the operational parameter of the hydrotreater.

44. The method of claim 43, further comprising determining whether the adjustment to an operational parameter of the hydrotreater improves a combined product yield of the hydrocracker, the fluid catalytic cracker, and the hydrotreater.

45. The method of claim 43, further comprising determining whether the adjustment to the operational parameter of the hydrotreater improves a combined product potential profit of the hydrotreater, hydrocracker, and the fluid catalytic cracker.

46. The method of any of claims 1 to 43, wherein the machine learning model is trained to maximize a combined profit of the products of the hydrocracker and the fluid catalytic cracker.

47. The method of any of claims 40 to 46, further comprising implementing the adjustment to the operational parameter of the hydrotreater.

48. The method of claim 47, wherein implementing the adjustment to the operational parameter of the hydrotreater, includes adjusting the operational parameter of the hydrotreater.

49. The method of any of claims 40 to 48, wherein generating, by the machine learning model, the adjustment to the operational parameter of the hydrotreater is constrained by the hydrotreater achieving a target property of a product of the hydrotreater.

50. The method of any of claims 40 to 49, wherein the target property of a product of the hydrotreater includes a target sulfur content.

51. The method of any of claims 40 to 50, wherein the target property of a product of the hydrotreater includes a target volumetric swell.

52. The method of any of claims 40 to 51, wherein the target property of a product of the hydrotreater includes a target nitrogen content.

53. The method of any of claims 40 to 52, wherein the target property of a product of the hydrotreater includes a target nitrogen content.

900. The method of any of claims 1 to 53, further comprising generating, by the machine learning model, an arrangement of feedstocks to be processed in by the fluid catalytic cracker.

901. The method of claim 900, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker is made to increase catalyst life of the catalysts of the fluid catalytic cracker.

902. The method of any of claims 900 to 901, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker includes a period of time for processing of a first feedstock followed by a longer period of time for processing a second feedstock.

903. The method of any of claims 900 to 902, wherein the second feedstock includes one or more feedstocks that cause less deactivation of the catalysts of the fluid catalytic cracker than the first feedstock.

54. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 1 to 903.

55. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 1 to 903.

56. A system for enhancing fluid production for a hydrotreatment, the system comprising: a hydrotreater to receive a feedstock and refine the feedstock into a refined feedstock; a plurality of sensors to measure a parameter associated with the hydrotreater and each positioned at one of (a) proximate the hydrotreater or (b) within the hydrotreater; a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the hydrotreater and to control aspects of fluid flowing to or from the hydrotreater; one or more sample collection assemblies to collect samples of the fluid associated with the hydrotreater; one or more sample analysis assemblies to analyze each collected sample to provide properties of collected samples; and a hydrotreater controller in signal communication with the hydrotreater, the plurality of sensors, the plurality of refinery operation control devices, and the one or more sample analysis assemblies, and storing a trained machine learning model, the hydrotreater controller configured to: determine an output including predicted properties of feedstock and parameter settings of the plurality of refinery operation control devices and the hydrotreater based on application of one or more of (i) data measured by the plurality of sensors, (ii) data corresponding to analysis from the one or more sample analysis assemblies, or (iii) a target product and corresponding target properties to the trained machine learning model, and adjust one or more of an amount of feedstock or type of feedstock and parameters associated with the plurality of refinery operation control devices and hydrotreater based on the output to enhance production of the refined feedstock.

57. A method for enhancing control of a hydrotreatment operation associated with a petroleum refining operation, the method comprising: supplying feedstock to a hydrotreater associated with the petroleum refining operation, the feedstock having one or more feedstock properties; analyzing a feedstock sample via a first analyzer to provide feedstock sample properties;predicting one or more feedstock sample properties associated with the feedstock sample based on (A) the feedstock sample properties and (B) a first output from application of the feedstock sample properties to a first trained machine learning model; operating the hydrotreater to produce one or more unit materials, the one or more unit materials having one or more unit materials properties, and the one or more unit materials comprising one or more of gas, gasoline, diesel, or isobutanes; analyzing the unit material sample via a second analyzer to provide unit material sample properties; predicting one or more unit material sample properties associated with the unit material sample based on (C) the unit material sample properties and (D) a second output from application of the unit material sample properties to a second trained machine learning model; and controlling, during the hydrotreatment operation, based on the feedstock sample properties and the one or more unit material sample properties, one or more of:(a) one or more feedstock properties associated with the feedstock supplied to the hydrotreater;(b) one or more unit product materials properties associated with the unit product materials;(c) operation of the hydrotreater; or(d) operation of one or more upstream equipment or downstream equipment, so that the controlling, during the hydrotreatment operation, causes the hydrotreatment operation to produce one or more of:(i) one or more intermediate materials each having one or more properties within a range of one or more target properties of the one or more intermediate materials,(ii) one or more unit product materials each having one or more properties within a range of one or more target properties of the one or more unit product materials, or(iii) one or more downstream materials each having one or more properties within a range of one or more target properties of the one or more downstream materials, and thereby to cause the hydrotreatment operation to achieve material outputs that more accurately and responsively converge on one or more of the target properties.

58. A system for enhancing fluid production for a hydrocracker operation, the system comprising: a hydrocracker unit to receive a feed and produce one or more fluids; a plurality of sensors to measure a parameter associated with the hydrocracker unit and each positioned at one of (a) proximate the hydrocracker unit or (b) within the hydrocracker unit; a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the hydrocracker unit and to control aspects of fluid flowing to or from the hydrocracker unit; one or more sample collection assemblies to collect samples of the fluid associated with the hydrocracker unit; one or more sample analysis assemblies to analyze each collected sample to provide properties of collected samples; and a hydrocracker controller in signal communication with the hydrocracker unit, the plurality of sensors, the plurality of refinery operation control devices, and the one or more sample analysis assemblies, and storing a trained machine learning model, the hydrocracker controller configured to: determine an output including predicted properties of feedstock and parameter settings of the plurality of refinery operation control devices and the hydrocracker unit based on application of one or more of (i) data measured by the plurality of sensors, (ii) data corresponding to analysis from the one or more sample analysis assemblies, or (iii) a target product and corresponding target properties to the trained machine learning model, and adjust one or more of an amount of feed or type of feed and parameters associated with the plurality of refinery operation control devices and hydrocracker unit based on the output to enhance fluid production.

59. A method compri sing : receiving, by a machine learning model, feedstock data from one or more refinery equipment providing feedstock to a hydrotreater, the feedstock data indicative of one or more operating parameters of the one or more refinery equipment, the machine learning model trained on historical feedstock data and historical data of operation of the hydrotreater, the machine learning model having a target product specification for a first product from the hydrotreater;predicting, by the machine learning model, a property of the feedstock based on the feedstock data; and generating, by the machine learning model, an output indicative of an adjustment to a parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification based on the feedstock data, the historical feedstock data, and historical data of operation of the hydrotreater.

60. The method of claim 59, wherein the feedstock data includes data indicative of an analyzed sample of the feedstock.

61. The method of any of claims 59 to 60, further comprising receiving, by a machine learning model, hydrotreater sensor data from a sensor disposed to measure a parameter associated with the hydrotreater and each positioned at one of (a) proximate the hydrotreater or (b) within the hydrotreater, wherein generating the output indicative of an adjustment to operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification is further based on the hydrotreater sensor data.

