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

Abstract

Methods and systems for improving operation of a gasoline desulfurization system may include receiving gasoline blending pool data, the gasoline blending pool data indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each current component of a gasoline blending pool. A machine learning model generates a light naphtha sulfur target for a light naphtha produced from a selective hydrogenation reactor of a gasoline desulfurization system based on the gasoline blending pool data and generates a heavy naphtha sulfur target for a desulfurized heavy naphtha produced from a hydrodesulfurization reactor of the gasoline desulfurization system based on the gasoline blending pool data. The machine learning model may be trained on historical data of the gasoline blending pool and historical data of operating parameters, resulting product attributes, and composition from the gasoline desulfurization system, including sulfur concentration and octane number.

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 compositions 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 gasoline blending pool data, the gasoline blending pool data indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each current component of a gasoline blending pool. Then a machine learning model generates a light naphtha sulfur target for a light naphtha produced from a selective hydrogenation reactor of a gasoline desulfurization system based on the gasoline blending pool data and generates a heavy naphtha sulfur target for a desulfurized heavy naphtha produced from a hydrodesulfurization reactor of the gasoline desulfurization system based on the gasoline blending pool data. The machine learning model has been trained on historical data of the gasoline blending pool, historical data of operating parameters and resulting product attributes and composition from the gasoline desulfurization system, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

[0008] Another embodiment includes a method including receiving feedstock data from a sensor package, the feedstock data being indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed a gasoline desulfurization system. A machine learning model generates a recommended change to an operating parameter of a selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target. The method further includes sending instructions to a controller to implement the recommended change to the operating parameter of the selective hydrogenation reactor. The operating parameter of the selective hydrogenation reactor is then changed to implement the recommended change.

[0009] In yet another embodiment, a method includes receiving a heavy naphtha product data from a sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenationreactor. A machine learning model generates a recommended change to an operating parameter of a hydrodesulfurization reactor of a gasoline desulfurization system to produce a desulfurized heavy naphtha product having a sulfur concentration closer to a heavy naphtha sulfur target based on the heavy naphtha product data and the heavy naphtha sulfur target. The method further includes sending instructions to a controller to implement the recommended change to the operating parameter of the selective hydrogenation reactor, wherein the operating parameter of the selective hydrogenation reactor is changed to implement the recommended change.

[0010] Another embodiment includes a method including receiving feedstock data from a sensor package, the data being indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed into a gasoline desulfurization system. A machine learning model predicts when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system, and generates a change to an operating parameter of the selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target. The change includes a changing rate over time of the operating parameter of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system.

[0011] In another embodiment, a method includes receiving a heavy naphtha product data from a sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenation reactor. A machine learning model predicts when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into a hydrodesulfurization reactor of a gasoline desulfurization system, and generates a change to an operating parameter of the hydrodesulfurization reactor based on the heavy naphtha product data and a heavy naphtha sulfur target; wherein the change includes a changing rate over time of the operating parameter of the hydrodesulfurization reactor based on when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

[0012] In yet another embodiment, a method includes receiving business data, the business data including feedstock costs, and a current cost of operating each process of a gasoline desulfurization system. The method further includes receiving feedstock data, the data being indicative of one or more of a composition of a feedstock or a feed rate of the feedstock being fed into a gasoline desulfurization system and receiving gasoline blendingpool data, the gasoline blending pool data indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each component of the gasoline blending pool. The machine learning model generates a plurality of simulations based on the feedstock data, the business data, and gasoline blending pool data operating at different operational parameters of the gasoline desulfurization system and generates a cost, octane, and estimated sulfur concentration of a light naphtha produced from splitter downstream from a selective hydrogenation reactor for each simulation. The machine learning model generates a cost, octane, and estimated sulfur concentration of a desulfurized heavy naphtha produced from the splitter downstream from a hydrodesulfurization reactor for each simulation. The machine learning model has been trained on historical data of the gasoline blending pool, historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

[0013] Another embodiment of the disclosure is directed to a system for enhancing fluid production for a gasoline desulfurization operation. The system may include a gasoline desulfurization unit to receive a feed and produce one or more fluids. The system may include a plurality of sensors to measure a parameter associated with the gasoline desulfurization unit and each positioned at one of (a) proximate the gasoline desulfurization unit or (b) within the gasoline desulfurization unit. The system may include a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the gasoline desulfurization unit and to control aspects of fluid flowing to or from the gasoline desulfurization unit. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the gasoline desulfurization unit. 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 gasoline desulfurization controller in signal communication with the gasoline desulfurization 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 gasoline desulfurization controller may be configured to (1) determine an output including predicted properties of the feedstock and parameter settings of the plurality of refinery operation control devices and the gasoline desulfurization 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 correspondingtarget properties to the trained machine learning model, and (2) adjust one or more of the amount of feed or type of feed and parameters associated with the plurality of refinery operation control devices and gasoline desulfurization unit based on the output to enhance fluid production.

[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 diagrams 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 processes within a refinery.

[0021] FIG. 5A is a schematic diagram of the different products of the refinery and other sources that may be added to each of the blending pools.

[0022] FIG. 6 is a schematic diagram of a machine learning model that may be used with the various processes within a refinery.

[0023] FIG. 7 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. 8 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. 9 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. 10 is a schematic diagram of a gasoline desulfurization control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0027] FIG. 10A is a schematic diagram of an alternative embodiment of a gasoline desulfurization control system.

[0028] FIG. 11A and FIG. 11B are simplified diagrams of control systems to enhance to enhance fluid production at refinery, according to an embodiment of the disclosure.

[0029] FIG. 12 is a flow chart illustrating enhanced fluid production at a refinery, according to an embodiment of the disclosure.

[0030] FIG. 13 is a flow chart of a method, according to at least one embodiment of the present disclosure.

[0031] FIG. 14 is a flow chart of a method, according to at least one embodiment of the present disclosure.

[0032] FIG. 15 is a flow chart of another method, according to at least one embodiment of the present disclosure.

[0033] FIG. 16 is a flow chart of yet another method, according to at least one embodiment of the present disclosure.

[0034] FIG. 17 is a flow chart of a method, according to at least one embodiment of the present disclosure.

[0035] FIG. 18 is a flow chart of another method, according to at least one embodiment of the present disclosure.DETAILED DESCRIPTION

[0036] 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 theappended 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.

[0037] 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 a plurality of trained machine learning models that are trained to enhance fluid production of one or more refinery operations or sub-operations.

[0038] 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 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 ordistillation 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.

[0039] 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 may include 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.

[0040] 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 datapoints and / or parameters to train a machine learning model. Such data may include historical data set and / or a currently generated data set including an outcome. 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 suboperation 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, each outcome 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.

[0041] 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).

[0042] 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 / orvalidation. 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 remining 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 achieved that threshold, then the controller may output the trained machine learning model for further use.

[0043] 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 learning model 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.

[0044] 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 aselected 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.

[0045] 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 or substantially 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.

[0046] 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.

[0047] 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 or classifier. 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 a pump, 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.

[0048] 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.

[0049] 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 or classifier. 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.

[0050] 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.

[0051] 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) achieve 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.

[0052] Thus, rather than attempting to adjust operations at a significant delay, a refinery’s operations may be adjusted in-real time or at time intervals shorter than in typical optimization operations (such typical optimization operations including operations by 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.

[0053] 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 or classifiers 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.

[0054] 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, therefinery controller 101 and / or the plurality of operation controllers 102 may include, for example, a trained machine learning model or classifier, 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 102 to 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.

[0055] 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.

[0056] In an embodiment, the operation controller 102 may include a local enhancement circuitry 184. The local enhancement circuitry 184 may include a trained machine learningmodel 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 achieve 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.

[0057] 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-infrared spectroscopic 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.

[0058] 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 toperform the various operations described herein. Devices such as smartphones, laptop computers, and tablet computers are generally collectively referred to as mobile devices.

[0059] 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.

[0060] 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.

[0061] 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 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.

[0062] 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 ormeter 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 after multiple 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 result in 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 anotherembodiment, 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.

[0063] 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 optimization algorithm (which may also be referred to as the local enhancement module) to generate or determine targets.

[0064] 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 / oroperation 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.

[0065] 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 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 140, 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).

[0066] In an embodiment, the refinery controller 101 and / or operation controllers 102 may obtain data in 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.

[0067] 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.

[0068] 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. 1A-1B and below in connection with FIGS. 3-18.

[0069] 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 processingcircuitry 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.

[0070] 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, the software instructions may specifically configure the processing circuitry 202 to perform the algorithms and / or operations described herein when the software instructions are executed.

[0071] 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.

[0072] 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.

[0073] 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 circuitry 208 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.

[0074] 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 achieving the target product’s properties).

[0075] In another, the modeling circuitry 208 may train and / or include a plurality of machine learning models or classifiers. Each of the plurality of machine learning models or classifiers may correspond to a selected refinery operation and / or sub-operation.

[0076] 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. 3-18. 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).

[0077] 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, as described above in connection with FIGS. 1A-1B and below in connection with FIGS. 3- 18. 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.

[0078] 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. 3-18. The equipment and device adjustment circuitry 212 may further utilize communications circuitry 206 to transmit signals to adjust equipment and / or devices utilized.

[0079] 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, orcommunications 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.

[0080] 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.

[0081] 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.

[0082] 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 3-18) may be a computer program product comprising software instructionsstored 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.

[0083] FIG. 3 is 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 and receive 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.

[0084] 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. Forexample, 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.

[0085] 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 target properties 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).

[0086] 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.

[0087] 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.

[0088] 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 plots and / or values, gain directions and / or magnitude, and / or gain distributions, among other factors.

[0089] 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.

[0090] As noted, data may be obtained in 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.

[0091] 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 or classifiers. Each trained machine learning model may be trained to recognize a specificor 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.

[0092] 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.

[0093] 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 orparameter. Another model may be trained to utilize the outputs of each of those plurality of models, in addition to data.

[0094] 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.

[0095] 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 achieve 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 the 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.

[0096] 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.

[0097] FIG. 5 is a schematic diagram of a refinery 500. Crude oil 502 is initially processed by an atmospheric distillation tower 504 into its constituent parts. Some of the lightest parts from the atmospheric distillation tower 504 is a gas 506 that is sent to gas processing 508.

[0098] Optionally, the atmospheric distillation tower 504 may include or be connected to additional separation equipment 551, such as a splitter, a separator, a debutanizer, a deisobutanizer, a deisopentanizer, or other distillation column designed for a specific cut off. A deisopentanizer is a distillation column that may be used to further split lighter components, such as isopentane and lighter molecules from the light naphtha 536.

[0099] Gas processing 508 separates sour gas 509 from the gas 506 and other gas 507 resulting from different processes throughout the refinery into different gases and diverts the sour gas 509 to amine treating 516. The remainder of the gas 506 and other gas 507 is passed to the mercaptan treater 510 that separates mercaptans from the fuel gas. The fuel gas may be finished as liquid petroleum gas 512 (“LPG”). The resulting butane 514 may be kept as a finished product, sent to the gasoline blending pool 543, or sent to the C4 isomerization 591 to be processed into isobutane 593 as a finished product or further sent to an alkylation unit 594.

[0100] Amine treating 516 separates the hydrogen sulfide gas 520 from the refinery fuel 518. The hydrogen sulfide gas 520 and the hydrogen sulfide gas collected from processes throughout the refinery 500 is sent to the sulfur plant 522. The sulfur plant 522 converts the hydrogen sulfide gas 520 into sulfur 530 as a finished product.

[0101] Additionally, sour water 532 collected from processes throughout the refinery 500 is sent to the sour water steam stripper 528. The sour water steam stripper 528 uses steam 534 to remove hydrogen sulfide gas 526 from the sour water 532. The hydrogen sulfide gas 526 is also sent to the sulfur plant 522.

[0102] The atmospheric distillation tower 504 separates light naphtha 536 from the crude oil 502. The light naphtha 536 is sent to a hydrotreater 538 that removes sulfur from the light naphtha 536. The desulfurized light naphtha is then sent to isomerization 540 to be processed into isomerate 542. Isomerate 542 are isomers of the light naphtha that have higher octane values. The isomerate 542 may then be sent to the gasoline blending pool 543.

[0103] The atmospheric distillation tower 504 separates heavy naphtha 544 from the crude oil 502. Optionally, the heavy naphtha 544 may be sent to a splitter 545. The splitter 545 may include one or more splitter columns that separate heavier C7+ naphtha molecules from lighter components of the heavy naphtha 544. The heavier C7+ naphtha may be sentto the hydrotreater 554 to be desulfurized and produced as jet fuel 558. The remaining lighter components of the heavy naphtha may be sent to a hydrotreater 546 to remove sulfur from the heavy naphtha 544 prior to being sent to the catalytic reformer 548 to be converted into reformate 550.

[0104] Alternatively, the heavy naphtha 544 may be sent directly to the hydrotreater 546 to remove sulfur from the heavy naphtha 544 prior to being sent to the catalytic reformer 548. From the hydrotreater 546, the desulfurized heavy naphtha is sent to a catalytic reformer 548 to be processed into reformate 550. Reformate 550 includes high-octane branched and cyclic hydrocarbons, such as benzene, toluene, xylene, and ethylbenzene. The reformate 550 may then be sent to the gasoline blending pool 543.

[0105] Another product from the atmospheric distillation tower 504 is jet fuel 552. The jet fuel 552 may include kerosene and other equivalent hydrocarbons. The jet fuel 552 may be sent from the atmospheric distillation tower 504 to a hydrotreater 554 to remove contaminants from the jet fuel 552, such as sulfur and mercaptans. Once desulfurized, the finished jet fuel 558 may be sold or sent to a jet fuel blending pool to be blended with other products and additives and then made available for sale and distribution. The jet fuel 552 may also be separated into kerosene.

[0106] Further, diesel 560 is another product separated from crude oil 502 by the atmospheric distillation tower 504. The diesel 560 is sent from the atmospheric distillation tower 504 to a hydrotreater 562 to remove sulfur. The desulfurized diesel 564 may then be sold or sent to a diesel blending pool.

[0107] The atmospheric distillation tower 504 also separates atmospheric gas oil 566 and atmospheric bottoms 568 from the crude oil 502. The atmospheric bottoms 568 may be sent to vacuum distillation 570 where the atmospheric bottoms 568 may be further separated into light vacuum gas oil 572 (“LVGO”), medium vacuum gas oil 599 (“MVGO), heavy vacuum gas oil 584 (“HVGO”), and vacuum residuum 521. Vacuum distillation 570 may also be configured to separate the atmospheric bottoms 568 into more or less components.

[0108] The atmospheric gas oil 566, the LVGO 572, the MVGO 599, and deasphalted oil 501 from solvent deasphalting 598 (“SDA”) may be sent to a hydrotreater 574 to remove sulfur, then fed into a fluid catalytic cracker 576. The fluid catalytic cracker 576 processes the atmospheric gas oil 566 and the light vacuum gas oil 572 into naphtha 578, jet fuel 581, diesel 582, fuel oil 583, and butenes and pentenes 592.

