Systems, analyzers, controllers, and associated methods of operating a resid destruction unit

Abstract

Embodiments of systems and methods for operating a resid destruction unit include generating an input comprising target parameters and one or more sensor outputs from one or more sensors disposed in and / or proximate the resid destruction unit, one or more sample analyses from sample collection and analysis assemblies disposed in and / or proximate the resid destruction unit. A machine learning model may be applied to the input to generate an output comprising one or more predicted parameters. A machine learning model may be applied to the predicted parameters and the target parameters to generate an output indicative of operating conditions of the resid destruction unit achieve the target parameters. Related methods and systems are also disclosed.

Description

SYSTEMS, ANALYZERS, CONTROLLERS, AND ASSOCIATED METHODS OFOPERATING A RESID DESTRUCTION UNITCROSS-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 the destruction of resid and / or enhance fluid production of one or more fluids associated with a resid destruction system, such as gas oil, naphtha, distillate, and liquified petroleum gases (LPGs) using machine learning models during the refining operations and sub-operations associated with the resid destruction system.BACKGROUND1CT

[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. Furthermore, a variety of starting feedstock are utilized at a refinery and 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 process changes to one operation affects other (e.g., downstream) operations 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) unit operations. However, those controllers and monitoring devices utilize algorithms that require expert personnel and 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, be exposed to different environmental conditions, and / or may otherwise differ from 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, such resid destruction. 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 and / or to enhance the destruction of resid to enhance the fluid production. Such fluids may includehydrocarbons 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, as well as enhancing resid destruction and the formation of more valuable fluids. 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 system for enhancing fluid production for refining operations. The system may include a plurality of refining equipment each configured with refining equipment parameter settings to perform a refining sub-operation of a refinery operation on a flow of feedstock thereto. The system may include a plurality of sensors to measure a sensor parameter associated with the plurality of refining equipment and each one of the plurality of sensors positioned at one of: (a) proximate one of the plurality of refining equipment or (b) within one of the plurality of refining equipment. The system may include a plurality of refining operation control devices, each one of the plurality of refining operation control devices positioned proximate one of the plurality of the refining equipment and being either downstream or upstream thereof and having parameters to control an aspect and / or property of feedstock flowing to one of the plurality of refining equipment or a product flowing from one of the plurality of refining equipment. The system may include one or more sample collection assemblies to collect samples of the feedstock and products associated with each of the plurality of refining equipment. 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 plurality of operation controllers, each one of the plurality of operation controllers in signal communication with one of a plurality of subsets of the plurality of refining equipment, one of a plurality of subsets of the plurality of sensors, one of a plurality of subsets of the plurality of refining operation control devices, and the one or more sample analysis assemblies. Each of the plurality of operation controllers may include: (i) a plurality of predictive controls circuitries each storing one or more selected models3CT(machine learning models), the plurality of predictive controls circuitries configured to: determine an output based on application of data from the one of the plurality of subsets of the plurality of refining equipment, the one of the plurality of subsets of the plurality of sensors, the one of the plurality of subsets of the plurality of refining operation control devices, and target products and corresponding properties to the corresponding one or more selected models; (ii) a local enhancement circuity storing a machine learning model and configured to: in response to reception of the output from one or more of the plurality of predictive controls circuitries, determine one or more target setpoints based on application of the output and the data from the one of the plurality of subsets of the plurality of refining equipment, the one of the plurality of subsets of the plurality of sensors, the one of the plurality of subsets of the plurality of refining operation control devices to the corresponding one or more selected models; and (iii) a plurality of equipment and device controllers, each of the equipment and device controllers in signal communication with the local enhancement circuitry and configured to: in response to a determination of one or more target setpoints, adjust the one of the plurality of subsets of the plurality of refining equipment and the one of the plurality of subsets of the plurality of refining operation control devices to the one or more target setpoints.