62. The method of any of claims 59 to 61, further comprising receiving, by the machine learning model, product sensor data indicative of property of the product produced by the hydrotreater.

63. The method of claim 62, wherein the product sensor data includes data generated from analyzing a sample of the product produced by the hydrotreater.

64. The method of claim 63, wherein the product sensor data includes a spectra of the sample indicative of properties of the product produced by the hydrotreater.

65. The method of any of claims 62 to 64, wherein generating the output indicative of an adjustment to operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification is further based on the product sensor data.

66. The method of any of claims 59 to 65, wherein the target product specification is a sulfur concentration in the product.

67. The method of any of claims 59 to 66, wherein the machine learning model is trained to generate operational parameters of the hydrotreater to remove sulfur to a target sulfur concentration in the product.

68. The method of any of claims 59 to 67, wherein the machine learning model is trained to maximize sulfur removal at a lowest possible temperature.

69. The method of any of claims 59 to 68, further comprising indicating, by the machine learning model, an adjustment of hydrotreater temperature to achieve a maximum aromatic saturation of the feedstock within hydrotreater operation constraints.

70. The method of claim 69, wherein the machine learning model is trained on the hydrotreater operation constraints.

71. The method of any of claims 59 to 70, further comprising: receiving a downstream request for a property of the product from the hydrotreater; and generating, by the machine learning model, an adjustment of an operational parameter of the hydrotreater to achieve the property of the product from the hydrotreater.

72. The method of claim 71, wherein the downstream request is a target aromatic saturation of the feedstock.

73. The method of any of claims 59 to 70, wherein the hydrotreater is operated as a distillate hydrotreater or a gas oil hydrotreater.

74. The method of any of claims 59 to 72, further comprising generating, by the machine learning model, a feed rate of kerosene to hydrotreat with the hydrotreater and a feed rate of kerosene sent to a gasoline pool.

75. The method of any of claims 59 to 74, further comprising generating, by the machine learning model, a feed rate of naphtha to hydrotreat with the hydrotreater and a feed rate of naphtha sent to a gasoline pool.

76. The method of any of claims 59 to 75, further comprising generating, by the machine learning model, a feed rate of distillate to hydrotreat with the hydrotreater and a feed rate of distillate sent to a gasoline pool.

77. The method of any of claims 59 to 76, further comprising generating, by the machine learning model, an adjustment of a distillation flash target for the feedstock.

78. The method of any of claims 59 to 77, further comprising generating, by the machine learning model, an adjustment of a distillation cutoff for the feedstock to reduce hard to treat sulfur molecules in the feedstock.

79. The method of any of claims 59 to 78, wherein the feedstock includes gas oil.

80. The method of any of claims 59 to 79, wherein the feedstock includes diesel.

81. The method of any of claims 59 to 80, wherein the feedstock includes kerosene.

82. The method of any of claims 59 to 81, wherein the feedstock includes naphtha.

83. The method of any of claims 59 to 82, wherein the machine learning model is installed in a hydrotreater controller of the hydrotreater.

84. The method of any of claims 59 to 83, further comprising adjusting the parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification based on the output.

85. The method of any of claims 59 to 84, further comprising: receiving, by the machine learning model, the operational parameters of a fluid catalytic cracker, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the fluid catalytic cracker; andgenerating an adjustment to the operational parameters of the hydrotreater based on the operational parameters of the fluid catalytic cracker.

86. The method of claim 85, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing fluid catalytic cracker yield constrained by process constraints of the fluid catalytic cracker.

87. The method of claims 85 or 86, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing a value of the fluid catalytic cracker, the method further comprising: receiving market pricing of fluid catalytic cracker products and costs of producing the fluid catalytic cracker products; and generating a profit algorithm based on market pricing of fluid catalytic cracker products and costs of producing the fluid catalytic cracker products.

88. The method of any of claims 59 to 87, further comprising: receiving, by the machine learning model, the operational parameters of a fluid catalytic cracker, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the fluid catalytic cracker; and adjusting a feed rate of hydrotreated feedstocks from the hydrotreater to the fluid catalytic cracker, based on the properties of the product.

89. The method of any of claims 59 to 88, wherein the machine learning model is trained to maximize aromatic saturation of the feedstock in the hydrotreater.

90. The method of any of claims 59 to 89, further comprising generating, by the machine learning model, an adjustment to an operational parameter of the hydrotreater based on a target aromatic saturation of the feedstock.

91. The method of any of claims 59 to 90, wherein the machine learning model is trained to improve aromatic saturation of the feedstock in the hydrotreater to provide hydrotreated feedstock to a fluid catalytic cracker within operational constraints of the fluid catalytic cracker.

92. The method of any of claims 59 to 91, wherein the machine learning model is trained to improve a potential profit of the hydrotreater providing hydrotreated feedstock to the fluid catalytic cracker within operational constraints of the fluid catalytic cracker.

93. The method of any of claims 59 to 92, wherein the machine learning model is trained to maximize combined profit of the products of the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

94. The method of any of claims 59 to 93, wherein operational constraints of a fluid catalytic cracker include one or more of minimum regeneration temperature, wet gas compressor capacity, fluid catalytic cracker fractionation limits, surge drum capacity, charge pump capacity, heater capacity, reactor capacity, regenerator capacity, main air blower capacity, and gas concentration section limits.

910. The method of any of claims 85 to 94, further comprising generating, by the machine learning model, an arrangement of feedstocks to be processed in by the fluid catalytic cracker.

911. The method of claim 910, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker is made to increase catalyst life of the catalysts of the fluid catalytic cracker.

912. The method of any of claims 910 to 911, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker includes a period of time for processing of a first feedstock followed by a longer period of time for processing a second feedstock.

913. The method of any of claims 910 to 912, wherein the second feedstock includes one or more feedstocks that cause less deactivation of the catalysts of the fluid catalytic cracker than the first feedstock.

95. The method of any of claims 59 to 913, wherein the target product specification is a sulfur target, wherein the machine learning model is trained to maximize aromatic saturation constrained by the target product specification.

96. The method of any of claims 85 to 95, wherein the hydrotreater is operated as a distillate hydrotreater or a gas oil hydrotreater.

97. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 59 to 96.

98. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 59 to 96.

99. A method comprising: receiving, by a machine learning model, a first feedstock data from one or more refinery equipment providing a first feedstock to be hydrotreated, the first feedstock data indicative of one or more properties or composition of the first feedstock, the machine learning model having a target product specification for the first feedstock; receiving, by the machine learning model, current data of operational parameters of a first hydrotreater and a second hydrotreater, wherein the machine learning model trained on historical first feedstock data, historical data of operational parameters of a first hydrotreater, and historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater and the second hydrotreater; generating, by the machine learning model, a recommendation of one of the first hydrotreater or the second hydrotreater to send the first feedstock to be hydrotreated; and generating, by the machine learning model, an output indicative of an adjustment to a parameter of operation of the recommended one of the first hydrotreater or the second hydrotreater to hydrotreat the first feedstock to accurately produce the product to the target product specification for the first feedstock based on the feedstock data, the historical feedstock data, historical data of operational parameters of a first hydrotreater, and historical data of operational parameters of a second hydrotreater, and historical data of the feedstocks and products of the first hydrotreater and the second hydrotreater.