[0109] The butenes and pentenes 592 from the fluid catalytic cracker 576, as well as the isobutane 593 from the C4 isomerization 591, may be sent to an alkylation unit 594 wherethe feedstocks are processed into alkylate 596. The alkylate 596 may then be sold or sent to the gasoline blending pool 543. Alkylate 596 has a high octane rating and low RVP.

[0110] The naphtha 578 may be sent to a hydrotreater 580 to further remove sulfur from the naphtha 578. The desulfurized naphtha may then be sold or sent to the gasoline blending pool 543. The jet fuel 581 may also be passed through a hydrotreater 585 to remove sulfur and other contaminants and then may be sent to a jet fuel blending pool. The diesel 582 may also be passed through a hydrotreater 587 to remove sulfur and other contaminants and then sent to a diesel blending pool and may be sold or stored for later sale and distribution. Lastly, the fuel oil 583 may also be passed through a hydrotreater 585 to remove sulfur and other contaminants and then sent to a 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.

[0111] The MVGO 599, HVGO 584, and the deasphalted oil 501 may be sent to the hydrocracker 586 to be processed into gasoline 588, diesel 590, and jet fuel 595. The hydrocracker 586 may also produce smaller hydrocarbons that may be sent to gas processing 508. The hydrocracker 586 process integrates desulfurization and other contaminant removal, so further hydrotreating is not necessary for its products. The gasoline 588 may be sold or sent to the gasoline blending pool 543 shown in FIG. 5A. The diesel 590 may be sold or sent to a diesel blending pool. The jet fuel 595 may be sold or sent to a jet fuel blending pool.

[0112] The vacuum residuum 521 may be sent to solvent deasphalting 598 (“SDA”) or used directly in asphalt 519. In some applications, the vacuum residuum 521 may be sent to an asphalt blending pool. The SDA 598 may be used to extract lighter components from the vacuum residuum 521 using solvents such as propane, butane, pentane, or a combination of these hydrocarbons to extract deasphalted oil 501 from the vacuum residuum 521. As discussed above, the deasphalted oil 501 may be sent to the hydrocracker 586 or the fluid catalytic cracker 576 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 or pitch 579. The pitch 579 may also be used in the asphalt blending pool.

[0113] The vacuum residuum 521 and pitch 579 may also be sent to the coker 597 to be processed into naphtha 503 and processed through a gasoline desulfurization unit 556 (“GDU”). The GDU 556 may include one or more processes useful to remove contaminants from the naphtha 503 in addition to gasoline desulfurization, including hydrotreatingprocesses, catalysts, particulate catches, clay treaters, salt driers, and mercaptan treaters, separators, and steam strippers. The naphtha 503 may then be sent to the gasoline blending pool 543. The coker 597 also processes the vacuum residuum 521 and pitch 579 into jet fuel 511, diesel 513, fuel oil 515, and petroleum coke 517. The jet fuel 511 may be passed through a hydrotreater 523 and then sent to the jet fuel blending pool. The diesel 513 may be sent to a hydrotreater 525 to remove sulfur and then sent to the diesel blending pool. The fuel oil 515 may also be processed through hydrotreater 527 and then sent to a fuel oil blending pool.

[0114] The hydrotreaters 538, 546, 554, 562, 574, 580, 585, 587, 589, 523, 525, and 527 refer broadly to desulfurization and contaminant removal processes generally, including hydrotreating processes, clay treaters, salt driers, mercaptan treaters, gasoline desulfurization units, separators, steam strippers, filtration systems, and particulate catches.

[0115] 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 504 and vacuum distillation 570 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 gasoline desulfurization, isomerization, reformation, and alkylation processes not shown in FIG. 5 to process the products and components to meet regulatory requirements and customer specifications.

[0116] FIG. 5 A is a schematic diagram of hydrocarbon refinery products 505 produced by the refinery 500 and their movement into blending pools 524, including a gasoline blending pool 543. Each blending pool 524 stores products from various refinery product streams and products from other sources in storage tanks 575 and may use these products to blend these products into different formulations that meet various regulatory requirements, industry standards, and customer specifications.

[0117] As shown, the butane 514, the naphtha 503, naphtha 578, the isomerate 542, the reformate 550, the alkylate 596, and the gasoline 588 may all be sent to the gasoline blending pool 543. Additionally, materials may be obtained from other sources 529 including other refineries, third parties, or off the open market. Gasoline specifications may include research octane number (RON), motor octane number (MON), volatility or Reidvapor pressure (“RVP”), sulfur content, benzene content, aromatic content, olefin content, drivability index, distillation curve, vapor-liquid ratio, stability, and corrosion resistance.

[0118] For example, butane 531, naphtha 533, isomerate 535, reformate 537, alkylate 539, and gasoline 541 may be obtained from other sources 529 and added to the gasoline blending pool 543. Butane 531 may be blended in the gasoline blending pool 543 for use during seasons of cold temperatures because butane 531 increases the Reid Vapor Pressure (“RVP”) helping the gasoline blend vaporize and ignite in cold gasoline engines. Coker naphtha 503 may be included in a blended gasoline after hydrotreating but may include a higher proportion of olefins. Naphtha 578 from the fluid catalytic cracker 576, also known as FCC gasoline, may be a large contributor to the gasoline blending pool 543 and may also contain olefins and aromatic molecules. Thus, naphtha 503 and naphtha 578 may be passed through a catalytic reformer 548 to saturate the olefins and aromatics before being sent to the gasoline blending pool 543.

[0119] Isomerate 542 increases the octane rating of a blended gasoline without increasing aromatic content and is useful in meeting regulatory limits on benzene. Reformate 550 also increases the octane rating of a blended gasoline, but contains benzene. Alkylate 596 is a useful blending component of gasoline that increases octane rating and contains few, if any, aromatics. Hydrocracked gasoline 588, also known as hydrocracked naphtha, is low in sulfur and may be used to reduce sulfur content in a blended gasoline.

[0120] Additionally, natural gasoline 547, oxygenates 549, such as MTBE, ETBE, or ethanol may be sent to the gasoline blending pool 543. Natural gasoline 547 may be derived from natural gas liquids during gas processing. Natural gasoline 547 typically has a low octane rating and may be used as a low cost component in blended gasolines. Oxygenates 549 may help increase octane rating in blended gasolines and may help reduce carbon monoxide emissions.

[0121] The different components of the gasoline blending pool 543 are mixed to create the blended product attributes ordered by customers and that comply with regulatory requirements. In many applications, higher-octane components, like reformate and alkylate, balance lower-octane product streams. Further, aromatics, benzene, and olefins are often limited to meet regulatory standards.

[0122] In general, inputs into the gasoline blending pool 543 may include reformate, isomerate, alkylate, naphtha from different sources, butane, oxygenates, and other additives. Each formulation of blended gasoline from the gasoline blending pool 543 has attributes that may be measured and specified through testing including Reid vapor pressure(“RVP”) at different temperatures, olefin content, oxygenate content, benzene content, deposit formation potential, knock resistance index, long-term storage stability, aromatics content, vapor liquid ratio, vapor lock index, sulfur content, mercaptan sulfur, gum content, oxidation stability, corrosion, sulfur compliance risk score, fuel economy impact, driveability index, blend volatility control, and density. Each formulation of blended gasoline from the gasoline blending pool may be selected for production based on the current market price for similar formulations, forecasted demand, predicted profit per batch, regulatory constraints, and current inventory levels in the storage tanks of the gasoline blending pool 543.

[0123] All products produced via refining processes including but not limited to the refinery products 505 and products from other sources 529, including purchased products may be added to the blending pools 524 and blended into saleable products in compliance with regulatory and customer specified constraints. Blended formulations are often sold into markets with different specifications requiring strict product segregation, quality assurance, quality controls. Specifications for formulations may be based on location with regulations varying significantly between jurisdictions. Segregation of components within the blending pools 524 may be required by standards and regulatory requirements. For example, products blended with renewable products may occur at a terminal to avoid contamination of storage tanks within the blending pool.

[0124] A complication in developing blended formulations, is that some attributes are not proportional while some are proportional to the amount contributed. For example, density, sulfur content, and percent of aromatics, benzene, and olefins are generally proportional with the amount blended. In contrast, flash point, flash point stability, flash point variance, freeze point, viscosity, cetane number, and octane number are not perfectly linear due to interactions between components. In some applications, the presence of lower flash point components may significantly impact the resulting flash point of a blended formulation. Smoke point can be heavily influenced by high-smoke components. Reid Vapor pressure does not blend linearly because of vapor-liquid equilibrium effects. Pour point is not proportional because some components have disproportionate effects.

[0125] The gasoline blending pool 543 includes a plurality of storage tanks 575. Each storage tank 575 may contain one or more of the refinery products 505 or products from other sources 529. Each storage tank 575 may include a sensor package 577. Depending on the substance stored within, the sensor package 577 may include level sensors, such as radar level sensors, ultrasonic level sensors, float & tape gauges, magnetostrictive sensors,and capacitive sensors. The sensor package 577 may also include temperature sensors, such as resistance temperature detectors, thermocouples, and infrared sensors, and pressure sensors, such as hydrostatic pressure sensors and differential pressure sensors. The sensor package 577 may include gas and vapor sensors, such as volatile organic compound (“VOC”) sensors, hydrocarbon gas detectors, and oxygen sensors for monitoring leak detection and inert gas blanketing. The sensor package 577 may include flow rate sensors, such as coriolis flow meters, ultrasonic flow meters, and turbine flow meters for metering flow in and out of the storage tank 575. The sensor package 577 may also include density and composition analyzers, such as refractometers, spectroscopy, and gas chromatographs that may be used to monitor the condition of the materials within the storage tank 575. Further, the sensor package 577 may include safety and leak detection sensors, such as acoustic sensors, fiber optic sensors, and infrared gas leak detectors.

[0126] Data from the sensor package 577 may be used by a machine learning model to request certain refinery products to be produced at faster or slower rates or recommend that products and feedstocks from other sources, be ordered to fill a storage tank 575 in anticipation of blending a particular formulation for a client. The machine learning model may also make recommendations based on the quantities held in each storage tank 575 of particular products. In other words, the machine learning model may adjust a process producing a product based on the inventory of the product in one of the blending pools 524. The adjustment may be to change operating parameters, including one or more of temperature, pressure, and flowrates of a process, based on the inventory of a product in one of the blending pools 524. Alternatively, the machine learning model 800 of FIG. 6 may adjust a process to produce more of a product or products with a desired component distribution based on one or more of forecast information, demand predicted by the machine learning model 800 based on the business data 810 including regulatory changes and a predicted market price for the product during a future time period. The machine learning model 800 may create multiple simulations (e.g., published simulations, displayed simulations, transmitted simulations, published projections, displayed projections, and transmitted projections) based on the historical data 802 and the business data 810 that model potential prices for refinery products, then make a recommendation to adjust a process parameter based on a selected simulation. The simulation may be published on the engineering gateway 812 for approval or permission by a user. The machine learning model may also make recommendations based on the business information available to it including forecasts, pending customer orders, and general demand in the market.

[0127] A machine learning model 800, such as shown in FIG. 6, may be used to optimize each blended formulation blended from the gasoline blending pool 543. In some embodiments, the machine learning model obtains attribute data for each component in the blending pool and other components that may be sourced from other refineries, third- parties, and generally available in the market that be used to create a desired blended formulation that meets a customer’s specifications and regulatory requirements. The machine learning model may create multiple combinations that may be compared against the customer’s specifications and regulatory requirements. Each combination that meets the customer’s specifications and regulatory requirements are retained in a potential formulation list, and those that do not meet the customer’s specifications and regulatory requirements are discarded. The machine learning model may then access business data to obtain pricing information for each component, calculate a base formulation cost for each combination in the potential formulation list, and order the list from lowest cost to the highest cost. Alternatively, the machine learning model may use the component attributes to extrapolate the anticipated attributes of each blended formulation and may order the potential formulation list according to an anticipated attribute. For example, a potential formulation list for a diesel fuel may be ordered based on an anticipated cetane number, sulfur content, cold flow properties, or energy content, and then secondarily when multiple formulations have the same cetane number on another attribute. Similarly, a potential formulation list for gasoline may be ordered according to the gasoline specification attributes, such as octane rating.

[0128] FIG. 6 is a schematic diagram of a machine learning model 800 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 generate 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.

[0129] The machine learning model 800 is connected to historical database 802. The historical database 802 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 802 may be a copy of historical databases that are maintained and updated by the machine learning model 800. The machine learning model 800 is also connected to a training database 803 that includes the process data and historical information that is used for training of the machine learning model 800.

[0130] As discussed in relation to other figures of this disclosure, the machine learning model 800 may be used to analyze and improve specific processes, process controllers, and process interactions. The machine learning model 800 may also be used as a refinery controller.

[0131] In this embodiment, the machine learning model 800 may receive feedstock sensor, sample, and process data 804 from a feedstock process controller 814 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 800. For example, the feedstock process may include one or more of atmospheric distillation, vacuum distillation, filtration, hydrotreater, mercaptan treater, splitter, stripper, and separator.

[0132] Further, the machine learning model 800 may receive targeted process sensor, sample, and process data 806 from a targeted process controller 816 or directly from the sensors, sampling and testing systems, and labs obtaining data from the targeted process. The machine learning model 800 may also receive subsequent process sensor, sample, and process data 808 from a targeted process controller 816 or directly from the sensors, sampling and testing systems, and labs obtaining data from the targeted process. The feedstock sensor, sample, and process data 804, the targeted process sensor, sample, and process data 806, and the subsequent process sensor, sample, and process data 808 may include operating pressure and temperature data for the components of each process and sub-process. The feedstock sensor, sample, and process data 804, the targeted process sensor, sample, and process data 806, and the subsequent process sensor, sample, and process data 808 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.

[0133] The machine learning model 800 may also access business data 810. Business data 810 may include demand planning information, forecasting information for various products, current pricing information, product distribution information, and information regarding the current and anticipated regulatory landscape. Business data 810 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 810 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 810 may also include the inventory levels of various products that may be stored in the storage tanks 575 of the various blending pools.

[0134] The machine learning model 800 may also access lab data 811 that is obtained from laboratory analysis of samples collected from processes within a refinery and products produced by those processes. Lab data 811 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.

[0135] The machine learning model 800 may also access weather data 813 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 process operating parameters, feedstocks, the resulting products from those processes, and the effect of weather on the storage of products within the blending pools 524.

[0136] The machine learning model 800 may publish to and receive instructions from an engineering gateway 812. As used herein, “publish” means one or more of to simulate, send, or display. The engineering gateway 812 may act as a user interface for engineers to review process and machine learning model 800 data, recommendations, warnings, and requests. A user may use the engineering gateway 812 to assist the machine learning model 800 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 812 may also be used to facilitate active learning by the machine learning model 800. A user may use the engineering gateway 812 to provide feedback to the machine learning model 800. Feedback may include user approvals, user corrections, and user instructions of analysis, predictions, interpolations, data, and otherrecommendations that is provided by the machine learning model 800 through the engineering gateway 812.