[0008] Some embodiments of the disclosure are directed to a system for enhancing fluid production for a distillation operation. The system may include a distillation column to receive a feedstock and separate the feedstock into a plurality of fluids. The system may include a plurality of sensors to measure a parameter associated with the distillation column and each positioned at one of: (a) proximate the distillation column; or (b) within the distillation column. The system may include a plurality of refining operation control devices each positioned proximate and downstream or upstream of the distillation column and to control an aspect and / or property of fluid flowing to or from the distillation column. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the distillation column. The system may include one or more sample analysis assemblies to analyze each collected sample to provide properties of the collected samples. The system may include a distillation controller in signal communication with the distillation column, the plurality of sensors, the plurality of refining operation control devices, and the one or more sample analysis assemblies, and storing a trained machine learning model. The distillation controller may be configured to determine an output including predicted properties of the feedstock and parameter settings of the plurality of refining operation control devices and the distillation column based onapplication of one or more of: (i) data measured by the plurality of sensors; (ii) data corresponding to analysis from the one or more sample analysis assemblies; or (iii) a target product and corresponding target properties to the trained machine learning model; and adjust one or more of the amount of feedstock or type of feedstock and parameters associated with the plurality of refining operation control devices and distillation column based on the output to enhance production of the plurality of fluids.

[0009] Another embodiment of the disclosure is directed to a system for enhancing fluid production for a coker operation in a coker unit. The system may include a plurality of coke drums to receive heavy distillate, resid materials, or other fluids and each of the plurality of coke drums configured with refining equipment parameter settings to convert the heavy distillate, resid, or other fluids. The system may include a plurality of sensors to measure a parameter associated with the coker unit and each positioned at one of: (a) proximate the coker unit; and / or (b) within the coker unit. The system may include a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the coker unit and to control aspects of fluid flowing to or from the coker unit. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the coker 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 coker controller in signal communication with the coker 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 coker controller may be configured to: (1) determine an output including predicted properties of the heavy distillate, resid, or other fluids and parameter settings of the plurality of refinery operation control devices and the 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 (2) adjust one or more of the amount of heavy distillate, resid, or other fluids or type of heavy distillate, resid, or other fluids and parameters associated with the plurality of refinery operation control devices and coker unit based on the output to enhance production of fluid from the coker unit.

[0010] Another embodiment of the disclosure is directed to a system for enhancing fluid production for a solvent deasphalting (SDA) operation. The system may include an SDA unit to receive a feedstock and produce a fluid (e.g., deasphalted oil) and deasphalted asphalt (e.g., pitch). The system may include a plurality of sensors to measure a parameter5CTassociated with the SDA unit and each positioned at one of: (a) proximate the SDA unit; or (b) within the SDA unit. The system may include a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the SDA unit and to control aspects of fluid flowing to or from the SDA unit. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the SDA 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 an SDA controller in signal communication with the SDA 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 SDA 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 SDA 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 (2) adjust one or more of the amount of feedstock or type of feedstock and parameters associated with the plurality of refinery operation control devices and SDA unit based on the output to enhance the production of fluid from the SDA unit.

[0011] Another embodiment of the disclosure is directed to a controller to enhance fluid production of a refining operation. The controller may include a first plurality of inputs each in signal communication with one of a plurality of sensors to measure a set of first parameters associated with aspects of a refining operation and one or more refining suboperations. The controller may include a second plurality of inputs each in signal communication with one or more analyzers to analyze and provide properties of samples of fluids input to and output from each of a plurality of refining equipment. The controller may include a first plurality of inputs / outputs each in signal communication with one of the plurality of refining equipment, the controller configured to receive a set of second parameters associated with each of the plurality of refining equipment and to transmit instructions and selected parameters to cause each of the plurality of refining equipment to operate at the selected parameters. The controller may include a second plurality of inputs / outputs each in signal communication with one of a plurality of sub-controllers each configured to control one of the plurality of refining sub-operations. The controller may be configured to apply one or more of the set of first parameters, the set of second parameters, or the properties to a trained machine learning model, and determine an adjusted one ormore inputs, intermediaries, or operating parameters based on application of data received from the plurality of sensors, the one or more analyzers, the plurality of refining equipment, and the plurality of sub-controllers; and adjust, via the sub-controllers, a type and amount of refining equipment fluid inputs and refining equipment parameters based on the adjusted one or more inputs, intermediaries, or operating parameters.