100. The method of claim 99, further comprising: receiving, by the machine learning model, a second feedstock data from one or more refinery equipment providing a second feedstock to be hydrotreated, the second feedstockdata indicative of one or more properties or composition of the second feedstock, the machine learning model having a target product specification for the second feedstock; generating, by the machine learning model, a recommendation of one of the first hydrotreater or the second hydrotreater to send the second feedstock to be hydrotreated; and generating, by the machine learning model, an output indicative of an adjustment to a parameter of operation of the recommended one of the first hydrotreater or the second hydrotreater to hydrotreat the second feedstock to accurately produce the product to the target product specification for the second feedstock.

101. The method of any of claims 99 to 100, wherein the first feedstock data is indicative of one or more operating parameters of the one or more refinery equipment.

102. The method of any of claims 99 to 101, wherein the feedstock data includes data indicative of an analyzed sample of the feedstock.

103. The method of any of claims 99 to 102, further comprising receiving, by a machine learning model, hydrotreater sensor data from a sensor disposed to measure an operational parameter associated with the hydrotreater and positioned at one of (a) proximate the hydrotreater or (b) within the hydrotreater, wherein generating the output indicative of an adjustment to operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification is further based on the hydrotreater sensor data.

104. The method of any of claims 99 to 103, further comprising receiving, by the machine learning model, product sensor data indicative of property of the product produced by the hydrotreater.

105. The method of claim 104, wherein the product sensor data includes data generated from analyzing a sample of the product produced by the hydrotreater.

106. The method of claim 105, wherein the product sensor data includes a spectra of the sample indicative of properties of the product produced by the hydrotreater.

107. The method of any of claims 104 to 106, wherein generating the output indicative of an adjustment to operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification is further based on the product sensor data.

108. The method of any of claims 99 to 107, wherein the target product specification is a sulfur concentration in the product.

109. The method of any of claims 99 to 108, wherein the machine learning model is trained to generate operational parameters of the hydrotreater to remove sulfur to a predetermined threshold of sulfur concentration in the product.

110. The method of any of claims 99 to 109, wherein the machine learning model is trained to maximize sulfur removal at a lowest possible temperature.

111. The method of any of claims 99 to 110, further comprising indicating, by the machine learning model, an adjustment of hydrotreater temperature to achieve a maximum aromatic saturation of the feedstock within hydrotreater operation constraints.

112. The method of claim 111, wherein the machine learning model is trained on the hydrotreater operation constraints.

113. The method of any of claims 99 to 112, further comprising: receiving a downstream request for a property of the product from the hydrotreater; and generating, by the machine learning model, an adjustment of an operational parameter of the hydrotreater to achieve the property of the product from the hydrotreater.

114. The method of claim 113, wherein the downstream request for a property of the product from the hydrotreater is a target aromatic saturation of the feedstock.

115. The method of any of claims 99 to 114, further comprising generating, by the machine learning model, a feed rate of kerosene to hydrotreat with the hydrotreater and a feed rate of kerosene sent to a gasoline pool.

116. The method of any of claims 99 to 115, further comprising generating, by the machine learning model, a feed rate of naphtha to hydrotreat with the hydrotreater and a feed rate of naphtha sent to a gasoline pool.

117. The method of any of claims 99 to 116, further comprising generating, by the machine learning model, a feed rate of distillate to hydrotreat with the hydrotreater and a feed rate of distillate sent to a gasoline pool.

118. The method of any of claims 99 to 117, further comprising generating, by the machine learning model, an adjustment of a distillation flash target for the feedstock.

119. The method of any of claims 99 to 118, further comprising generating, by the machine learning model, an adjustment of a distillation cutoff for the feedstock to reduce hard to treat sulfur molecules in the feedstock.

120. The method of any of claims 99 to 119, wherein the feedstock includes gas oil.

121. The method of any of claims 99 to 120, wherein the feedstock includes diesel.

122. The method of any of claims 99 to 121, wherein the feedstock includes kerosene.

123. The method of any of claims 99 to 122, wherein the feedstock includes naphtha.

124. The method of any of claims 99 to 123, wherein the machine learning model is installed in a hydrotreater controller of the hydrotreater.

125. The method of any of claims 99 to 124, further comprising adjusting the parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to the target product specification based on the output.

126. The method of any of claims 99 to 125, further comprising:receiving, by the machine learning model, the operational parameters of a fluid catalytic cracker, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the fluid catalytic cracker; and generating an adjustment to the operational parameters of the hydrotreater based on the operational parameters of the fluid catalytic cracker.

127. The method of claim 126, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing fluid catalytic cracker yield constrained by process constraints of the fluid catalytic cracker.

128. The method of claim 126, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing a value from the fluid catalytic cracker, the method further comprising: receiving market pricing of fluid catalytic cracker products and costs of producing the fluid catalytic cracker products; and generating a profit algorithm based on market pricing of fluid catalytic cracker products and costs of producing the fluid catalytic cracker products.

129. The method of any of claims 99 to 128, further comprising: receiving, by the machine learning model, the operational parameters of a fluid catalytic cracker, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of a fluid catalytic cracker; and adjusting a feed rate of hydrotreated feedstocks from the hydrotreater to the fluid catalytic cracker, based on the properties of the product.

130. The method of any of claims 99 to 129, wherein the machine learning model is trained to maximize aromatic saturation of the feedstock in the hydrotreater.

131. The method of any of claims 99 to 130, further comprising generating, by the machine learning model, an adjustment to an operational parameter of the hydrotreater based on a target aromatic saturation of the feedstock.

132. The method of any of claims 99 to 131, wherein the machine learning model is trained to improve aromatic saturation of the feedstock in the hydrotreater to providehydrotreated feedstock to a fluid catalytic cracker within operational constraints of the fluid catalytic cracker.

920. The method of any of claims 1 to 53, further comprising generating, by the machine learning model, an arrangement of feedstocks to be processed in by the fluid catalytic cracker.

901. The method of claim 900, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker is made to increase catalyst life of the catalysts of the fluid catalytic cracker.

902. The method of any of claims 900 to 901, wherein the arrangement of feedstocks to be processed in by the fluid catalytic cracker includes a period of time for processing of a first feedstock followed by a longer period of time for processing a second feedstock.

903. The method of any of claims 900 to 902, wherein the second feedstock includes one or more feedstocks that cause less deactivation of the catalysts of the fluid catalytic cracker than the first feedstock.

133. The method of any of claims 126 to 132, wherein the hydrotreater is operated as a distillate hydrotreater or a gas oil hydrotreater.

134. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 99 to 133.

135. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 99 to 133.

136. A method compri sing : receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater;receiving, by the machine learning model, current operating parameters of the hydrotreater, wherein the machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater; generating, by the machine learning model, a prediction of a state of catalysts within a hydrotreater reactor based on the current operating parameters of the hydrotreater; and generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater based on the predicted state of the catalysts within the hydrotreater reactor and the feedstock data.

137. The method of claim 136, further comprising publishing, by the machine learning model, the prediction of a state of the catalysts within the hydrotreater reactor.