[0137] The machine learning model 800 may access, receive data from, and publish instructions to a feedstock process controller 814, a targeted process controller 816, and a subsequent process controller 818. In some embodiments, the machine learning model 800 may adjust an algorithm used by the targeted process controller 816 based on changes identified from the adjusted algorithm. Once an adjusted algorithm has been selected by the machine learning model 800, the machine learning model 800 requests approval to implement the adjusted algorithm through the engineering gateway 812. Upon receipt of approval, the machine learning model 800 saves a copy of the approval information in the historical data 802 and sends the approved adjusted algorithm to the targeted process controller 816.

[0138] The machine learning model 800 includes a data analysis module 822. The data analysis module 822 may be used by the machine learning model 800 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 800 may save the revised data set in either the training data 803 or the historical data 802 for retraining of the machine learning model 800, for analysis and adjustment of a targeted process, or for later use.

[0139] The machine learning model 800 includes an interpolation module 824. The interpolation module 824 may be used to analyze a data set and interpolate missing data from the data set. The interpolation module 824 may use the historical data 802 and current operational data from a process upstream from a targeted process to interpolate a composition of the feedstock that will be fed into the targeted process. For example, samples may be taken from the product of a process and the sample data recorded in connection with the operating parameters of the process. The interpolation module 824 may interpolate or infer the composition of the resulting product from the operating temperature of the process, the operating pressure of the process, or the feed rate of feedstock into the process. Consequently, based on the upstream data related to the processing of a feedstock, the machine learning model may interpolate, infer, or determine the composition of the feedstock that will be fed into a downstream process. Thus, feedstock composition data includes this upstream data related to a feedstock that may be used to determine the composition of the feedstock.

[0140] The interpolation module 824 may compare the historical data 802 with feedstock sensor, sample, and process data 804, the targeted process sensor, sample, and process data 806, and the subsequent process sensor, sample, and process data 808 to identify process changes. The interpolation module 824 may be able to identify and flag small deviations for further investigation. The interpolation module 824 works with the communication module 828 to publish the data regarding the flagged deviation for user review and guidance. For example, the interpolation module 824 may be used to identify miscalibrated sensors, misprocessed test data, or misprocessed samples that may be inaccurate. The machine learning model 800 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 812. The machine learning model 800 may use the interpolation module 824 to identify components that may be wearing out and catalysts that may be deactivated or poisoned. The interpolation module 824 may also communicate through the communication module 828 with the engineering gateway 812 to identify these potential process concerns to a user for further investigation.

[0141] The machine learning model 800 includes a prediction module 826. The prediction module 826 may analyze the feedstock sensor, sample, and process data 804, the targeted process sensor, sample, and process data 806, and the subsequent process sensor, sample, and process data 808 to predict changes in a process. For example, temperature excursions may be predicted in a process and the prediction module 826 may communicate through the communication module 828 with the engineering gateway 812 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 812 or provide different instructions for the machine learning model 800 to implement. In predicting changes in a process, prediction module 826 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 826 may provide a recommended maintenance window for the maintenance procedure to occur.

[0142] The machine learning model 800 includes a communication module 828. The communication module 828 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 828 may send or publish communications to the engineering gateway 812. The communication module 828may also communicate with the various process controllers and other models used by the refinery.

[0143] In some embodiments, the machine learning model 800 includes a confidence module 830. The confidence module 830 may review the analysis and recommendations of the different modules of the machine learning model 800 and assign a confidence level to the analysis and recommendations. For example, the confidence module 830 may produce SHAP values, or use LIME or anchors to analyze each algorithm that is adjusted or created by the machine learning model 800. The confidence module 830 may compare predicted or anticipated values from a simulation or model with actual data to generate a confidence level that may include a calculation of the standard deviation between the predicted or anticipated values and the actual measured data. The results may be published by the communication module 828 to the engineering gateway 812 to assist a user in reviewing each algorithm and the changes recommended by the machine learning model 800.

[0144] In some embodiments, the machine learning model 800 includes a regulatory module 832. The regulatory module 832 may access the business data 810 to assist in regulatory compliance. For example, the regulatory module 832 may flag a process that may be predicted by the prediction module 826 to move out of compliance (i.e., non- compliant). The regulatory module 832 may issue a warning through the communication module 828 to the engineering gateway 812. The regulatory module 832 may also identify windows of time when regulations are lessened and recommend changes to process parameters to reduce process costs. The regulatory module 832 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 832 may make recommendations for various process controllers to adjust their parameters to promote the production of blend components in-line with demand that may accompany regulatory changes.

[0145] In some embodiments, the machine learning model 800 includes a forecasting module 834. The forecasting module 834 may analyze business data 810 and recommend that various process controllers adjust their parameters to promote the production of components that may be in greater demand. The forecasting module 834 may also recommend maintenance windows for equipment producing products that may be in low demand during specific time frames. The forecasting module 834 may forecast pricing for various components based on the historical data 802 and the business data 810 and publish the forecasted pricing via the communication module 828 to the engineering gateway 812.

[0146] In some embodiments, the machine learning model 800 may include an authority module 836 and additional modules 838, 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 836 may track user instructions and approvals for various instructions and changes to be made by the machine learning model 800 to the various controllers throughout the refinery. In some applications, the authority module 836 may consider a recommendation from a module of the machine learning model 800 and automatically authorize the machine learning model 800 to issue an instruction to the targeted process controller 816 to make a change to an associated process. In other applications, the authority module 836 may direct the communication module 828 to request approval through the engineering gateway 812 before a recommendation may be implemented and instructions sent to the targeted process controller 816.

[0147] In some embodiments, the authority module 836 directs the communication module 828 to request approval through the engineering gateway 812 before a recommendation is implemented and / or instructions are sent to a targeted process controller. The recommendation may include one or more of changes or predicted changes in a nonlinear refinery process, such as the gasoline desulfurization processes where multiple variables, including pressure, temperature, and initial composition of the feedstock affect the composition of the product of the process. A nonlinear process is a process whose product attributes or composition are determined by a plurality of variables and thus may not be readily predictable. Additionally, a nonlinear product attribute is an attribute that does not blend on a linear basis. For example, when two jet fuels are blended together, many resulting attributes, such as flash point, of the blended jet fuel are not one-to-one by volume on a resulting attribute because of the effect of certain components in the blended jet fuels.

[0148] FIG. 7 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 hydrogenavailable 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.

[0149] 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.

[0150] FIG. 8 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. Each refinery 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 / orfeed. 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.

[0151] FIG. 9 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 controller and / 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 enhancementmodule 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 achieve 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.

[0152] 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.

[0153] 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 that specific 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.

[0154] 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 achieve 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 andsmaller percentages of gasoline and / or other fluids from other sources to achieve the octane and / or volatility limit.

[0155] FIG. 10 is a schematic diagram of a gasoline desulfurization control system 2400 to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. In embodiments, a section of the refinery corresponding to a gasoline desulfurization operation may include a gasoline desulfurization controller 2402. Similar to previously described controllers, the gasoline desulfurization controller 2402 may obtain data associated with the refinery equipment of the gasoline desulfurization operation, such as from a selective hydrogeneration reactor 2410, a splitter 2412, a hydrodesulfurization reactor 2414, a separator 2416, and a second splitter 2418. The gasoline desulfurization controller 2402 may also initiate capture of samples of fluids associated with the gasoline desulfurization operation. The sample collection and analysis assembly 2408 may then analyze the samples and produce properties and / or a spectra for each sample. The gasoline desulfurization controller 2402 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 2404 and / or predictive controls module 2406 to produce an output indicative of adjustment to parameters and / or feed. The gasoline desulfurization controller 2402 may then utilize the output to adjust various parameters and / or feed associated with the gasoline desulfurization operation via the local enhancement module 2404.

[0156] As shown, feedstocks 2420 of one or more of FCC naphtha, light straight run naphtha, natural gasoline, cocker naphtha, deisopentanizer overhead stream, and other sources of naphtha, such as a visbreaker or a pyrolysis process, as well as hydrogen gas in excess are fed into the selective hydrogenation reactor 2410. In one embodiment, the feed rate of feedstocks 2420 is increased by filling unused capacity with a portion of naphtha from the deisopentanizer 551 of Fig. 5. The naphtha from the deisopentanizer is composed of primarily C4 and C5 range material having an average road octane number or “RON” of about 87 and also has a low sulfur content. As the naphtha from the deisopentanizer is fed into the selective hydrogenation reactor 2410, a higher percentage of the resulting product 2422 from the selective hydrogenation reactor 2410 may be split into desulfurized light naphtha product 2426 by the splitter 2412. The higher percentage draw may offset the lower RON of the desulfurized light naphtha product 2426 with more C7 olefins included in the overhead splitter stream because of offset available from the low sulfur content of the deisopentanizer feedstock. In other words, the low sulfur content of the naphtha from the deisopentanizer may be offset by the sulfur included with the C7 olefins that is includedwith the desulfurized light naphtha product 2426, while still meeting the sulfur concentration target of the desulfurized light naphtha product 2426. Because a higher percentage is taken by the desulfurized light naphtha product 2426, the resulting heavy naphtha 2424 from the splitter bottoms of splitter 2412 is heavier and thus, has a higher road octane number.

[0157] A machine learning model or the gasoline desulfurization controller 2402 may adjust in real time the percentage of desulfurized light naphtha split from the resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer. Alternatively, the machine learning model or the gasoline desulfurization controller 2402 may predict an offset, publish the prediction to a user and the user may implement an adjustment to the percentage of desulfurized light naphtha split from the resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer. The adjustment may be based on an anticipated or previous average flow rate of naphtha from the deisopentanizer.

[0158] Feedstocks 2420 may be sampled and analyzed by a sensor package 2440 disposed to obtain a sample or measure the properties of the feedstocks 2420. The sensor package 2440 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, octane 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 2420 to be analyzed in a lab to generate lab data. Testing may include bromine titration to obtain a bromine number and maleic anhydride titration to obtain a maleic anhydride number or MAV number, both of which may be used to estimate the flow rate of hydrogen gas needed to be included in the feedstocks 2420. As used herein, the sensor package 2440 may include operating parameters and sensor information gathered from the operations processing the feedstock 2420. Consequently, the composition and attributes of the feedstock 2420 may be inferred from these operating parameters and sent to the sample collection and analysis assembly 2408. In general, the sensor package 2440 may measure the flow rate, sulfur content, bromine number, MAV number, distillation composition, octane number, temperature, and pressure of the naphtha feedstock, as wellas the flow rate, composition, and purity of hydrogen gas being mixed into the naphtha feedstock.

[0159] A challenge in operating the gasoline desulfurization system 2400 is the feed rate into the selective hydrogenation reactor 2410 and the composition of the feedstock 2420 changes frequently over time. Another challenge in operating the gasoline desulfurization system 2400 is that the more energy spent in removing sulfur from the feedstock 2420, the lower the octane of the products from the gasoline desulfurization system 2400. In other words, the closer the product can be to a sulfur target concentration, the higher the octane of the resulting products. Another challenge is that the resulting product streams from gasoline desulfurization system 2400 may be sent to different tanks for different purposes. For example, a sulfur target for the resulting light naphtha 2426 from the splitter 2412 may be different from a sulfur target for the resulting gasoline or heavy naphtha 2436. Consequently, different portions of the gasoline desulfurization system 2400 may be operated toward different, not necessarily aligned, targets.

[0160] In some embodiments, a machine learning model may determine a sulfur target for each product stream from the gasoline desulfurization system 2400 based on the current inventory and composition of the gasoline pool, as well as the predicted future inventory and composition of the gasoline pool. For example, a machine learning model may generate a light naphtha sulfur target for a light naphtha produced from a selective hydrogenation reactor of a gasoline desulfurization system based on gasoline blending pool data, and separately, for a heavy naphtha sulfur target for a desulfurized heavy naphtha produced from a hydrodesulfurization reactor of the gasoline desulfurization system based on the gasoline blending pool data. The gasoline blending pool data includes data that is indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each current component of a gasoline blending pool. The machine learning model may be trained on historical data of the gasoline blending pool, historical data of operating parameters, resulting product attributes, and composition from the gasoline desulfurization system. For example, the machine learning model may generate sulfur targets for the selective hydrogenation reactor’s light naphtha product and hydrodesulfurization reactor’s heavy naphtha product to maximize overall octane while achieving a combined sulfur target. The historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

[0161] The light naphtha sulfur target and the heavy sulfur target may also be based off of business data and market data accessed and received by the machine learning model. Forexample, if the business data indicates that the cost of operating the selective hydrogenation reactor to produce desulfurized light naphtha at a target is too expensive relative to market prices, the light sulfur target may be changed and other less expensive processes operated to make up the increase in the sulfur concentration of the desulfurized light naphtha with other components of the gasoline blending pool.

[0162] For example, the machine learning model may recommend or alter the operating parameters of each process of the gasoline desulfurization system based on business data including the sulfur target for the market where the blended gasoline product will be sold. Sulfur concentration regulations may vary significantly between jurisdictions. In addition, some jurisdictions may offer deviations from a sulfur concentration regulation as a sulfur credit that may be purchased. Each sulfur credit permits a gasoline blend sold in that jurisdiction to exceed the sulfur concentration regulation by a set amount. This business data may be used by the machine learning model to adjust the sulfur concentration target for the desulfurized light naphtha product and the desulfurized heavy naphtha product produced by the gasoline desulfurization system.

[0163] In some embodiments, the machine learning model may use the costs of production and feedstocks to help alter or recommend changes to the operating parameters of each process of the gasoline desulfurization system. In application, the machine learning model may use an equation based on business data similar to the following to balance the market costs of different factors affecting what sulfur target to use for the desulfurized light naphtha product and the desulfurized heavy naphtha product produced by the gasoline desulfurization system. An equation may include:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)Gasoline Desulfurization System Value may also take into account the cost of sulfur credits in the following equation:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)The machine learning model may also determine sulfur targets based on an equation such as:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)The overall sulfur target may be set by regulatory, customer, or strategic needs as interpreted by a user.

[0164] “PGLN flow” is the rate at which desulfurized light naphtha product is produced. “PGLN RON” is the road octane number of the desulfurized light naphtha product. “$ / PGLN Octane-BBL Incentive” is the incentive, profit, or price of each octane barrel of desulfurized light naphtha product. “PGLN Sulfur” is the sulfur concentration in the desulfurized light naphtha product. “Sulfur Credit Cost” may be the actual cost of the sulfur credit, regulatory constraints, customer constraints, or a monetary value determined by the machine learning model or a user. When shipping the desulfurized light naphtha product to a jurisdiction where the sulfur limits do not apply, the part of the function (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost)) may be treated as zero. “PGHN flow” is the rate at which desulfurized heavy naphtha product is produced. “PGHN RON” is the road octane number of the desulfurized light naphtha product. “$ / PGHN Octane-BBL” is the incentive, profit, or price of each octane barrel of the desulfurized heavy naphtha product. “PGHN Sulfur” is the sulfur concentration in the desulfurized heavy naphtha product. “Makeup H2 Flow” is the rate at which hydrogen gas is fed into the gasoline desulfurization system. “$ / SCF Hydrogen Cost” is the cost per cubic foot of hydrogen gas. In some applications, the S / PGLN Octane-BBL incentive may be the same as $ / PGHN Octane-BBL.