138. The method of any of claims 136 to 137, further comprising generating, by the machine learning model, a replacement recommendation of the catalysts within the hydrotreater reactor.

139. The method of any of claims 136 to 138, further comprising publishing, by the machine learning model, a replacement recommendation of the catalysts within the hydrotreater reactor.

140. The method of any of claims 136 to 139, further comprising publishing, by the machine learning model, the adjustment to the current operating parameters of the hydrotreater.

141. The method of any of claims 136 to 140, further comprising implementing the adjustment to the current operating parameters of the hydrotreater.

142. The method of any of claims 136 to 141, wherein the current feedstock data indicates a change of a property of feedstock to be fed into the hydrotreater reactor.

143. The method of claim 142, further comprising generating, by the machine learning model, a second adjustment to the current operating parameters of the hydrotreater based on the current feedstock data.

144. The method of any of claims 136 to 143, wherein the feedstock data is indicative of one or more operating parameters of one or more refinery equipment, the method further comprising generating, by the machine learning model, a predicted property of the feedstock for the hydrotreater based on the one or more operating parameters of the one or more refinery equipment.

145. The method of any of claims 136 to 144, wherein the feedstock data includes data indicative of an analyzed sample of the feedstock.

146. The method of any of claims 136 to 145, further comprising receiving, by a machine learning model, current hydrotreater sensor data from a sensor package disposed to measure an operational parameter of the hydrotreater, wherein generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater is based on the current hydrotreater sensor data.

147. The method of any of claims 136 to 146, further comprising generating, by the machine learning model, a predicted flow rate of hydrogen gas to quench a temperature within a reactor of the hydrotreater based on the feedstock data.

148. The method of claim 147, further comprising adjusting a feed rate of hydrogen gas to quench a temperature within the reactor of the hydrotreater to the predicted flow rate.

149. The method of any of claims 136 to 148, further comprising generating, by the machine learning model, a plurality of predicted hydrogen feed rates based on feedstock data and historical data, wherein each predicted hydrogen feed rate of the plurality of predicted hydrogen feed rates is for a different location within the hydrotreater.

150. The method of claim 149, wherein a first predicted hydrogen feed rate of the plurality of predicted hydrogen feed rates is for hydrogen gas injected before an inlet of a reactor of the hydrotreater, wherein a second predicted hydrogen feed rate of the plurality of predicted hydrogen feed rates is for hydrogen gas injected into the reactor of the hydrotreater between the inlet and an outlet of the reactor.

151. The method of claims 149 or 150, further comprising adjusting a plurality of hydrogen feed rates to the plurality of predicted hydrogen feed rates.

152. The method of any of claims 136 to 151, further comprising generating, by the machine learning model, a total hydrogen feed rate for the hydrotreater based on the current feedstock data.

153. The method of any of claims 136 to 152, further comprising: receiving, by the machine learning model, the operational parameters of a downstream process, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the downstream process; and generating an adjustment to the operational parameters of the hydrotreater based on the operational parameters of the downstream process.

154. The method of any of claims 136 to 153, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing downstream process yield constrained by process constraints of the downstream process.

155. The method of any of claims 136 to 154, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing a value of hydrotreater, the method further comprising: receiving market pricing of downstream process products and costs of producing the downstream process products; and generating a profit algorithm based on market pricing of downstream process products and costs of producing the downstream process products.

156. The method of any of claims 136 to 155, further comprising: receiving, by the machine learning model, the operational parameters of a downstream process, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the downstream process; andadjusting a feed rate of hydrotreated feedstocks from the hydrotreater to the downstream process, based on the properties of the product.

157. The method of any of claims 136 to 156, wherein the machine learning model is trained to maximize aromatic saturation of the feedstock in the hydrotreater.

158. The method of any of claims 136 to 157, wherein generating the adjustment to an operational parameter of the hydrotreater is based on a target aromatic saturation of the feedstock.

159. The method of any of claims 136 to 158, wherein the machine learning model is trained to improve aromatic saturation of the feedstock in the hydrotreater to provide hydrotreated feedstock to a downstream process within operational constraints of the downstream process.

160. The method of any of claims 153 to 159, wherein the downstream process is a fluid catalytic cracker.

161. The method of any of claims 157 to 160, wherein the hydrotreater is operated as a distillate hydrotreater or a gas oil hydrotreater.

162. The method of any of claims 153 to 161, wherein the downstream process is a hydrocarbon pool controller.

163. The method of any of claims 136 to 162, wherein the hydrotreater includes a first hydrotreating reactor and a second hydrotreating reactor in parallel, wherein the machine learning model is programmed to optimize catalyst usage while managing the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window.

164. The method of claims 163, wherein the machine learning model manages the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window through managingoperating temperatures of the first hydrotreating reactor and the second hydrotreating reactor.

165. The method of claim 163, wherein the first catalyst replacement window overlaps with the second catalyst replacement window.

166. The method of claim 163, wherein the first catalyst replacement window does not overall with the second catalyst replacement window.

167. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about twenty-five percent to about seventy-five percent deactivated.

168. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about fifty percent deactivated.

169. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about twenty-five percent to about seventy-five percent of its predicted useful life remaining.

170. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about fifty percent of its predicted useful life remaining.

171. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has less than twenty-five percent of its predicted useful life remaining.

172. The method of claim 163, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the secondhydrotreating reactor has more than seventy-five percent of its predicted useful life remaining.

173. The method of any of claims 136 to 172, wherein the hydrotreater includes a first hydrotreating reactor and a second hydrotreating reactor in parallel, the method further comprising generating, by the machine learning model, a predicted state of the catalysts within the first hydrotreating reactor and the second hydrotreating reactor based on the feedstock data, the operational parameters, and product data of the first hydrotreating reactor and the second hydrotreating reactor.

174. The method of claim 173, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstream process, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor based on the predicted state of the catalysts within the first hydrotreating reactor and the second hydrotreating reactor.

175. The method of any of claims 173 to 174, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstream process, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor to achieve a lower operating temperature in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property.

176. The method of any of claims 173 to 175, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstream process, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor to achieve a lower ratio of hydrogen to feedstock in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property.

177. The method of any of claims 136 to 176, further comprising:generating, by the machine learning model, a value for the products of the hydrotreater, a value for costs of operating the hydrotreater, a value for hydrogen gas fed to the hydrotreater, and a value for the feedstocks fed to the hydrotreater; and generating a recommendation, by the machine learning model, to improve the value of the hydrotreater operation based on the value for the products of the hydrotreater, less the value for the costs of operating the hydrotreater, less the value of the hydrogen gas fed to the hydrotreater, and less the value for the feedstocks fed to the hydrotreater while achieving a target product property.

178. The method of claim 177, further comprising adjusting an operational parameter of the hydrotreater based on the recommendation to improve a value of the hydrotreater operation.

179. The method of any of claims 136 to 178, further comprising: receiving, by the machine learning model, a target product property of the product hydrotreated by the hydrotreater; and generating, by the machine learning model, a predicted minimum operating temperature required to hydrotreat the feedstock to achieve the target product property.

180. The method of claim 179, further comprising adjusting the operating temperature of the hydrotreater to the predicted minimum operating temperature required to hydrotreat the feedstock to achieve the target product property.

181. The method of claim 179, further comprising generating, by the machine learning model, a confidence interval of temperature above the predicted minimum operating temperature required to hydrotreat the feedstock to achieve the target product property based on the historical data and predicted current state of the hydrotreater reactor.

182. The method of claim 181, further comprising adjusting the operating temperature of the hydrotreater to within the generated confidence interval of temperature above the predicted minimum operating temperature.

183. The method of any of claims 136 to 182, further comprising generating, by the machine learning model, an optimized feed rates of feedstock and hydrogen gas tomaximize catalyst life constrained by a target product property of a product hydrotreated by the hydrotreater.

184. The method of any of claims 136 to 183, further comprising generating, by the machine learning model, a target inlet temperature of a reactor and a target outlet temperature of the reactor based on the predicted state of the catalysts, the feedstock data, and a target product property of a product hydrotreated by the hydrotreater.

185. The method of claim 184, wherein the target product property is sulfur content in the product.

186. The method of claim 185, wherein the target product property is measured by a sulfur analyzer.

187. The method of claims 184 or 186, wherein the target product property is measured from a sample to generate lab data, wherein the machine learning model correlates the feedstock data and an operating temperature of the hydrotreater with the lab data to generate a predicted target property.

188. The method of any of claims 183 to 184, wherein the target product property is a desired aromatic saturation rate of the product hydrotreated by the hydrotreater.

189. The method of any of claims 183 to 188, further comprising adjusting the parameter of operation of the hydrotreater to hydrotreat the feedstock to accurately produce the product to a target product specification.

190. The method of any of claims 175 to 189, wherein a target product property is one or more of an octane-barrels target, a cetane number target, a pour point target, a smoke point target, or a flash point target.

191. The method of any of claims 177 to 190, further comprising accessing current business data, wherein the values are based on current business data.

192. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 136 to 191.

193. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 136 to 191.

194. A method compri sing : receiving, by a machine learning model, current feedstock data indicative of a property of a feedstock for a hydrotreater; receiving, by the machine learning model, current operating parameters of the hydrotreater, wherein the machine learning model is trained on historical data indicative of feedstocks hydrotreated by the hydrotreater, the operating parameters used by the hydrotreater, and products produced by the hydrotreater, wherein the machine learning model is trained to meet a sulfur target, while maximizing aromatic saturation constrained by available hydrogen gas feed rates; generating, by the machine learning model, a target aromatic saturation based on the sulfur target, the feedstock data, and available hydrogen gas feed rates; and generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater based on the target aromatic saturation of the feedstocks within the hydrotreater, the sulfur target, the feedstock data, and available hydrogen gas feed rates.

195. The method of claim 194, further comprising generating, by the machine learning model, a prediction of a state of catalysts within a hydrotreater reactor based on the current operating parameters of the hydrotreater.

196. The method of claim 195, further comprising publishing, by the machine learning model, the prediction of a state of the catalysts within the hydrotreater reactor.

197. The method of any of claims 195 to 196, further comprising generating, by the machine learning model, a replacement recommendation of the catalysts within the hydrotreater reactor.

198. The method of any of claims 194 to 197, further comprising publishing, by the machine learning model, a replacement recommendation of catalysts within a hydrotreater reactor.

199. The method of any of claims 194 to 198, further comprising publishing, by the machine learning model, the adjustment to the current operating parameters of the hydrotreater.

200. The method of any of claims 194 to 199, further comprising implementing the adjustment to the current operating parameters of the hydrotreater.

201. The method of any of claims 194 to 200, wherein the current feedstock data indicates a change of a property of feedstock to be fed into a hydrotreater reactor.

202. The method of claim 201, further comprising generating, by the machine learning model, a second adjustment to the current operating parameters of the hydrotreater based on the current feedstock data.

203. The method of any of claims 194 to 202, further comprising generating, by the machine learning model, a predicted minimum operating temperature required to hydrotreat the feedstock to achieve the sulfur target and the target aromatic saturation of the feedstocks within a hydrotreater reactor.

204. The method of claim 203, wherein generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater includes adjusting an operating temperature to achieve the predicted minimum operating temperature.

205. The method of claim 203, further comprising generating, by the machine learning model, a confidence interval of temperature above the predicted minimum operating temperature required to hydrotreat the feedstock to achieve a target product property based on the historical data and predicted current state of the hydrotreater reactor.

206. The method of claim 205, wherein generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater includes adjusting theoperating temperature of the hydrotreater to the generated confidence interval of temperature above the predicted minimum operating temperature.

207. The method of any of claims 194 to 206, wherein the feedstock data is indicative of one or more operating parameters of one or more refinery equipment, the method further comprising generating, by the machine learning model, a predicted property of the feedstock for the hydrotreater based on the one or more operating parameters of the one or more refinery equipment.

208. The method of any of claims 194 to 207, wherein the feedstock data includes data indicative of an analyzed sample of the feedstock.

209. The method of any of claims 194 to 208, further comprising receiving, by a machine learning model, current hydrotreater sensor data from a sensor package disposed to measure an operational parameter of the hydrotreater, wherein generating, by the machine learning model, an adjustment to the current operating parameters of the hydrotreater is based on the current hydrotreater sensor data.

210. The method of any of claims 194 to 209, further comprising generating, by the machine learning model, a predicted flow rate of hydrogen gas to quench a temperature within a reactor of the hydrotreater based on the feedstock data.

211. The method of claim 210, further comprising adjusting a feed rate of hydrogen gas to quench a temperature within the reactor of the hydrotreater to the predicted flow rate.

212. The method of any of claims 194 to 211, further comprising generating, by the machine learning model, a plurality of predicted hydrogen feed rates based on feedstock data and historical data, wherein each predicted hydrogen feed rate of the plurality of predicted hydrogen feed rates is for a different location within the hydrotreater.

213. The method of claim 212, wherein a first predicted hydrogen feed rate of the plurality of predicted hydrogen feed rates is for hydrogen gas injected before an inlet of a reactor of the hydrotreater, wherein a second predicted hydrogen feed rate of the pluralityof predicted hydrogen feed rates is for hydrogen gas injected into the reactor of the hydrotreater between the inlet and an outlet of the reactor.

214. The method of claims 212 or 213, further comprising adjusting a plurality of hydrogen feed rates to the plurality of predicted hydrogen feed rates.

215. The method of any of claims 194 to 214, further comprising generating, by the machine learning model, a total hydrogen feed rate for the hydrotreater based on the current feedstock data.

216. The method of any of claims 194 to 215, further comprising generating, by the machine learning model, an optimized feed rates of feedstock and hydrogen gas to maximize catalyst life constrained by a target property of a product hydrotreated by the hydrotreater.

217. The method of any of claims 194 to 216, further comprising generating, by the machine learning model, a target inlet temperature of a reactor and a target outlet temperature of the reactor based on a predicted state of catalysts and a target property of a product hydrotreated by the hydrotreater.

218. The method of claim 217, wherein the target property is sulfur content in the product.

219. The method of claim 218, wherein the target property is measured by a sulfur analyzer.

220. The method of claim 218, wherein the target property is measured from a sample to generate lab data, wherein the machine learning model correlates the feedstock data and an operating temperature of the hydrotreater with the lab data to generate a predicted target property.