[0165] The desulfurized light naphtha product and the desulfurized heavy naphtha product may be used to create premium gasoline blends and regular gasoline blends. In some applications, the desulfurized light naphtha product may have a higher octane number than the desulfurized heavy naphtha product. The desulfurized light naphtha product and the desulfurized heavy naphtha product may be sent to the same tank within the gasoline blending pool, while in other applications, the desulfurized light naphtha product and the desulfurized heavy naphtha product may be sent to different tanks within the gasoline blending pool. In some configurations, one or both of the desulfurized light naphtha productand the desulfurized heavy naphtha product may be sent to different jurisdictions with different sulfur concentration regulations.

[0166] In application, PGLN RON may be inferred from the feedstocks being fed into the selective hydrogenation reactor, while PGHN RON may be inferred from the operating parameters of the hydrodesulfurization reactor, especially the change in temperature across the first hydrodesulfurization reactor. However, there is a threshold where all of the available mercaptan has been treated and any temperature increase above that point is over treating and causes a drop in octane of the desulfurized heavy naphtha product. At this point, the change in temperature across the first hydrodesulfurization reactor, whether or not more hydrodesulfurization reactors or finishing reactors are used, no longer correlates with the octane number of the desulfurized heavy naphtha product. Therefore, a machine learning model trained on the historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of the splitter after the selective hydrogenation reactor, the feedstock attributes, and the attributes and properties of the resulting desulfurized heavy naphtha product, may be used to control the first hydrodesulfurization reactor to minimize overtreatment by the hydrodesulfurization reactor. A difficulty in this control process is the long time constant in changing reactor temperature and measuring the resulting octane in the resulting desulfurized heavy naphtha product. Consequently, the predictive analysis of the machine learning model may be particularly valuable in minimizing overtreatment.

[0167] Alternatively, the machine learning model may receive a gasoline blending pool target sulfur concentration based on regulatory or customer requirements. Then, based on the current and predicted product streams from isomerization, alkylation, catalytic reforming, gasoline desulfurization, hydrotreating, and hydrocracking and their respective octanes and sulfur concentrations, the machine learning model may generate a sulfur concentration target for each of the product streams of the gasoline desulfurization system 2400. Once a target is determined, the machine learning model through the gasoline desulfurization controller 2402 determines the operating parameters for each process of the gasoline desulfurization system 2400 based on the target and the composition and feed rate of the feedstock 2420.

[0168] Alternatively, the machine learning model may determine a sulfur target for each product stream from the gasoline desulfurization system 2400 based on the predicted use or destination of the products from the gasoline desulfurization system 2400. For example, a product stream produced during a production window may be predicted to be sold to ajurisdiction with regulations permitting a higher sulfur concentration in gasoline. Consequently, a high sulfur concentration target may be selected, resulting in a higher octane number of the product stream. This may allow the product stream to be blended with lower cost lower octane products.

[0169] In order to better control to a target, the machine learning model and the gasoline desulfurization controller 2402 may use the data from the sensor package 2440 to change the operating parameters of the selective hydrogenation reactor 2410 in real time based on the predicted composition and feed rate of the feedstocks 2420. The feedstocks 2420 may include streams from a coker, a fluid catalytic cracker, and an atmospheric distillation tower that may be mixed inline and fed into the selective hydrogenation reactor 2410. Consequently, the composition of the feedstock 2420 may be constantly changing.

[0170] In some embodiments, the machine learning model and the gasoline desulfurization controller 2402 may use the data from the sensor package 2440 to generate a predicted composition of feedstock 2420, a predicted feed rate, and a predicted beginning time for the predicted composition of the feedstock 2420 to be fed into the selective hydrogenation reactor 2410. These predictions may then be used to determine the operating parameters to improve the operation of the selective hydrogenation reactor 2410 toward an operational goal. The predictions may be published to a user interface for approval or may be implemented by the operational goal may be to produce a product at or below a desired sulfur content and to minimize the octane loss from the resulting product. Alternatively, the machine learning model may control the operation of gasoline desulfurization system 2400 to increase profitability while reducing the operating costs of the gasoline desulfurization system 2400. The machine learning model may also control the gasoline desulfurization system 2400 to increase the service interval of the catalysts used in the gasoline desulfurization system 2400. The machine learning model may also control the gasoline desulfurization system 2400 to sharpen the cut in each of the splitters 2412, 2418. The machine learning model may also control the gasoline desulfurization system 2400 may generate targets for each process and control each process to a target based on the overall operational targets of the gasoline desulfurization system 2400.

[0171] In some embodiments, a bypass 2460 may be used when feed rates of feedstock 2420 exceed the operational limits of the gasoline desulfurization system 2400. The bypass 2460 may send excess feedstock 2420 to a holding tank, the hydrodesulfurization reactor 2414, or another process. The machine learning model may also control the upstreamprocesses to minimize usage of the bypass 2460 or to reduce large changes in the feed rate of the feedstock 2420 into the gasoline desulfurization system 2400.

[0172] The selective hydrogenation reactor 2410 converts light mercaptans and light sulfides into heavy mercaptans and heavy sulfides. Because each sulfur atom in a molecule is equivalent to about three carbon atoms in lowering the boiling temperature of the molecule, the heavy mercaptans are easier to separate from the hydrocarbons that do not contain sulfur in the splitter 2412. For example, methyl mercaptan (CH3SH) will boil like C4s and Thiophene (C4H4S) will boil like C7s. Consequently, the resulting product 2422 from the selective hydrogenation reactor 2410 is sent to a splitter 2412 to be split into a low sulfur light naphtha 2426.

[0173] The selective hydrogenation reactor 2410 may also convert di olefins through hydrogenation to olefins. A small amount of olefins may be saturated. A small amount of isomerization may also occur. Some of the resulting hydrogen sulfide gas may also combine with olefins to become heavier mercaptans through recombination.

[0174] The operating parameters of the selective hydrogenation reactor 2410 that may be adjusted include operating temperature, operating pressure, and percent hydrogen of the feedstock 2420. A high operating temperature may increase the formation of heavy mercaptans and heavy sulfides but also decrease the octane of the products. Increased temperatures may also decrease the olefin content in the light naphtha 2426 stream from the splitter 2412 because the increased temperature increases the likelihood that the olefin in the feedstock 2420 will combine with a sulfide to form a heavy mercaptan. In the event that an increase in diolefin content of the feedstock 2420 is determined, the operating temperature may be increased to increase the conversion of diolefins to olefins. Predicted lower diolefin content may permit a machine learning model or the predictive controls 2406 of the gasoline desulfurization controller 2402 to lower the operating temperature to manage toward a target sulfur concentration in the light naphtha 2426 stream from the splitter 2412. A lower operating temperature results in less olefin loss, longer catalyst life, and minimized risk of vaporization. Typically, the selective hydrogenation reactor 2410 is operated with the feedstock in the liquid phase as the gaseous phase reduces the ability of the process to remove the light sulfur compounds. Further, vaporization may impact the flow of feedstock and products through the selective hydrogenation reactor 2410. Consequently, operating temperature may be limited to a temperature below the boiling point of the feedstock.

[0175] A higher operating pressure may increase diolefin conversion to olefins and mercaptan conversion to heavy mercaptan. A higher operating pressure in the selective hydrogenation reactor 2410 may also reduce coke formation and polymerization of the feedstock as well as increase hydrogen dissolution in the feedstock. Further, a higher operating pressure may reduce vaporization risk.

[0176] Too high of a percent hydrogen of the feedstock 2420 may increase vaporization but too low may affect the longevity of the catalyst. Further, hydrogen consumption in the selective hydrogenation reactor 2410 is exothermic so improved control of the percent of hydrogen of the feedstock 2420 may result in better operating temperature control over the selective hydrogenation reactor 2410.

[0177] The selective hydrogenation reactor 2410 may be a single reactor or a plurality of reactors that may be arranged to feed products of the selective hydrogenation reactor 2410 into one or more splitters 2412. The selective hydrogenation reactor 2410 may include sensors to measure the inlet temperature of the feedstock 2420, the exothermic heat released by the hydrogenation of the naphtha feedstock 2420 within the selective hydrogenation reactor 2410, the pressure drop across the selective hydrogenation reactor 2410, and the phase of the feedstock 2420 within the selective hydrogenation reactor 2410.

[0178] A machine learning model may generate changes for a process controller such as a control algorithm, changes to an existing control algorithm, recommended changes to an operating parameter that may be published, including displayed or sent, to a user for approval or alteration. The changes or recommended changes may be based on a rate of cooling and a rate of heating available to the selective hydrogenation reactor. In some embodiments, the machine learning model may receive weather data indicative of current weather, and then determine the rate of cooling and the rate of heating available to the selective hydrogenation reactor by the machine learning model based on the weather data. The machine learning model may be trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data. The change or recommended change may include a changing rate over time to an operating parameter, such as temperature of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system. By changing the rate of change over time, the process may better accommodate the input and variability of different feedstocks, the mixing of the different feedstocks within theselective hydrogenation reactor 2410, and variances of the different feedstocks within the selective hydrogenation reactor.For example, a feedstock with high percentage of light naphtha is being processed in the selective hydrogenation reactor and a high olefin, high sulfur content feedstock may be indicated by feedstock data to be fed into the selective hydrogenation reactor 2410. The machine learning model may slowly begin to raise an operating temperature of a first catalyst bed of the selective hydrogenation reactor 2410 before the high olefin, high sulfur content feedstock is fed into the selective hydrogenation reactor 2410. After a delay, the rate of heat input increases and then slows as the measured operating temperature nears a predicted operating temperature for processing the high olefin, high sulfur content feedstock. Further, after a delay, an operating temperature of a second catalyst bed or a second selective hydrogenation reactor may be to accommodate the change in feedstock reaching the second catalyst bed. Additionally, a feed rate of hydrogen gas may be increased to accommodate an increase in predicted sulfur content. The feed rate may also change in a nonlinear manner to better match the feed rate of the naphtha and olefin feedstock. A change in feedstock data above a predetermined threshold may cause the machine learning model may generate an updated control algorithm or simulation for approval for implementation for one or more of the processes of the gasoline desulfurization system 2400 in order to better control the processes for the change in feedstock. In some embodiments, the control algorithm or simulation for approval for implementation for one or more of the processes of the gasoline desulfurization system 2400 may be generated at least once per day. In some embodiments, the control algorithm or simulation for approval for implementation for one or more of the processes of the gasoline desulfurization system 2400 may be generated at least every 12 hours. In some embodiments, the control algorithm or simulation for approval for implementation for one or more of the processes of the gasoline desulfurization system 2400 may be generated at least every 6 hours.

[0179] As products 2422 exit the selective hydrogenation reactor 2410, the products 2422 from the selective hydrogenation reactor 2410 may be passed through a sensor package 2442. The sensor package 2442 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, octane 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 products from the selective hydrogenation reactor 2410 to be analyzed in a lab to generate lab data. Testing may include bromine titration to obtain a bromine number and maleic anhydride titration to obtain MAV number, both of which may be used to determine the effectiveness of the conversion of diolefins in the feedstock 2420 into olefins in the products 2422 from the selective hydrogenation reactor 2410. In the event that the MAV number of the bromine number is 0, the operating temperature of the selective hydrogenation reactor 2410 may be too high or the flow rate of feedstock 2420 to low.

[0180] The products 2422 from the selective hydrogenation reactor 2410 include light olefins, heavy mercaptans, heavy sulfides with small quantities of diolefins and light sulfurs. These products may be passed through a splitter 2412 that may be operated to maximize octane retention with the overhead stream containing a sulfur concentration near a selected sulfur target. In some configurations, the splitter 2412 may be split the products 2422 from the selective hydrogenation reactor 2410 into a sweet purge 2427 of C4 and smaller molecules, a light naphtha 2426 including part of the C4 to part of the C6 hydrocarbons, and heavy naphtha 2424 that includes part of the C6 and C7 and greater molecules. The sweet purge 2427 may be measured by a sensor package 2445 to determine its hydrogen content, which may be used to adjust the hydrogen content of the feedstock 2420. The sweet purge 2427 may be set to gas processing or another process to separate the unused hydrogen gas, butane, and other light molecules.

[0181] In general, the cutoff between the light naphtha and the heavy naphtha may be around the condensation point of thiophene with the resulting light naphtha having a very low concentration of sulfur. The heavy naphtha 2424 will contain a majority of the sulfur from the feedstock 2420 in the form of the heavy mercaptans and heavy sulfides. Depending on the feedstock composition, the expected ratio of light naphtha 2426 to heavy naphtha 2424 may range from about 20% to 50%.

[0182] The splitter 2412 may be controlled to sharpen a desired cut off around the boiling point of thiophene by controlling reboil in the bottoms of the splitter 2412 and controlling reflux to improve fractionation of the vaporized naphtha components in the column of the splitter 2412. In other words, the operating parameters of the splitter 2412 include the feed rate of the products 2422, the heat input into the splitter 2412 to reboil the products 2422, column pressure, and controlling overhead cooling by use of reflux to maximize or improvethe sharpness in the cutoff between the light naphtha 2426 in the overhead stream and heavy naphtha 2424 in the bottoms stream or to target a sulfur concentration in the light naphtha 2426 of the overhead stream. Further, the efficiency of the splitter 2412 in separating the light naphtha 2426 in the overhead stream and heavy naphtha 2424 in the bottoms stream may also be affected by the composition of the products 2422, the splitter column design, and limits on available heat and cooling used to support reboil and reflux within the splitter 2412.

[0183] A sensor package 2444 may be disposed to determine the composition and attributes of the light naphtha 2426 from the splitter 2412. A sensor package 2446 may also be disposed to determine the composition and attributes of the heavy naphtha 2426 from the splitter 2412. The sensor packages 2444, 2445, and 2446 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, octane 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 products from the splitter 2412 to be analyzed in a lab to generate lab data. In particular, the sensor packages 2444, 2445, and 2446 may measure the flow rates of light naphtha 2426 to heavy naphtha 2424

[0184] The operational parameters of the splitter 2412 include controlling reboil in the bottoms of the splitter 2412 and controlling reflux to improve fractionation of the components of vaporized products 2422 in the column of the splitter 2412. In other words, the operating parameters of the splitter 2412 include the feed rate of the products 2422, the heat input into the splitter 2412 to reboil the products 2422, splitter pressure, and controlling overhead cooling by use of reflux to sharpen the cutoff between the heavy naphtha 2424 in the bottoms stream, the light naphtha 2426 from the overhead stream, and the sweet purge 2427. Further, the effectiveness of the splitter 2412 in separating products 2422 may be dependent on the cutoff point, the splitter 2412 column design, and limits on available heat and cooling used to support reboil and reflux.