221. The method of any of claims 194 to 220, further comprising: receiving, by the machine learning model, the operational parameters of a downstream process, the machine learning model trained on historical data of theoperational parameters, feedstock properties, and product properties of the downstream process; and generating an adjustment to the operational parameters of the hydrotreater based on the operational parameters of the downstream process.

222. The method of any of claims 194 to 221, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing downstream process yield constrained by process constraints of the downstream process.

223. The method of any of claims 194 to 222, wherein generating an adjustment to the operational parameters of the hydrotreater is further based on the machine learning model optimizing value of the hydrotreater, the method further comprising: receiving market pricing of downstream process products and costs of producing the downstream process products; and generating a profit algorithm based on market pricing of downstream process products and costs of producing the downstream process products.

224. The method of any of claims 194 to 223, further comprising: receiving, by the machine learning model, the operational parameters of a downstream process, the machine learning model trained on historical data of the operational parameters, feedstock properties, and product properties of the downstream process; and adjusting a feed rate of hydrotreated feedstocks from the hydrotreater to the downstream process, based on the properties of the product.

225. The method of any of claims 194 to 224, wherein generating the adjustment to an operational parameter of the hydrotreater is based on a target aromatic saturation of the feedstock.

226. The method of any of claims 194 to 225, wherein a downstream process is a fluid catalytic cracker.

227. The method of any of claims 194 to 226, wherein a downstream process is a hydrocarbon pool controller.

228. The method of any of claims 194 to 227, wherein the hydrotreater is operated as a distillate hydrotreater or a gas oil hydrotreater.

229. The method of any of claims 194 to 228, wherein the hydrotreater includes a first hydrotreating reactor and a second hydrotreating reactor in parallel, wherein the machine learning model is programmed to optimize catalyst usage while managing the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window.

230. The method of claims 229, wherein the machine learning model manages the first hydrotreating reactor toward a first catalyst replacement window and the second hydrotreating reactor toward a second catalyst replacement window through managing operating temperatures of the first hydrotreating reactor and the second hydrotreating reactor.

231. The method of claim 229, wherein the first catalyst replacement window overlaps with the second catalyst replacement window.

232. The method of claim 229, wherein the first catalyst replacement window does not overlap with the second catalyst replacement window.

233. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about twenty-five percent to about seventy-five percent deactivated.

234. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor is about fifty percent deactivated.

235. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about twenty-five percent to about seventy-five percent of its predicted useful life remaining.

236. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has about fifty percent of its predicted useful life remaining.

237. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has less than twenty-five percent of its predicted useful life remaining.

238. The method of claim 229, wherein the first catalyst replacement window is predicted by the machine learning model to overlap when a catalyst of the second hydrotreating reactor has more than seventy-five percent of its predicted useful life remaining.

239. The method of any of claims 194 to 238, wherein the hydrotreater includes a first hydrotreating reactor and a second hydrotreating reactor in parallel, the method further comprising generating, by the machine learning model, a predicted state of catalysts within the first hydrotreating reactor and the second hydrotreating reactor based on the feedstock data, the operational parameters, and product data of the first hydrotreating reactor and the second hydrotreating reactor.

240. The method of claim 239, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstream process, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor based on the predicted state of the catalysts within the first hydrotreating reactor and the second hydrotreating reactor.

241. The method of any of claims 239 to 240, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstreamprocess, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor to achieve a lower WABT in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property.

242. The method of any of claims 239 to 241, wherein the feedstocks include a first feedstock from a first upstream process and a second feedstock from a second upstream process, the method further comprising generating, by the machine learning model, a distribution decision directing the first feedstock to the first hydrotreating reactor and the second feedstock to the second hydrotreating reactor to achieve a lower ratio of hydrogen to feedstock in the first hydrotreating reactor and the second hydrotreating reactor while achieving a target product property.

243. The method of any of claims 194 to 242, further comprising: generating, by the machine learning model, a value for the products of the hydrotreater, a value for costs of operating the hydrotreater, a value for the hydrogen gas fed to the hydrotreater, and a value for the feedstocks fed to the hydrotreater; and generating a recommendation, by the machine learning model, to improve the value of the hydrotreater operation based on the value for the products of the hydrotreater, less the value for the costs of operating the hydrotreater, less the value of the hydrogen gas fed to the hydrotreater, and less the value for the feedstocks fed to the hydrotreater while achieving a target product property.

244. The method of any of claims 194 to 243, wherein the target property is one or more of an octane-barrels target, a sulfur concentration target, a cetane number target, a pour point target, a smoke point target, or a flash point target.

245. The method of claim 243, further comprising adjusting an operational parameter of the hydrotreater based on the recommendation to improve the value of the hydrotreater operation.

246. The method of any of claims 243 to 245, further comprising accessing current business data, wherein the values are based on current business data.

247. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 194 to 246.

248. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 194 to 246.

249. A method comprising: generating, by a machine learning model, a value for products of a hydrotreater, a value for hydrogen gas fed to the hydrotreater, and a value for feedstocks fed to the hydrotreater, a value for the products of a hydrocracker, a value for the hydrogen gas fed to the hydrocracker, a value for the feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker, wherein a product of the hydrotreater includes hydrotreated gas oil fed to the fluid catalytic cracker, wherein a product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker, wherein the machine learning model was trained with historical data including feedstock data, operational data, and product data of the hydrotreater, the hydrocracker, and the fluid catalytic cracker; and generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve a combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrotreater and the hydrocracker, and the value for the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

250. The method of claim 249, further comprising generating, by the machine learning model, a value for costs of operating the hydrotreater, a value for the costs of operating the hydrocracker, and a value for the costs of operating the fluid catalytic cracker, wherein generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improvea combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker is further based on the value for the costs of operating the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

251. The method of claims 249 or 250, further comprising adjusting the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker.

252. The method of any of claims 249 to 251, further comprising accessing current business data, wherein the values are based on the current business data.

253. The method of claim 252, wherein the business data includes a current market price for each of the products of the fluid catalytic cracker and hydrocracker.

254. The method of any of claims 249 to 253, further comprising generating an adjustment, by the machine learning model, to a cutoff of the gas oil produced by atmospheric distillation or vacuum distillation to provide a more desirable feedstock property in the feedstocks of the hydrotreater or hydrocracker.

255. The method of any of claims 249 to 254, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker is constrained by a target product property of one or more products of the hydrotreater, the hydrocracker, or the fluid catalytic cracker.

256. The method of claim 255, wherein the target property is aromatic saturation.

257. The method of claim 255, wherein the target property is maximum aromatic saturation.

258. The method of claim 255, wherein the target property is maximum aromatic saturation limited by sulfur content.

259. The method of claim 255, wherein the target property is octane-barrels.

260. The method of any of claims 249 to 259, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker is constrained by limitations of the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

261. The method of claim 260, wherein the limitations include catalyst deactivation states of catalysts of the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

262. The method of any of claims 249 to 261, wherein the generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker includes: generating a prediction of a deactivation state of a catalyst of the fluid catalytic cracker; and generating an adjustment in an amount of unconverted gas oil from the hydrocracker to the fluid catalytic cracker to slow deactivation of the catalyst of the fluid catalytic cracker.

263. The method of any of claims 249 to 262, wherein a product of the fluid catalytic cracker is a feedstock to a gasoline desulfurization unit, the method further comprising generating, by the machine learning model, an adjustment to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker is based on product data from the gasoline desulfurization unit.

264. The method of claim 263, further comprising generating, by a machine learning model, values for products of the gasoline desulfurization unit, wherein a product of the fluid catalytic cracker is a feedstock to a gasoline desulfurization unit, the method further comprising generating, by the machine learning model, an adjustment to an operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker is based on the value of the product of the gasoline desulfurization unit.

265. The method of any of claims 249 to 264, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrotreater, the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrotreater, the hydrocracker, and the fluid catalytic cracker is based on an objective function including the value of a product of the hydrocracker, the value of a product of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrotreater and the hydrocracker, and the value of the feedstocks fed to the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

266. The method of claim 265, wherein the objective function includes the value of costs of operating the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

267. The method of claim 265, wherein the objective function includes the value for a product of a gasoline desulfurization unit, wherein the gasoline desulfurization unit receives a feedstock from the fluid catalytic cracker.

268. The method of any of claims 265 to 267, wherein a feedstock is a byproduct from an upstream process producing a more valuable product, wherein a value of the byproduct is discounted to reflect a higher value of the products from the upstream process.

269. The method of claim 268, wherein the byproduct is an unconverted gas oil processed by the hydrocracker to be processed in a fluid catalytic cracker.

270. The method of any of claims 265 to 269, wherein the objective function includes a positive value of a product of the hydrocracker based on a property of the product of the hydrocracker and a positive value of a product of the fluid catalytic cracker based on a property of the product of the fluid catalytic cracker.

271. The method of claim 270, wherein the objective function includes a negative value of costs of operating the hydrotreater, the hydrocracker, and the fluid catalytic cracker.

272. The method of claim 270, wherein the objective function includes a positive value for a product of a gasoline desulfurization unit based on based on a property of the productof the gasoline desulfurization unit, wherein the gasoline desulfurization unit receives a feedstock from the fluid catalytic cracker.

273. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 249 to 272.

274. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 249 to 272.

275. A method compri sing : generating, by a machine learning model, a value for products of a hydrocracker, a value for hydrogen gas fed to the hydrocracker, a value for feedstocks fed to the hydrocracker, a value for the products of a fluid catalytic cracker, and a value for the feedstocks fed to the fluid catalytic cracker, wherein a product of the hydrocracker includes unconverted gas oil fed to the fluid catalytic cracker, wherein the machine learning model was trained with historical data including feedstock data, operational data, and product data of the hydrocracker and the fluid catalytic cracker; and generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to improve a combined value of the hydrocracker and the fluid catalytic cracker, based on the value for the products of the hydrocracker, the value of the products of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrocracker, and the value for the feedstocks fed to the hydrocracker and the fluid catalytic cracker.

276. The method of claim 275, further comprising generating, by the machine learning model, a value for costs of operating the hydrocracker and a value for the costs of operating the fluid catalytic cracker, wherein generating an adjustment, by the machine learning model, to an operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to improve a combined value of the hydrocracker and the fluid catalytic cracker is further based on the value for the costs of operating the hydrocracker and the fluid catalytic cracker.

277. The method of claims 275 or 276, further comprising adjusting the operational parameter of one or more of the hydrocracker or the fluid catalytic cracker.

278. The method of any of claims 275 to 277, further comprising accessing business data, wherein the values are based on current business data.

279. The method of claim 278, wherein the business data includes a current market price for each of the products of the fluid catalytic cracker and hydrocracker.

280. The method of any of claims 275 to 279, further comprising generating an adjustment, by the machine learning model, to a cutoff of the gas oil produced by atmospheric distillation or vacuum distillation to provide a more desirable feedstock property in the feedstocks of the hydrocracker.

281. The method of any of claims 275 to 280, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrocracker, or the fluid catalytic cracker to improve the combined value of the hydrocracker and the fluid catalytic cracker is constrained by a target product property of one or more products of the hydrocracker or the fluid catalytic cracker.

282. The method of claim 281, wherein the target property is aromatic saturation.

283. The method of claim 281, wherein the target property is maximum aromatic saturation.

284. The method of claim 281, wherein the target property is maximum aromatic saturation limited by sulfur content.

285. The method of claim 281, wherein the target property is octane-barrels.

286. The method of any of claims 275 to 285, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to improve the combined value of the hydrocracker and thefluid catalytic cracker is constrained by limitations of the hydrocracker and the fluid catalytic cracker.

287. The method of claim 286, wherein the limitations include catalyst deactivation states of catalysts of the hydrocracker and the fluid catalytic cracker.

288. The method of any of claims 275 to 287, wherein a product of the fluid catalytic cracker is a feedstock to a gasoline desulfurization unit, the method further comprising generating, by the machine learning model, an adjustment to an operational parameter of one or more of the hydrocracker or the fluid catalytic cracker is based on product data from the gasoline desulfurization unit.

289. The method of claim 288, further comprising generating, by a machine learning model, values for products of the gasoline desulfurization unit, wherein a product of the fluid catalytic cracker is a feedstock to the gasoline desulfurization unit, the method further comprising generating, by the machine learning model, an adjustment to an operational parameter of one or more of a hydrotreater, the hydrocracker, the fluid catalytic cracker, or the gasoline desulfurization unit is based on the value of the product of the gasoline desulfurization unit.

290. The method of any of claims 275 to 289, wherein generating the adjustment, by the machine learning model, to the operational parameter of one or more of the hydrocracker or the fluid catalytic cracker to improve the combined value of the hydrocracker and the fluid catalytic cracker is based on an objective function including the value of a product of the hydrocracker, the value of a product of the fluid catalytic cracker, the value of the hydrogen gas fed to the hydrocracker, and the value of the feedstocks fed to the hydrocracker and the fluid catalytic cracker.

291. The method of claim 290, wherein the obj ective function includes the value of costs of operating the hydrocracker and the fluid catalytic cracker.

292. The method of claim 290, wherein the objective function includes the value for a product of a gasoline desulfurization unit, wherein the gasoline desulfurization unit receives a feedstock from the fluid catalytic cracker.

293. The method of any of claims 275 to 292, wherein a feedstock is a byproduct from an upstream process producing a more valuable product, wherein a value of the byproduct is discounted to reflect a higher value of the products from the upstream process.

294. The method of claim 293, wherein the byproduct is an unconverted gas oil processed by the hydrocracker to be processed in a fluid catalytic cracker.

295. The method of any of claims 290 to 294, wherein the objective function includes a positive value of a product of the hydrocracker, a positive value of a product of the fluid catalytic cracker, a negative value of the hydrogen gas fed to the hydrocracker.

296. The method of claim 295, wherein the objective function includes a negative value of costs of operating the hydrocracker and the fluid catalytic cracker.

297. The method of claim 295, wherein the objective function includes a positive value for a product of a gasoline desulfurization unit, wherein the gasoline desulfurization unit receives a feedstock from the fluid catalytic cracker.

298. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 275 to 297.

299. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 275 to 297.