[0185] The splitter 2412 may include sensors the measure the inlet feed rate of products 2422, the temperatures and pressures throughout the splitter 2412, and the outlet feed rates of heavy naphtha 2424, the light naphtha 2426, and the sweet purge 2427. The sensor data from the splitter 2412, and sensor packages 2444, 2445, 2446 may be used to determinedistillation information and effectiveness, including actual cutoff points, reflux -to-feed ratio, pressure drop across the splitter 2412, MAV number, bromine number, octane of the heavy naphtha 2424 and the light naphtha 2426, sulfur concentrations, light naphtha 2426 to heavy naphtha 2424 ratio, boiling points of the light naphtha 2426 and heavy naphtha 2424. The heavy naphtha 2424 is then sent to the hydrodesulfurization reactor 2414 to separate the sulfur from the heavy naphtha 2424.

[0186] The hydrodesulfurization reactor 2414 may include a plurality of hydrodesulfurization reactors, catalyst beds within each hydrodesulfurization reactor, and polishing reactors (not shown) arranged in parallel or in series. The heavy naphtha 2424 from the splitter 2412 may be mixed with hydrogen gas 2429 prior to being fed into the hydrodesulfurization reactor 2414. Additionally, hydrogen gas may be fed directly into the hydrodesulfurization reactor 2414 as a way to provide a cooler temperatures within the hydrodesulfurization reactor 2414 acting as a quench to reactor temperature. The hydrodesulfurization reactor 2414 operates to remove sulfur from the heavy naphtha 2424 through hydrodesulfurization. In hydrodesulfurization, hydrogen in the presence of a catalyst is able to separate the sulfur atom from the heavy sulfides and heavy mercaptan in the heavy naphtha 2424 to produce hydrogen sulfide and sulfur free hydrocarbons. Unfortunately, olefins in the heavy naphtha 2424 may also be saturated reducing the octane of the resulting heavy naphtha 2428 from the hydrodesulfurization reactor 2414. Additionally, mercaptan may reform from the hydrogen sulfide gas recombining with the available olefins as an exothermic equilibrium reaction.

[0187] The operating parameters of the hydrodesulfurization reactor 2414 include operating temperature, hydrogen gas 2429 feed rate, and operating pressure. The operating temperatures include inlet temperature, the change in temperature across each catalyst bed, hydrogen gas quench temperatures, and the outlet temperature of the resulting heavy naphtha 2428. Each of the temperatures within the hydrodesulfurization reactor may vary over time and may be used by a machine learning model to prevent temperature variations that may affect the reactions within the hydrodesulfurization reactor. In particular, hydrogen gas may be fed into the hydrodesulfurization reactor 2414 at different locations and act to quench or lower temperatures in regions of the hydrodesulfurization reactor 2414 to control temperature variations within the hydrodesulfurization reactor 2414.

[0188] In general, the operating temperature may be managed to a weighted average bed temperature or WABT. In general, the hydrodesulfurization reactor 2414 is operated so that the heavy naphtha 2424 is fully vaporized throughout the hydrodesulfurization reactor2414. The higher the WABT, the more hydrogen sulfide gas is formed and sulfur removed from the heavy naphtha 2428. The higher the WABT also results in increased olefin saturation resulting in octane loss and hydrogen consumption. The machine learning model used to adjust the gasoline desulfurization controller may adjust the operating temperatures of the hydrodesulfurization reactor 2414 to target a desired sulfur content while maximizing octane retention of the resulting heavy naphtha 2428. Further, hydrogen gas quench may be used to target localized temperatures that exceed desired process temperatures. Additionally, the exothermic heat of reaction may be determined for each hydrodesulfurization reactor or finishing reactor and may be used to estimate hydrogen consumption.

[0189] The hydrogen gas 2429 feed rate is also an important operating parameter that may be presented as a hydrogen to hydrocarbon ratio. The recombination reaction between hydrogen sulfide gas recombining with the available olefins may be reduced by increasing the hydrogen to hydrocarbon ratio. The rate of recombination is dependent on the partial pressure of hydrogen sulfide and the partial pressure of olefins in the hydrodesulfurization reactor 2414 and the piping from the hydrodesulfurization reactor 2414 to the separator 2416 or a finishing reactor. In the event a finishing reactor is used, the finishing reactor operates to raise the temperature of the resulting heavy naphtha 2428 in the presence of a catalyst to discourage the recombination reaction and facilitate the separation of hydrogen sulfide from the resulting heavy naphtha 2428 in the splitter 2418.

[0190] Each hydrodesulfurization reactor 2414 has a desired operating pressure based on its design. When the hydrodesulfurization reactor 2414 is operated at the desired operating pressure, the recombination reactions may be reduced.

[0191] As discussed above, a machine learning model may generate changes for a process controller such as a control algorithm, changes to an existing control algorithm, recommended changes to an operating parameter that may be published, including displayed or sent, to a user for approval or alteration. The changes or recommended changes may be based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor 2414. In some embodiments, the machine learning model may receive weather data indicative of current weather, and then determine the rate of cooling and the rate of heating available to the hydrodesulfurization reactor 2414 by the machine learning model based on the weather data. The machine learning model may be trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data. The change orrecommended change may include a changing rate over time to an operating parameter, such as temperature of the hydrodesulfurization reactor 2414 based on when the feedstock corresponding to the feedstock data will be fed into the hydrodesulfurization reactor 2414 of the gasoline desulfurization system. By changing the rate of change over time, the process may better accommodate the input of different feedstocks, the mixing of the different feedstocks within the hydrodesulfurization reactor 2414, and variances of the different feedstocks within the hydrodesulfurization reactor 2414. For example, a hydrogen gas quench may be varied with a changing feed rate of heavy naphtha product from the selective hydrogenation reactor and when the changing feed rate heavy naphtha product will be fed into the hydrodesulfurization reactor 2414.

[0192] As the resulting heavy naphtha 2428 exits the hydrodesulfurization reactor 2414, the resulting heavy naphtha 2428 may be measured, analyzed, and sampled by a sensor package 2448. The resulting heavy naphtha 2428 may then be fed into a separator 2416 that may be used to separate unreacted hydrogen 2430 from the resulting heavy naphtha 2428. The hydrogen 2430 may be recycled and fed back into the hydrodesulfurization reactor 2414 or to the selective hydrogeneration reactor 2410. Temperatures and pressures within the separator 2416 may be kept at levels to discourage the recombination reaction between hydrogen sulfide and the olefins in the resulting heavy naphtha 2428. The hydrogen 2430 and the separated heavy naphtha 2432 may be respectively measured, analyzed, and sampled by a sensor package 2450 and 2452.

[0193] The sensor packages 2448, 2450, and 2452 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, octane 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 resulting heavy naphtha 2428, the hydrogen 2430, and the separated heavy naphtha 2432 to be analyzed in a lab to generate lab data. In particular, the sulfur concentration, octane, temperature, composition, and pressure of the resulting heavy naphtha 2428, the hydrogen 2430, and the separated heavy naphtha 2432.

[0194] From the separator 2416, the separated heavy naphtha 2432 may be fed into the splitter 2418. The splitter 2418 may also be referred to as a stabilizer. The splitter 2418separates the off gas 2434, including hydrogen sulfide and light hydrocarbons, from the heavy naphtha 2436 which results in low sulfur concentration in the heavy naphtha 2436. The off gas 2434 may be sent to gas processing or amine treating. The off gas 2434 may be respectively measured, analyzed, and sampled by a sensor package 2454.

[0195] The heavy naphtha 2436 from the splitter 2418 may be sent to the gasoline pool. The heavy naphtha 2436 may be respectively measured, analyzed, and sampled by a sensor package 2456. The sensor packages 2454 and 2456 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, octane 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 resulting heavy naphtha 2436 and the off gas 2434 to be analyzed in a lab to generate lab data. In particular, the octane and sulfur content of the naphtha 2436 may be measured and used to adjust operating parameters of the hydrodesulfurization reactor 2414. Pressure drop across the selective hydrogeneration reactor 2410 and the hydrodesulfurization reactor 2414 may be measured and used to identify fouling in their catalyst beds or bed support failure.

[0196] A machine learning model may receive a desulfurized heavy naphtha product data from a the sensor packages 2448 and 2456. The desulfurized heavy naphtha product data may be indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor 2414 after removal of hydrogen sulfide gas. In some embodiments, the desulfurized heavy naphtha product data may indicate a concentration of sulfur in the desulfurized heavy naphtha. The machine learning model may generate an amendment to an existing algorithm, a recommended change, or a new change to an operating parameter of the hydrodesulfurization reactor 2414 based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target. The Reid vapor pressure of the desulfurized heavy naphtha product may also be determined by sensor package 2456 and may be used to control the operating parameters of the splitter 2418.

[0197] For a gasoline desulfurization operation, the gasoline desulfurization controller 2402 may optimize sulfur removal from gasoline. As a sulfur amount in a targeted product becomes lower, a reactor’s temperature may be increased to achieve the same amount of sulfur removal in the target product. However, this temperature increase does not result ina linear removal of sulfur from the resulting product. Higher temperatures may result in other negative results in the process including cracking of the hydrocarbons and loss of octane. Such behavior occurs due to the remaining sulfur in the product being difficult to treat and remove (in other words, the removing sulfur in a product including larger amounts of sulfur requires less energy than removing sulfur in a product including smaller amounts of sulfur). As such, a linear model may be less effective for control when utilized with for a target product sulfur content at lower values. Thus, use of a linear model results in less stable control, potential to go off product-specification (in other words, produce an inaccurate target product), and potential to prematurely deactivate catalyst.

[0198] Alternatively, a machine learning model may manage a light naphtha sulfur target and a heavy naphtha sulfur target as part of an overall average sulfur target for the combined streams while also maximizing the overall octane-barrels produced by the combined product of the gasoline desulfurization unit. Further, the machine learning model may manage one or more selective hydrogenation unit reactor variables, including temperature and pressure, and one or more splitter variables to produce the light naphtha product with the highest octane-barrels given a light naphtha sulfur target. The machine learning model may manage hydrodesulfurization reactor temperatures and other operating variables to produce a heavy naphtha product with maximum octane-barrels with a sulfur target.

[0199] FIG. 10A is a schematic diagram of an alternative embodiment of a gasoline desulfurization control system 2500. As shown, feedstock 2550 may be fed into a selective hydrogeneration reactor feedstock drum 2552. The feedstock 2550 From the selective hydrogeneration reactor feedstock drum 2552, the feedstock 2550 is fed serially into selective hydrogeneration reactors 2554 and 2556. Once processed in the selective hydrogeneration reactors 2554 and 2556, the processed feedstock 2557 is fed into a splitter 2558 where the processed feedstock 2557 is split into hydrogen gas 2560, desulfurized light naphtha 2562, and heavy naphtha 2564 including heavy sulfides and mercaptans.

[0200] The heavy naphtha 2564 is then fed into a hydrodesulfurization reactor feedstock drum 2566, where the heavy naphtha 2564 is fed into a hydrodesulfurization reactor 2568. In this configuration, the hydrodesulfurization reactor 2568 processes the sulfur from the heavy sulfides and mercaptans in the heavy naphtha 2564 into hydrogen sulfide and heavy naphtha 2569. The hydrogen sulfide and heavy naphtha 2569 is then moved into a hydrodesulfurization reactor finishing heater 2570 where the hydrogen sulfide and heavy naphtha 2569 is heated. The heated hydrogen sulfide and heavy naphtha 2569 is then passed into a finishing reactor 2572 where a catalyst is used to push the reaction equilibrium awayfrom the reformation of mercaptan and toward the formation of hydrogen sulfide and heavy naphtha 2569. The formation of mercaptan is favored in environments with a high temperature, low H2 partial pressure, high olefin, and high H2S concentration. The outlet of the hydrodesulfurization reactor 2568 is susceptible to recombinant mercaptan formation due to the heat generated and reduced hydrogen from exothermic olefin saturation.

[0201] The hydrogen sulfide and heavy naphtha 2569 is then passed into a separator 2574. The separator 2574 separates hydrogen gas 2576 that may be mixed into the heavy naphtha 2564 being fed into the hydrodesulfurization reactor 2568 from heavy naphtha 2578. The heavy naphtha 2578 may then be passed into a splitter 2580. The splitter 2580 separates sour gas 2582, including the hydrogen sulfide and light hydrocarbons from desulfurized heavy naphtha 2584. The desulfurized heavy naphtha 2584 may be sent to a gasoline blending pool.

[0202] In this configuration of the gasoline desulfurization control system 2500, a machine learning model may recommend a lower operating temperature of the hydrodesulfurization reactor 2568 while meeting a heavy naphtha sulfur concentration target. The finishing reactor 2572 is operated at a higher temperature within a range favorable for the catalyst to favor the formation of hydrogen sulfide and heavy naphtha 2569. Specifically, the machine learning model may generate recommendations and changes to the operating parameters to increase the temperature differential between the inlet of the hydrodesulfurization reactor 2568 and the operating temperature of the finishing reactor 2572. A lower temperature of the hydrodesulfurization reactor 2568 is favorable as it reduces olefin saturation while preserving octane-barrels to hit the heavy naphtha sulfur concentration target, while a higher temperature in the finishing reactor 2572 reduces recombination mercaptan formation to increase the removal of hydrogen sulfide from the heavy naphtha 2584. To increase the temperature differential, cooling may be applied to the heavy naphtha 2564 before it reaches the inlet of the hydrodesulfurization reactor 2568. A lower inlet temperature may allow the machine learning model to reduce quench cooling while maintaining lower rates of olefin saturation.

[0203] FIG. 11A and FIG. 11B are simplified diagrams of control systems 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, oneor 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.

[0204] 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.

[0205] 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 one 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.

[0206] 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.

[0207] 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.

[0208] 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.

[0209] In FIG. 11B, predictive controls 2714 may connect to subsets of each of the components described in FIG. 11 A. 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.

[0210] 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.

[0211] FIG. 12 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 operation controller 2501 and / or predictive controls 2514. 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 or one or more processors of the operation controller 2701. In other embodiments, method 2800 may be implemented in or included in components of FIGS. 1 A-28B. 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.

[0212] At block 2802, the one or more predictive controls may each obtain data from corresponding sources. In such an example, the each 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.

[0213] 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.

[0214] At block 2806, each predictive controls may determine whether the sub-parameters are different than currently set parameters. If the parameters are different, then, in some embodiments and at block 2808, 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 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.

[0215] At block 2810, the operation controller 2701 may determine updated parameters and / or fluid compositions 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, as well 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.

[0216] At block 2812, the operation controller 2701 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 maycompare 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.

[0217] 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.

[0218] FIG. 13 is flow chart of a method, according to at least one embodiment of the present disclosure. The method includes receiving gasoline blending pool data at 2910. The gasoline blending pool data is indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each current component of a gasoline blending pool. A machine learning model generates a light naphtha sulfur target for a light naphtha produced from a selective hydrogenation reactor of a gasoline desulfurization system based on the gasoline blending pool data at 2920 and generates a heavy naphtha sulfur target for a desulfurized heavy naphtha produced from a hydrodesulfurization reactor of the gasoline desulfurization system based on the gasoline blending pool data at 2930. The machine learning model has been trained on historical data of the gasoline blending pool and historical data of operating parameters, resulting product attributes, and composition from the gasoline desulfurization system. The historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

[0219] FIG. 14 is flow chart of a method, according to at least one embodiment of the present disclosure. The method includes receiving feedstock data from a sensor package at 3010. The feedstock data is indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed into a gasoline desulfurization system. The method further including generating, by a machine learning model, a recommended change to an operating parameter of a selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target at 3020 and implementing the recommended change by adjusting the operating parameter of the selective hydrogenation reactor at 3030.