300. A method compri sing : receiving business data, by a machine learning model, wherein the machine learning model is trained on data indicative of one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, products of the fluid catalytic cracker, wherein the business data is related to the one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, products of the fluid catalytic cracker;generating, by the machine learning model, values for the one or more feedstocks to a hydrocracker, operational parameters of the hydrocracker, products of the hydrocracker, one or more feedstocks of a fluid catalytic cracker, operational parameters of the fluid catalytic cracker, products of the fluid catalytic cracker; and generating, by the machine learning model, one or more adjustments to operational parameters of the hydrocracker and fluid catalytic cracker, wherein the hydrocracker provides unconverted gas oil to the fluid catalytic cracker based on the values to improve an overall value of the hydrocracker and fluid catalytic cracker operation.

301. The method of claim 300, wherein generating, by the machine learning model, one or more adjustments to operational parameters of the hydrocracker and fluid catalytic cracker is further based on an objective function including a total value of products less a total cost of feedstocks less operating costs subject to operational limits and engineer specified-constraints.

302. The method of any of claims 300 to 301, further comprising implementing the one or more adjustments to parameters of the hydrocracker and fluid catalytic cracker.

303. The method of any of claims 300 to 302, wherein the machine learning model is trained on the data indicative of one or more feedstocks to a hydrotreater, operational parameters of the hydrotreater, or products of the hydrotreater, wherein the business data is related to the one or more feedstocks to the hydrotreater, operational parameters of the hydrotreater, or products of the hydrotreater, wherein generating, by the machine learning model, values includes generating values for the one or more feedstocks to a hydrotreater, operational parameters of the hydrotreater, or products of the hydrotreater, wherein generating, by the machine learning model, one or more adjustments includes further generating one or more adjustments to operational parameters of the hydrotreater based on the values to improve an overall value of the hydrocracker, fluid catalytic cracker, and hydrotreater operation.

304. The method of claim 303, wherein the business data includes costs of operating the hydrotreater and a current pricing of hydrogen gas.

305. The method of any of claims 300 to 304, wherein the business data includes current market pricing of the products of the hydrocracker and fluid catalytic cracker.

306. The method of any of claims 300 to 305, wherein the business data includes costs of operating the hydrocracker and fluid catalytic cracker.

307. The method of any of claims 300 to 303, wherein the business data includes data indicative of a forecast, wherein generating, by the machine learning model, one or more adjustments is constrained by the forecast of demand for the products of the hydrocracker and fluid catalytic cracker.

308. The method of any of claims 300 to 307, wherein the method of claim 242 is used with each update to the business data.

309. The method of any of claims 300 to 308, further comprising publishing the one or more adjustments to a user for approval to implement the one or more adjustments.

310. The method of any of claims 307 to 309, wherein generating, by the machine learning model, the one or more adjustments is further based on known unique constraints of specific equipment.

311. The method of any of claims 300 to 310, wherein the business data is received from a local enhancement portion of a trained machine learning model.

312. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 300 to 310.

313. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 300 to 310.

314. A method compri sing : receiving operational parameters of a fluid catalytic cracker;generating a prediction of a deactivation state of a catalyst of the fluid catalytic cracker based on the operational parameters of the fluid catalytic cracker; generating an adjustment in an amount of unconverted gas oil from a hydrocracker to the fluid catalytic cracker; generating a prediction of an effect on the deactivation state of a catalyst of the fluid catalytic cracker based on the generated adjustment in the amount of unconverted gas oil from the hydrocracker to the fluid catalytic cracker; and generating an adjustment to an operational parameter of the hydrocracker to produce the amount of unconverted gas oil from the hydrocracker.

315. The method of claim 314, further comprising implementing the adjustment to the operational parameter of the hydrocracker to produce the amount of unconverted gas oil from the hydrocracker.

316. The method of any of claims 314 to 315, wherein a machine learning model performs the method of: receiving operational parameters of a fluid catalytic cracker; generating a prediction of a deactivation state of a catalyst of the fluid catalytic cracker based on the operational parameters of the fluid catalytic cracker; generating an adjustment in an amount of unconverted gas oil from a hydrocracker to the fluid catalytic cracker; generating a prediction of an effect on the deactivation state of a catalyst of the fluid catalytic cracker based on the generated adjustment in the amount of unconverted gas oil from the hydrocracker to the fluid catalytic cracker; and generating an adjustment to an operational parameter of the hydrocracker to produce the amount of unconverted gas oil from the hydrocracker, wherein the machine learning model has been trained with historical data indicative of properties and composition of a feedstock of the fluid catalytic cracker, operational parameters of the fluid catalytic cracker, properties and composition of products of the fluid catalytic cracker, properties and composition of a feedstock of the hydrocracker, operational parameters of the hydrocracker, and properties and composition of products of the hydrocracker.

317. The method of claim 316, further comprising receiving, by the machine learning model, business data indicative of market pricing of the products of the hydrocracker andthe fluid catalytic cracker, costs of feedstocks for the hydrocracker and fluid catalytic cracker, and costs of operation of the hydrocracker and fluid catalytic cracker, wherein the machine learning model is programmed to maximize an overall value of the hydrocracker and fluid catalytic cracker.

318. The method of any of claims 314 to 317, wherein generating the adjustment to the operational parameter of the hydrocracker to produce the amount of unconverted gas oil from the hydrocracker is constrained by a priority to produce a product of the hydrocracker having a target product property.

319. The method of claim 318, wherein the target product property is sulfur content in the product.

320. The method of claim 319, wherein the target product property is measured by a sulfur analyzer.

321. The method of claims 318 or 319, wherein the target product property is measured from a sample to generate lab data, wherein a machine learning model correlates feedstock data and an operating temperature of a hydrotreater with the lab data to generate a predicted target property.

322. The method of claim 318, wherein the target product property is a desired aromatic saturation rate of the product from the hydrocracker.

323. The method of claim 318, wherein the target product property is a desired octane- barrels of the product from the hydrocracker.

324. The method of claim 318, wherein the target product property is a desired octane number of the product from the hydrocracker.

325. The method of claim 318, wherein the target product property is a desired conversion rate of the product from the hydrocracker.

326. The method of claim 318, wherein the target product property is a desired cetane number of the product from the hydrocracker.

327. The method of any of claims 314 to 326, wherein generating the adjustment to the operational parameter of the hydrocracker to produce the amount of unconverted gas oil from the hydrocracker is constrained by a priority to produce a product of the fluid catalytic cracker having a target property.

328. The method of claim 327, wherein the target product property is a desired octane- barrels of the product from the fluid catalytic cracker.

329. The method of claim 327, wherein the target product property is a desired octane number of the product from the fluid catalytic cracker.

330. The method of claim 327, wherein the target product property is a desired cetane number of the product from the fluid catalytic cracker.

331. The method of claim 316, wherein the machine learning model is programmed to optimize a deactivation rate of the catalyst of the fluid catalytic cracker.

332. The method of any of claims 314 to 331, wherein the effect is a deceleration of a deactivation rate of the catalyst of the fluid catalytic cracker.

333. The method of any of claims 314 to 331, wherein the effect is an acceleration of a deactivation rate of the catalyst of the fluid catalytic cracker.

334. A memory including instructions that causes a processor to perform the instructions including the method of any of claims 314 to 333.

335. A computing system including a processor and a memory including instructions that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 314 to 333.

Images