[0220] FIG. 15 is flow chart of a method, according to at least one embodiment of the present disclosure. The method includes receiving a heavy naphtha product data from a sensor package. The heavy naphtha product data is indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenation reactor at 3110. The method further includes generating, by a machine learning model, a recommended changeto an operating parameter of a hydrodesulfurization reactor of a gasoline desulfurization system to produce a desulfurized heavy naphtha product having a sulfur concentration closer to a heavy naphtha sulfur target based on the heavy naphtha product data and the heavy naphtha sulfur target at 3120 and sending instructions to a controller to implement the recommended change to the operating parameter of the hydrodesulfurization reactor at 3130.

[0221] FIG. 16 is a flow is flow chart of another method, according to at least one embodiment of the present disclosure. The method includes receiving feedstock data from a sensor package at 3210. The feedstock data is indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed a gasoline desulfurization system. A machine learning model predicts when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system at 3220 and generates a change to an operating parameter of the selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target at 3230. The change includes a changing rate over time of the operating parameter of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system.

[0222] FIG. 17 is a flow is flow chart of yet another method, according to at least one embodiment of the present disclosure. The method includes receiving a heavy naphtha product data from a sensor package at 3310. The heavy naphtha product data is indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenation reactor. A machine learning model predicts when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into a hydrodesulfurization reactor of a gasoline desulfurization system at 3320 and generates a change to an operating parameter of the hydrodesulfurization reactor based on the heavy naphtha product data and a heavy naphtha sulfur target at 3330. The change includes a changing rate over time of the operating parameter of the hydrodesulfurization reactor based on when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

[0223] FIG. 18 is a flow is flow chart of a method, according to at least one embodiment of the present disclosure. The method includes receiving business data at 3410. The business data includes feedstock costs, and a current cost of operating each process of a gasoline desulfurization system. The method further includes receiving feedstock data at3420. The feedstock data is indicative of one or more of a composition of a feedstock or a feed rate of the feedstock being fed into the gasoline desulfurization system. The method includes receiving gasoline blending pool data at 3430. The gasoline blending pool data is indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each component of the gasoline blending pool. A machine learning model generates a plurality of simulations based on the feedstock data, the business data, and gasoline blending pool data operating at different operational parameters of the gasoline desulfurization system at 3440 and generates a cost, octane, and estimated sulfur concentration of a light naphtha produced from splitter downstream from a selective hydrogenation reactor for each simulation at 3450. The machine learning model generates a cost, octane, and estimated sulfur concentration of a desulfurized heavy naphtha produced from the splitter downstream from a hydrodesulfurization reactor for each simulation at 3460. The machine learning model has been trained on historical data of the gasoline blending pool, historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system. The historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

[0224] 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

CLAIMSWHAT IS CLAIMED IS:

1. A method comprising: receiving gasoline blending pool data, the gasoline blending pool data indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each current component of a gasoline blending pool; generating, by a machine learning model, a light naphtha sulfur target for a light naphtha produced from a selective hydrogenation reactor of a gasoline desulfurization system based on the gasoline blending pool data; and generating, by the machine learning model, a heavy naphtha sulfur target for a desulfurized heavy naphtha produced from a hydrodesulfurization reactor of the gasoline desulfurization system based on the gasoline blending pool data, wherein the machine learning model has been trained on historical data of the gasoline blending pool and historical data of operating parameters, resulting product attributes, and composition from the gasoline desulfurization system, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

2. The method of claim 1, wherein generating by a machine learning model, the light naphtha sulfur target is also based on enhancing retained octane-barrels of the light naphtha produced from the selective hydrogenation reactor.

3. The method of claims 1 or 2, wherein generating by a machine learning model, the heavy naphtha sulfur target is also based on enhancing retained octane-barrels of the desulfurized heavy naphtha produced from the hydrodesulfurization reactor.

4. The method of any of claims 1 to 3, wherein generating by a machine learning model, the light naphtha sulfur target is also based on operational constraints of the selective hydrogenation reactor, wherein generating by a machine learning model, the heavy naphtha sulfur target is also based on operational constraints of the hydrodesulfurization reactor.

5. The method of any of claims 1 to 4, further comprising:receiving feedstock data from a sensor package, the feedstock data being indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed into a gasoline desulfurization system; and generating, by a machine learning model, a recommended change to an operating parameter of the selective hydrogenation reactor to produce a light naphtha product having a sulfur concentration closer to the light naphtha sulfur target based on the feedstock data and the light naphtha sulfur target.

6. The method of claim 5, further comprising adjusting the operating parameter of the selective hydrogenation reactor.

7. The method of claim 5, further comprising sending instructions to a controller to implement the recommended change to the operating parameter of the selective hydrogenation reactor, wherein the operating parameter of the selective hydrogenation reactor is changed to implement the recommended change.

8. The method of any of claims 5 to 7, wherein the feedstock includes a naphtha.

9. The method of any of claims 5 to 8, wherein the feedstock includes a hydrogen gas.

10. The method of any of claims 5 to 9, wherein receiving feedstock data includes the operating parameters of a process upstream from the gasoline desulfurization system, the method further comprising interpolating, by the machine learning model, a feedstock composition from the operating parameters of a process upstream from the gasoline desulfurization system.

11. The method of any of claims 5 to 10, further comprising predicting, by a machine learning model, when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system, wherein the recommended change to the operating parameter is also based on when the feedstock will be fed into the selective hydrogenation reactor.

12. The method of any of claims 5 to 11, wherein the recommended change to the operating parameter includes a change to a feed rate of feedstock, the method furthercomprising sending a portion of the feedstock to a bypass to reduce the feed rate of feedstock into the selective hydrogenation reactor.

13. The method of any of claims 5 to 12, wherein the recommended change includes a change to an operating temperature of the selective hydrogenation reactor.

14. The method of any of claims 5 to 13, wherein the recommended change is based on a rate of cooling and a rate of heating available to the selective hydrogenation reactor.

15. The method of claim 14, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the selective hydrogenation reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

16. The method of any of claims 5 to 15, wherein the recommended change includes a changing rate over time to an operating temperature of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system.

17. The method of any of claims 1 to 16, further comprising: receiving market pricing data for the light naphtha product; generating a second recommended change to an operating parameter of the selective hydrogenation reactor based on the market pricing data; amending the recommended change with the second recommended change.

18. The method of any of claims 1 to 16, further comprising: receiving a heavy naphtha product data from a second sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from the selective hydrogenation reactor; generating, by a machine learning model, a recommended change to an operating parameter of the hydrodesulfurization reactor to produce a desulfurized heavy naphthaproduct having a sulfur concentration closer to the heavy naphtha sulfur target based on the heavy naphtha product data and the heavy naphtha sulfur target.

19. The method of claim 18, further comprising sending instructions to a controller to implement the recommended change to the operating parameter of the hydrodesulfurization reactor.

20. The method of claim 19, further comprising changing the operating parameter of the hydrodesulfurization reactor to implement the recommended change.

21. The method of any of claims 18 to 20, further comprising receiving the heavy naphtha product data includes the operating parameters of a process upstream from the hydrodesulfurization reactor; and interpolating, by the machine learning model, a composition of the heavy naphtha product from the operating parameters of a process upstream from the hydrodesulfurization reactor.

22. The method of any of claims 18 to 21, further comprising predicting, by a machine learning model, when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor, wherein the recommended change to the operating parameter is also based on when the heavy naphtha product will be fed into the hydrodesulfurization reactor.

23. The method of any of claims 18 to 22, wherein the recommended change includes a changing rate over time to an operating temperature of the hydrodesulfurization reactor based on when the heavy naphtha product from the selective hydrogenation reactor corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

24. The method of any of claims 18 to 23, wherein the recommended change includes a change to an operating temperature of the hydrodesulfurization reactor.

25. The method of any of claims 18 to 24, wherein the recommended change is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

26. The method of claim 25, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the hydrodesulfurization reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

27. The method of any of claims 18 to 26, wherein the recommended change to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

28. The method of any of claims 18 to 27, wherein the recommended change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

29. The method of any of claims 18 to 27, further comprising splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

30. The method of claim 29, further comprising receiving a desulfurized heavy naphtha product data from a third sensor package, the desulfurized heavy naphtha product data being indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor after removal of hydrogen sulfide gas, wherein the desulfurized heavy naphtha product data indicates a concentration of sulfur in the desulfurized heavy naphtha.

31. The method of claim 30, further comprising: generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target; and amending the recommended change with the second recommended change.

32. The method of any of claims 1 to 31, further comprising: receiving market pricing data for the desulfurized heavy naphtha product; generating a third recommended change to an operating parameter of the hydrodesulfurization reactor based on the market pricing data; amending the recommended change with the third recommended change.

33. The method of any of claims 1 to 32, further comprising: publishing a recommended change to a user.

34. The method of claim 33, further comprising: receiving approval to implement the recommended change.

35. The method of any of claims 1 to 34, further comprising publishing a report of implemented changes to the operating parameters of the gasoline desulfurization system for a period of time.

36. The method of any of claims 5 to 35, wherein each recommended change is automatically implemented by the machine learning model.

37. The method of any of claims 5 to 36, further comprising generating, by the machine learning model, a control algorithm to control a process of the gasoline desulfurization system to produce a light naphtha product having a sulfur concentration closer to the light naphtha sulfur target, wherein the recommended change is the control algorithm for the process of the gasoline desulfurization system.

38. The method of claim 37, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in feedstock above a predetermined threshold has occurred, the control algorithm is generated.

39. The method of any of claims 5 to 38, further comprising generating, by the machine learning model, a control algorithm to control a process of the gasoline desulfurization system to produce a desulfurized heavy naphtha product having a sulfur concentrationcloser to the heavy naphtha sulfur target, wherein the recommended change is the control algorithm for the process of the gasoline desulfurization system.

40. The method of claim 39, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in feedstock above a predetermined threshold has occurred, the control algorithm is generated.

41. The method of any of claims 1 to 40, wherein a feedstock for the selective hydrogenation reactor includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from the resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

42. The method of any of claims 1 to 41, wherein the machine learning model is trained on the historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of a splitter splitting a product of the selective hydrogenation reactor, properties of feedstock, and properties of the produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

43. The method of any of claims 1 to 42, wherein the machine learning model is trained on the historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and the attributes and properties of the produced desulfurized light naphtha product and the desulfurized heavy naphtha product, the method further comprising: receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)44. The method of claim 43, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)45. The method of any of claims 1 to 43, further comprising generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)46. A memory including instructions and a machine learning model that causes a processor to perform instructions including the method of any of claims 1 to 45.

47. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 1 to 45.

48. A method comprising: receiving feedstock data from a sensor package, the feedstock data being indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed into a gasoline desulfurization system; generating, by a machine learning model, a recommended change to an operating parameter of a selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target; andimplementing the recommended change by adjusting the operating parameter of the selective hydrogenation reactor.

49. The method of claim 48, wherein implementing the recommended change includes sending instructions to a controller to implement the recommended change to the operating parameter of the selective hydrogenation reactor.

50. The method of claims 48 or 49, wherein the machine learning model has been trained on historical data of the feedstock data, the operating parameters and resulting product attributes and composition of the gasoline desulfurization system, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration.

51. The method of any of claims 48 to 50, wherein the light naphtha sulfur target is based on enhancing retained octane-barrels of the light naphtha produced from the selective hydrogenation reactor.

52. The method of any of claims 48 to 51 , wherein the light naphtha sulfur target is also based on operational constraints of the selective hydrogenation reactor53. The method of any of claims 48 to 52, further comprising generating, by the machine learning model, a control algorithm to control a process of the gasoline desulfurization system to produce a light naphtha product having a sulfur concentration closer to the light naphtha sulfur target, wherein the recommended change is the control algorithm for the process of the gasoline desulfurization system.

54. The method of claim 53, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in feedstock above a predetermined threshold has occurred, the control algorithm is generated.

55. The method of any of claims 48 to 53, further comprising generating, by the machine learning model, a control algorithm to control a process of the gasoline desulfurization system to produce a desulfurized heavy naphtha product having a sulfurconcentration closer to a heavy naphtha sulfur target, wherein the recommended change is the control algorithm for the process of the gasoline desulfurization system.

56. The method of claim 55, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in feedstock above a predetermined threshold has occurred, the control algorithm is generated.

57. The method of any of claims 48 to 56, wherein the feedstock includes a naphtha.

58. The method of any of claims 48 to 57, wherein the feedstock includes a hydrogen gas.

59. The method of any of claims 48 to 58, wherein receiving feedstock data includes the operating parameters of a process upstream from the gasoline desulfurization system, the method further comprising interpolating, by the machine learning model, a feedstock composition from the operating parameters of a process upstream from the gasoline desulfurization system.

60. The method of any of claims 48 to 59, further comprising predicting, by a machine learning model, when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system, wherein the recommended change to the operating parameter is also based on when the feedstock will be fed into the selective hydrogenation reactor.

61. The method of any of claims 48 to 60, wherein the recommended change to the operating parameter includes a change to a feed rate of feedstock, the method further comprising sending a portion of the feedstock to a bypass to reduce the feed rate of feedstock into the selective hydrogenation reactor.

62. The method of any of claims 48 to 61, wherein the recommended change includes a change to an operating temperature of the selective hydrogenation reactor.

63. The method of any of claims 48 to 62, wherein the recommended change is based on a rate of cooling and a rate of heating available to the selective hydrogenation reactor.

64. The method of claim 63, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the selective hydrogenation reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

65. The method of any of claims 48 to 64, further comprising: receiving market pricing data for a light naphtha product; generating a second recommended change to an operating parameter of the selective hydrogenation reactor based on the market pricing data; amending the recommended change with the second recommended change.

66. The method of any of claims 48 to 64, wherein the recommended change includes a changing rate over time to an operating temperature of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system.

67. The method of any of claims 48 to 66, further comprising: receiving a heavy naphtha product data from a second sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from the selective hydrogenation reactor; generating, by a machine learning model, a recommended change to an operating parameter of a hydrodesulfurization reactor to produce a desulfurized heavy naphtha product having a sulfur concentration closer to a heavy naphtha sulfur target based on the heavy naphtha product data and the heavy naphtha sulfur target.

68. The method of claim 67, wherein the heavy naphtha sulfur target is also based on enhancing retained octane-barrels of the desulfurized heavy naphtha produced from the hydrodesulfurization reactor.

69. The method of any of claims 67 to 68, wherein the heavy naphtha sulfur target is also based on operational constraints of the hydrodesulfurization reactor.

70. The method of claim 67, wherein receiving the heavy naphtha product data includes the operating parameters of a process upstream from the hydrodesulfurization reactor; and interpolating, by the machine learning model, a composition of the heavy naphtha product from the operating parameters of a process upstream from the hydrodesulfurization reactor.

71. The method of any of claims 67 to 70, further comprising predicting, by a machine learning model, when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor, wherein the recommended change to the operating parameter is also based on when the heavy naphtha product will be fed into the hydrodesulfurization reactor.

72. The method of any of claims 67 to 71, wherein the recommended change includes a changing rate over time to an operating temperature of the hydrodesulfurization reactor based on when the heavy naphtha product from the selective hydrogenation reactor corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

73. The method of any of claims 67 to 72, wherein the recommended change includes a change to an operating temperature of the hydrodesulfurization reactor.

74. The method of any of claims 67 to 73, wherein the recommended change is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

75. The method of claim 74, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the hydrodesulfurization reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operatingparameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

76. The method of any of claims 67 to 75, wherein the recommended change to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

77. The method of any of claims 67 to 76, wherein the recommended change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

78. The method of any of claims 67 to 76, further comprising splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

79. The method of claim 78, further comprising receiving a desulfurized heavy naphtha product data from a third sensor package, the desulfurized heavy naphtha product data being indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor after removal of hydrogen sulfide gas, wherein the desulfurized heavy naphtha product data indicates a concentration of sulfur in the desulfurized heavy naphtha.

80. The method of claim 79, further comprising: generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target; and amending the recommended change with the second recommended change.

81. The method of any of claims 48 to 80, further comprising: receiving market pricing data for a desulfurized heavy naphtha product; generating a third recommended change to an operating parameter of a hydrodesulfurization reactor based on the market pricing data; amending the recommended change with the third recommended change.

82. The method of any of claims 48 to 81, further comprising: publishing a recommended change to a user.

83. The method of claim 82, further comprising: receiving approval to implement the recommended change.

84. The method of any of claims 48 to 83, further comprising publishing a report of implemented changes to the operating parameters of the gasoline desulfurization system for a period of time.

85. The method of any of claims 48 to 84, wherein each recommended change is automatically implemented by the machine learning model.

86. The method of any of claims 48 to 85, wherein the feedstock includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from a resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

87. The method of any of claims 48 to 86, wherein the machine learning model is trained on historical data of the operating parameters of a hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of a splitter splitting a product of selective hydrogenation reactor, properties of feedstock, and the properties of produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

88. The method of any of claims 48 to 87, wherein the machine learning model is trained on historical data of the operating parameters of a hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and properties of a desulfurized light naphtha product and a desulfurized heavy naphtha product, the method further comprising:receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)89. The method of claim 88, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)90. The method of any of claims 48 to 88, further comprising generating the operating parameters of the selective hydrogenation reactor and a hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)91. A memory including instructions and a machine learning model that causes a processor to perform the instructions including the method of any of claims 48 to 90.

92. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 48 to 90.

93. A method compri sing :receiving a heavy naphtha product data from a sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenation reactor; generating, by a machine learning model, a recommended change to an operating parameter of a hydrodesulfurization reactor of a gasoline desulfurization system to produce a desulfurized heavy naphtha product achieving a heavy naphtha sulfur target based on the heavy naphtha product data and the heavy naphtha sulfur target; and implementing the recommended change by adjusting the operating parameter of the hydrodesulfurization reactor.

94. The method of claim 93, sending instructions to a controller to implement the recommended change to the operating parameter of the selective hydrogenation reactor.

95. The method of claims 93 or 94, wherein receiving the heavy naphtha product data includes the operating parameters of a process upstream from the hydrodesulfurization reactor, the method further comprising interpolating, by the machine learning model, a composition of the heavy naphtha product from the operating parameters of a process upstream from the hydrodesulfurization reactor.

96. The method of any of claims 93 or 95, wherein the heavy naphtha sulfur target is based on enhancing retained octane-barrels of the desulfurized heavy naphtha produced from the hydrodesulfurization reactor.

97. The method of any of claims 93 to 96, wherein the heavy naphtha sulfur target is based on operational constraints of the hydrodesulfurization reactor.

98. The method of any of claims 93 to 97, further comprising predicting, by a machine learning model, when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor, wherein the recommended change to the operating parameter is also based on when the heavy naphtha product will be fed into the hydrodesulfurization reactor.

99. The method of any of claims 93 to 98, wherein the recommended change includes a changing rate over time to an operating temperature of the hydrodesulfurization reactorbased on when the heavy naphtha product from the selective hydrogenation reactor corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

100. The method of any of claims 93 to 99, wherein the recommended change includes a change to an operating temperature of the hydrodesulfurization reactor.

101. The method of any of claims 93 to 100, wherein the recommended change is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

102. The method of claim 101, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the hydrodesulfurization reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

103. The method of any of claims 93 to 102, wherein the recommended change to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

104. The method of any of claims 93 to 103, wherein the recommended change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

105. The method of any of claims 93 to 103, further comprising splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

106. The method of claim 105, further comprising receiving a desulfurized heavy naphtha product data from a second sensor package, the desulfurized heavy naphtha product data being indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor after removal of hydrogen sulfidegas, wherein the desulfurized heavy naphtha product data indicates a concentration of sulfur in the desulfurized heavy naphtha.

107. The method of claim 106, further comprising: generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target; and amending the recommended change with the second recommended change.

108. The method of any of claims 93 to 107, further comprising: generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on market pricing data; amending the recommended change with the second recommended change.

109. The method of any of claims 93 to 108, further comprising: publishing a recommended change to a user.

110. The method of claim 109, further comprising: receiving approval to implement the recommended change.

111. The method of any of claims 93 to 110, further comprising publishing a report of implemented changes to the operating parameters of the gasoline desulfurization system for a period of time.

112. The method of any of claims 93 to 111, wherein each recommended change is automatically implemented by the machine learning model.

113. The method of any of claims 93 to 112, wherein feedstock includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from a resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

114. The method of any of claims 93 to 113, wherein the machine learning model is trained on historical data of the operating parameters of the hydrodesulfurization reactor,the operating parameters of the selective hydrogenation reactor, the operating parameters of a splitter splitting a product of the selective hydrogenation reactor, properties of feedstock, and properties of the produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

115. The method of any of claims 93 to 114, wherein the machine learning model is trained on historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and properties of a desulfurized light naphtha product and the desulfurized heavy naphtha product, the method further comprising: receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)116. The method of claim 115, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)117. The method of any of claims 93 to 115, further comprising generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)118. A memory including instructions and a machine learning model that causes a processor to perform the instructions including the method of any of claims 93 to 117.

119. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 93 to 117.

120. A method compri sing : receiving feedstock data from a sensor package, the feedstock data being indicative of a change to one or more of a composition of a feedstock or a feed rate of the feedstock being fed a gasoline desulfurization system; predicting, by a machine learning model, when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system; and generating, by a machine learning model, a change to an operating parameter of the selective hydrogenation reactor based on the feedstock data and a light naphtha sulfur target; wherein the change includes a changing rate over time of the operating parameter of the selective hydrogenation reactor based on when the feedstock corresponding to the feedstock data will be fed into a selective hydrogenation reactor of the gasoline desulfurization system.

121. The method of claim 120, wherein the machine learning model has been trained on historical data of feedstock composition, including changes in feedstock over time, and historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system changing over time, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration.

122. The method of any of claims 120 to 121, wherein the light naphtha sulfur target is based on enhancing retained octane-barrels of the light naphtha produced from the selective hydrogenation reactor.

123. The method of any of claims 120 to 122, wherein the light naphtha sulfur target is based on operational constraints of the selective hydrogenation reactor.

124. The method of any of claims 120 to 123, wherein the feedstock includes a naphtha.

125. The method of any of claims 120 to 124, wherein the feedstock includes a hydrogen gas.

126. The method of any of claims 120 to 125, wherein receiving feedstock data includes the operating parameters of a process upstream from the gasoline desulfurization system, the method further comprising interpolating, by the machine learning model, a feedstock composition from the operating parameters of a process upstream from the gasoline desulfurization system.

127. The method of any of claims 120 to 126, wherein the change to the operating parameter includes a change to a feed rate of feedstock, the method further comprising sending a portion of the feedstock to a bypass to reduce the feed rate of feedstock into the selective hydrogenation reactor.

128. The method of any of claims 120 to 127, wherein the change includes a change to an operating temperature of the selective hydrogenation reactor.

129. The method of any of claims 120 to 128, wherein the change is based on a rate of cooling and a rate of heating available to the selective hydrogenation reactor.

130. The method of claim 129, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the selective hydrogenation reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parametersand resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

131. The method of any of claims 120 to 130, further comprising: receiving market pricing data for a light naphtha product; generating a second recommended change to an operating parameter of the selective hydrogenation reactor based on the market pricing data; amending the change with the second recommended change.

132. The method of any of claims 120 to 131, further comprising: receiving a heavy naphtha product data from a sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from the selective hydrogenation reactor; predicting, by a machine learning model, when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into a hydrodesulfurization reactor of the gasoline desulfurization system; and generating, by a machine learning model, a change to an operating parameter of the hydrodesulfurization reactor based on the heavy naphtha product data and a heavy naphtha sulfur target; wherein the change includes a changing rate over time of the operating parameter of the hydrodesulfurization reactor based on when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

133. The method of claim 132, wherein the machine learning model has been trained on historical data of heavy naphtha product data composition, including changes in heavy naphtha product data over time, and historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system changing over time, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration.

134. The method of any of claims 132 or 133, wherein generating by a machine learning model, the heavy naphtha sulfur target is based on enhancing retained octane-barrels of the desulfurized heavy naphtha produced from the hydrodesulfurization reactor.

135. The method of any of claims 132 to 134, wherein the heavy naphtha sulfur target is based on operational constraints of the hydrodesulfurization reactor.

136. The method of any of claims 132 to 135, further comprising sending instructions to a controller to implement the change to the operating parameter of the hydrodesulfurization reactor, wherein changing the operating parameter of the hydrodesulfurization reactor to implement the change.

137. The method of any of claims 132 to 136, wherein the change includes a change to an operating temperature of the hydrodesulfurization reactor.

138. The method of any of claims 132 to 137, wherein the change is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

139. The method of any of claims 132 to 138, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the hydrodesulfurization reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

140. The method of any of claims 132 to 139, wherein the change to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

141. The method of any of claims 132 to 140, wherein the change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

142. The method of any of claims 132 to 140, further comprising splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

143. The method of any of claims 132 to 142, further comprising receiving a desulfurized heavy naphtha product data from a second sensor package, the desulfurized heavy naphtha product data being indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor after removal of hydrogen sulfide gas, wherein the desulfurized heavy naphtha product data indicates a concentration of sulfur in the desulfurized heavy naphtha.

144. The method of any of claims 132 to 143, further comprising: generating a recommended change to an operating parameter of the hydrodesulfurization reactor based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target; and amending the change with the recommended change.

145. The method of any of claims 132 to 144, further comprising: receiving market pricing data for the desulfurized heavy naphtha product; generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on the market pricing data; amending the change with the second recommended change.

146. The method of any of claims 120 to 145, wherein the feedstock includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from a resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

147. The method of any of claims 120 to 146, wherein the machine learning model is trained on historical data of the operating parameters of a hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of a splitter splitting a product of the selective hydrogenation reactor, properties of feedstock, and properties of produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

148. The method of any of claims 120 to 147, wherein the machine learning model is trained on historical data of the operating parameters of a hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and properties of a desulfurized light naphtha product and a desulfurized heavy naphtha product, the method further comprising: receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN R0N)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)149. The method of claim 148, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN R0N)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)150. The method of any of claims 120 to 148, further comprising generating the operating parameters of the selective hydrogenation reactor and a hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)151. A memory including instructions and a machine learning model that causes a processor to perform the instructions including the method of any of claims 120 to 149.

152. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 120 to 150.

153. A method compri sing : receiving a heavy naphtha product data from a sensor package, the heavy naphtha product data being indicative of a composition or an attribute of a heavy naphtha product from a selective hydrogenation reactor; predicting, by a machine learning model, when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into a hydrodesulfurization reactor of a gasoline desulfurization system; and generating, by a machine learning model, a change to an operating parameter of the hydrodesulfurization reactor based on the heavy naphtha product data and a heavy naphtha sulfur target; wherein the change includes a changing rate over time of the operating parameter of the hydrodesulfurization reactor based on when the heavy naphtha product corresponding to the heavy naphtha product data will be fed into the hydrodesulfurization reactor of the gasoline desulfurization system.

154. The method of claim 153, wherein the machine learning model has been trained on historical data of heavy naphtha product data composition, including changes in heavy naphtha product data over time, and historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system changing over time, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration.

155. The method of any of claims 153 or 154, wherein the heavy naphtha sulfur target is based on enhancing retained octane-barrels of the desulfurized heavy naphtha produced from the hydrodesulfurization reactor.

156. The method of any of claims 153 to 155, wherein the heavy naphtha sulfur target is based on operational constraints of the hydrodesulfurization reactor.

157. The method of any of claims 153 to 156, further comprising sending instructions to a controller to implement the change to the operating parameter of the hydrodesulfurizationreactor, wherein changing the operating parameter of the hydrodesulfurization reactor to implement the change.

158. The method of any of claims 153 to 154, wherein the change includes a change to an operating temperature of the hydrodesulfurization reactor.

159. The method of any of claims 153 to 158, wherein the change is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

160. The method of any of claims 153 to 159, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the hydrodesulfurization reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

161. The method of any of claims 153 to 160, wherein the change to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

162. The method of any of claims 153 to 161, wherein the change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

163. The method of any of claims 153 to 161, further comprising splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

164. The method of any of claims 153 to 163, further comprising receiving a desulfurized heavy naphtha product data from a second sensor package, the desulfurized heavy naphtha product data being indicative of a composition or an attribute of the desulfurized heavy naphtha product from the hydrodesulfurization reactor after removal of hydrogen sulfidegas, wherein the desulfurized heavy naphtha product data indicates a concentration of sulfur in the desulfurized heavy naphtha.

165. The method of any of claims 153 to 164, further comprising: generating a recommended change to an operating parameter of the hydrodesulfurization reactor based on the desulfurized heavy naphtha product data and the heavy naphtha sulfur target; and amending the change with the recommended change.

166. The method of any of claims 153 to 165, further comprising: receiving market pricing data for the desulfurized heavy naphtha product; generating a second recommended change to an operating parameter of the hydrodesulfurization reactor based on the market pricing data; amending the change with the second recommended change.

167. The method of any of claims 153 to 166, further comprising: publishing the change to a user.

168. The method of claim 167, further comprising: receiving approval to implement the change.

169. The method of any of claims 153 to 168, further comprising publishing a report of implemented changes to the operating parameters of the gasoline desulfurization system for a period of time.

170. The method of any of claims 153 to 169, wherein each recommended change is automatically implemented by the machine learning model.

171. The method of any of claims 153 to 170, wherein feedstock includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from a resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

172. The method of any of claims 153 to 171, wherein the machine learning model is trained on historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of a splitter splitting a product of the selective hydrogenation reactor, properties of feedstock, and properties of produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

173. The method of any of claims 153 to 172, wherein the machine learning model is trained on historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and properties of a desulfurized light naphtha product and the desulfurized heavy naphtha product, the method further comprising: receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)174. The method of claim 173, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHN RON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)175. The method of any of claims 153 to 173, further comprising generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)176. A memory including instructions and a machine learning model that causes a processor to perform the instructions including the method of any of claims 153 to 175.

177. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 153 to 175.

178. A method compri sing : receiving business data, the business data including feedstock costs, and a current cost of operating each process of a gasoline desulfurization system; receiving feedstock data, the feedstock data being indicative of one or more of a composition of a feedstock or a feed rate of the feedstock being fed into the gasoline desulfurization system; receiving gasoline blending pool data, the gasoline blending pool data indicative of one or more of a sulfur concentration, a current inventory, or an octane number of each component of the gasoline blending pool; generating, by a machine learning model, a plurality of simulations based on the feedstock data, the business data, and gasoline blending pool data operating at different operational parameters of the gasoline desulfurization system; generating, by a machine learning model, a cost, octane, and estimated sulfur concentration of a light naphtha produced from splitter downstream from a selective hydrogenation reactor for each simulation; generating, by a machine learning model, a cost, octane, and estimated sulfur concentration of a desulfurized heavy naphtha produced from the splitter downstream from a hydrodesulfurization reactor for each simulation, wherein the machine learning model has been trained on historical data of the gasoline blending pool, historical data of the operating parameters and resulting product attributes and composition of the gasolinedesulfurization system, wherein the historical data of the operating parameters and resulting product attributes and composition include sulfur concentration and octane number.

179. The method of claim 178, further comprising publishing simulation data related to one or more of the simulations of the gasoline desulfurization system, the data including the cost and estimated sulfur concentration of the light naphtha produced from a splitter downstream from the selective hydrogenation reactor and the cost and estimated sulfur concentration of the desulfurized heavy naphtha produced from a second splitter downstream from the hydrodesulfurization reactor.

180. The method of any of claims 178 to 179, further comprising: receiving a selection of one of the one or more of the simulations of the gasoline desulfurization system; and implementing the selected simulation of the gasoline desulfurization system by changing the operating parameters of the gasoline desulfurization system to align with operating parameters of the selected simulation.

181. The method of any of claims 178 to 180, further comprising: selecting one of the one or more of the simulations of the gasoline desulfurization system; and implementing the selected simulation of the gasoline desulfurization system by changing the operating parameters of the gasoline desulfurization system to align with operating parameters of the selected simulation.

182. The method of any of claims 180 to 181, further comprising: generating a control algorithm for a controller of the selective hydrogenation reactor from the selected simulation; and updating the controller of the selective hydrogenation reactor with the control algorithm for a controller of the selective hydrogenation reactor.

183. The method of claims 182, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change infeedstock above a predetermined threshold has occurred, the control algorithm for the controller of the selective hydrogenation reactor is generated.

184. The method of any of claims 180 to 183, further comprising: generating a control algorithm for a controller of the hydrodesulfurization reactor from the selected simulation; and updating the controller of the hydrodesulfurization reactor with the control algorithm for a controller of the hydrodesulfurization reactor.

185. The method of claims 184, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in feedstock above a predetermined threshold has occurred, the control algorithm for the controller of the hydrodesulfurization reactor is generated.

186. The method of any of claims 180 to 185, wherein the plurality of simulations are generated at least once per day.

187. The method of any of claims 180 to 184, wherein the plurality of simulations are generated at least every 12 hours.

188. The method of any of claims 180 to 184, wherein the plurality of simulations are generated at least every 6 hours.

189. The method of any of claims 180 to 188, further comprising determining, by the machine learning model, from the feedstock data that a change in feedstock above a predetermined threshold has occurred, wherein the plurality of simulations are generated when the machine learning model determines that a change in feedstock above a predetermined threshold has occurred.

190. The method of any of claims 180 to 189, wherein the feedstock includes a naphtha.

191. The method of any of claims 180 to 190, wherein the feedstock includes a hydrogen gas.

192. The method of any of claims 180 to 191, wherein receiving feedstock data includes the operating parameters of a process upstream from the gasoline desulfurization system, the method further comprising interpolating, by the machine learning model, a feedstock composition from the operating parameters of a process upstream from the gasoline desulfurization system.

193. The method of any of claims 180 to 192, further comprising predicting, by a machine learning model, when the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system, wherein the simulation is also based on when the feedstock will be fed into the selective hydrogenation reactor.

194. The method of any of claims 180 to 193, wherein the simulation includes sending a portion of the feedstock to a bypass to reduce the feed rate of feedstock into the selective hydrogenation reactor.

195. The method of any of claims 180 to 194, wherein the simulation includes a change to an operating temperature of the selective hydrogenation reactor.

196. The method of any of claims 180 to 195, wherein the simulation is based on a rate of cooling and a rate of heating available to the selective hydrogenation reactor.

197. The method of claim 196, further comprising: receiving a weather data indicative of current weather; and determining the rate of cooling and the rate of heating available to the selective hydrogenation reactor by the machine learning model based on the weather data, wherein the machine learning model has been trained on historical data of the operating parameters and resulting product attributes and composition of the gasoline desulfurization system and an associated historical weather data.

198. The method of any of claims 180 to 197, wherein the simulation includes a changing rate over time to an operating temperature of the selective hydrogenation reactor based onwhen the feedstock corresponding to the feedstock data will be fed into the selective hydrogenation reactor of the gasoline desulfurization system.

199. The method of any of claims 197 to 198, further comprising determining, by the machine learning model, from the weather data that a change in weather above a predetermined threshold has occurred, wherein when the machine learning model has determined the change in weather above a predetermined threshold has occurred, the simulation is updated.

200. The method of any of claims 180 to 199, wherein the simulation includes a change to an operating temperature of the hydrodesulfurization reactor.

201. The method of claim 200, wherein the simulation includes a changing rate over time to an operating temperature of the hydrodesulfurization reactor.

202. The method of any of claims 180 to 201, wherein the simulation is based on a rate of cooling and a rate of heating available to the hydrodesulfurization reactor.

203. The method of any of claims 180 to 202, wherein the simulation to the operating parameter includes a change to feed rates of hydrogen at different locations within the hydrodesulfurization reactor based on temperature variations within the hydrodesulfurization reactor.

204. The method of any of claims 180 to 203, wherein the simulation to the operating parameter includes a change to the operating parameter includes reducing an inlet temperature of the hydrodesulfurization reactor and increasing the operating temperature of a finishing reactor.

205. The method of any of claims 180 to 204, wherein the feedstock includes naphtha from a deisopentanizer, the method further comprising adjusting a percentage of desulfurized light naphtha split from the resulting product of the selective hydrogenation reactor based on the percentage of the feedstock that is naphtha from the deisopentanizer.

206. The method of any of claims 180 to 205, wherein the simulation further includes splitting hydrogen sulfide gas from a desulfurized heavy naphtha from the hydrodesulfurization reactor.

207. The method of any of claims 180 to 206, wherein the machine learning model is trained on the historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, the operating parameters of the splitter after the selective hydrogenation reactor, the feedstock attributes, and the attributes and properties of the produced desulfurized heavy naphtha product, the method further comprising predicting the operating parameters to control the hydrodesulfurization reactor to reduce overtreatment by the hydrodesulfurization reactor based on the composition of feedstock being fed into the hydrodesulfurization reactor.

208. The method of any of claims 180 to 207, wherein the machine learning model is trained on the historical data of the operating parameters of the hydrodesulfurization reactor, the operating parameters of the selective hydrogenation reactor, feedstock properties, and the attributes and properties of the produced desulfurized light naphtha product and the desulfurized heavy naphtha product, the method further comprising: receiving business data, the business data including an incentive, profit, or price for each octane barrel of the desulfurized light naphtha product and the desulfurized heavy naphtha product; and generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) + (PGHN flow)(PGHN R0N)($ / PGHN Octane-BBL) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)209. The method of claim 208, wherein the business data includes a sulfur credit cost, wherein generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Gasoline Desulfurization System Value = (PGLN flow)(PGLN RON)($ / PGLN Octane- BBL Incentive) - (PGLN flow)(PGLN Sulfur)(Sulfur Credit Cost) + (PGHN flow)(PGHNRON)($ / PGHN Octane-BBL) - (PGHN flow)(PGHN Sulfur)(Sulfur Credit Cost) - (Makeup H2 Flow)($ / SCF Hydrogen Cost)210. The method of any of claims 180 to 208, further comprising generating the operating parameters of the selective hydrogenation reactor and the hydrodesulfurization reactor based on:Overall Sulfur target=[(PGLN flow)*PGLN sulfur+(PGHN flow)*PGHN sulfur] / (PGHN flow +PGLN flow)211. A memory including instructions and a machine learning model that causes a processor to perform the instructions including the method of any of claims 180 to 210.

212. A computing system including a processor and a memory including instructions and a machine learning model that cause the processor to perform the instructions, wherein the instructions include the method of any of claims 180 to 210.

213. A system for enhancing fluid production for a gasoline desulfurization operation, the system comprising: a gasoline desulfurization unit to receive a feed and produce one or more fluids; a plurality of sensors to measure a parameter associated with the gasoline desulfurization unit and each positioned at one of (a) proximate the gasoline desulfurization unit or (b) within the gasoline desulfurization unit; a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the gasoline desulfurization unit and to control aspects of fluid flowing to or from the gasoline desulfurization unit; one or more sample collection assemblies to collect samples of the fluid associated with the gasoline desulfurization unit; one or more sample analysis assemblies to analyze each collected sample to provide properties of collected samples; and a gasoline desulfurization controller in signal communication with the gasoline desulfurization 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 gasoline desulfurization controller configured to:determine an output including predicted properties of a feedstock and parameter settings of the refinery operation control device and the gasoline desulfurization 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 the amount of feed or type of feed and parameters associated with the refinery operation control device and gasoline desulfurization unit based on the output to enhance fluid production.

214. The system of claim 213, wherein the machine learning model has been trained on historical data of measured parameters associated with the gasoline desulfurization unit, historical data of refinery operation control devices, and sensor and sample data from the gasoline desulfurization system.

215. The system of claim 214, wherein the historical data of the sensor and sample data from the gasoline desulfurization system include sulfur concentration and octane number.

216. The system of any of claims 213 to 215, wherein the target product and corresponding target properties include a light naphtha sulfur target, the gasoline desulfurization controller predicting octane-barrels of a light naphtha product based on the light naphtha sulfur target.

217. The system of claim 216, wherein the gasoline desulfurization controller adjusts one or more of an the amount of feed or type of feed and parameters associated with the refinery operation control device and gasoline desulfurization unit to produce the predicted octane-barrels of the light naphtha product based on the light naphtha sulfur target.

218. The system of claim 217, wherein the gasoline desulfurization controller adjusts a reactor’s temperature to produce the predicted octane-barrels of the light naphtha product based on the light naphtha sulfur target.

219. The system of any of claims 213 to 218, wherein the target product and corresponding target properties include a heavy naphtha sulfur target, the gasolinedesulfurization controller predicting octane-barrels of a heavy naphtha product based on the heavy naphtha sulfur target.

220. The system of claim 219, wherein the gasoline desulfurization controller adjusts one or more of an the amount of feed or type of feed and parameters associated with the refinery operation control device and gasoline desulfurization unit to produce the predicted octane-barrels of the heavy naphtha product based on the heavy naphtha sulfur target.

221. The system of any of claims 219 to 220, wherein the gasoline desulfurization controller adjusts a reactor’s temperature to produce the predicted octane-barrels of the heavy naphtha product based on the heavy naphtha sulfur target.

222. The system of any of claims 219 to 221, wherein the target product and corresponding target properties include the light naphtha sulfur target, the heavy naphtha sulfur target, and an overall average sulfur target for the combined properties of the light naphtha product and the heavy naphtha product while enhancing overall octane-barrels of the combined properties of the produced light naphtha product and the heavy naphtha product.

223. A system for enhancing fluid production of a gasoline desulfurization operation, the system comprising: a gasoline desulfurization controller in signal communication with a gasoline desulfurization unit, a plurality of sensors, a plurality of refinery operation control devices, and storing a machine learning model on historical data of the refinery operations, including gasoline desulfurization unit operations, resulting properties of produced light naphtha product and heavy naphtha product from the gasoline desulfurization unit operations, operations upstream of the gasoline desulfurization unit, and feedstock properties, the gasoline desulfurization controller configured to: receive a light naphtha sulfur target for the produced light naphtha product and a heavy naphtha sulfur target for the produced heavy naphtha product; generate, by the machine learning model, an output including predicted properties of the light naphtha product and the heavy naphtha product and parameter settings of the gasoline desulfurization unit based on (i) data measured by a plurality of sensors disposed to measure upstream operations processing the feedstock, (ii)data corresponding to analysis from one or more sample analysis assemblies of a feedstock sample, and (iii) the light naphtha sulfur target and the heavy naphtha sulfur target; and adjust one or more parameters associated with the gasoline desulfurization unit the output of parameter settings.

224. The system of claim 223, wherein the historical data of the refinery operations include sulfur concentration and octane number of the produced light naphtha product and the heavy naphtha product.

225. The system of any of claims 223 to 224, wherein the output includes a prediction of octane-barrels of the light naphtha product based on a light naphtha sulfur target.

226. The system of claim 225, wherein the gasoline desulfurization controller adjusts one or more parameters associated with the gasoline desulfurization unit the output of parameter settings to produce the predicted octane-barrels of the light naphtha product based on the light naphtha sulfur target.

227. The system of any of claims 225 to 226, wherein the gasoline desulfurization controller adjusts a temperature of a selective hydrogeneration reactor of the gasoline desulfurization unit to produce the predicted octane-barrels of the light naphtha product based on the light naphtha sulfur target.

228. The system of any of claims 223 to 227, wherein the output includes a prediction of octane-barrels of the heavy naphtha product based on a heavy naphtha sulfur target.

229. The system of claim 228, wherein the gasoline desulfurization controller adjusts one or more parameters associated with the gasoline desulfurization unit the output of parameter settings to produce the predicted octane-barrels of the heavy naphtha product based on the heavy naphtha sulfur target.

230. The system of any of claims 228 to 229, wherein the gasoline desulfurization controller adjusts a temperature of a hydrodesulfurization reactor of the gasolinedesulfurization unit to produce the predicted octane-barrels of the heavy naphtha product based on the heavy naphtha sulfur target.

231. The system of any of claims 223 to 230, wherein the target product and corresponding target properties include the light naphtha sulfur target, the heavy naphtha sulfur target, and an overall average sulfur target for combined properties of the produced light naphtha product and the heavy naphtha product while enhancing overall octane- barrels of the combined properties of the produced light naphtha product and the heavy naphtha product.I l l

Images