[0012] Another embodiment of the disclosure is directed to a system for enhancing fluid production for a resid destruction operation. The system may include a resid destruction unit to receive a feed, which may include one or more sources of residuum (also referred to as “resid” herein), and produce one or more fluids. The system may include a plurality of sensors to measure a parameter associated with the resid destruction unit and each positioned at one of: (a) proximate the resid destruction unit; and / or (b) within the resid destruction unit. The system may include a plurality of refinery operation control devices each positioned proximate and downstream or upstream of the resid destruction unit and to control aspects of fluid flowing to or from the resid destruction unit. The system may include one or more sample collection assemblies to collect samples of the fluid associated with the resid destruction 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 resid destruction controller in signal communication with the resid destruction 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 resid destruction 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 resid destruction 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 (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 resid destruction unit based on the output to enhance fluid production.

[0013] 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.7CTAccordingly, 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

[0014] 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.

[0015] 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.

[0016] FIG. 2 is a simplified diagram that illustrates an apparatus for enhanced fluid production at a refinery, according to an embodiment of the disclosure.

[0017] FIG. 3 is another simplified diagram of a refining control system to enhance fluid production at a refinery, according to an embodiment of the disclosure.

[0018] 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.

[0019] FIG. 5 is a schematic diagram of a refinery, according to an embodiment of the disclosure.

[0020] FIG. 6A is a schematic diagram of a resid destruction unit including a resid destruction control system to enhance fluid production at a portion of a refinery, according to at least one embodiment of the disclosure.

[0021] FIG. 6B is a simplified schematic diagram of a coker unit including a coker controller, according to at least one embodiment of the disclosure.

[0022] FIG. 6C is a schematic diagram of a Solvent Deasphalting Unit (SDA) control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure.

[0023] FIG. 7 is a schematic diagram of a machine learning model that may be used with the various processes within a refinery, according to at least one embodiment of the disclosure.

[0024] FIG. 8A and FIG. 8B are simplified diagrams of control systems to enhance to enhance fluid production at refinery, according to an embodiment of the disclosure.

[0025] FIG. 9 is a flow chart illustrating enhanced fluid production at a refinery, according to an embodiment of the disclosure.

[0026] FIG. 10 is a simplified flow diagram of a method of operating a resid destruction unit, according to at least one embodiment of the disclosure.DETAILED DESCRIPTION

[0027] 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.

[0028] As used herein, a level of a coke drum refers to a percent of the coke drum that is filled with coke and may refer to a percentage of the available volume (or height) of the coke drum in which coke can be deposited based on a process limit, equipment safety limit, or other predefined limit for a 100 percent filled drum. For example, in some instances, a coke drum may be defined as full (a level of 100 percent) when the level of coke in the coke drum is about 15 feet from the top. The level of the coke drum may refer to the9CTpercentage of the coke drum that is filled with coke with respect to the predefined full level. The level or volume of coke that may be help within a coke drum may be specific to the particular coke drum (e.g., based on the dimensions of the particular coke drum).

[0029] As used herein, a cycle time of a coker means and includes a total duration required to complete one full operational cycle for a coke drum to be filled with hot residuum where the resid is thermally cracked into lighter hydrocarbons and solid coke; switching between coke drums where the coke drum that is filled with coke is taken out of fluid communication with the resid; steam-out and blowdown wherein residual hydrocarbons are purged from the filled coke drum; drum cooling where the filled coke drum is cooled; and coke cutting where the coke in the filled coke drum is removed.

[0030] 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.

[0031] 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. Non-limiting examples of fluids used herein may refer to a liquids, gases, vapors, or other materials. Some materials may be solid at room temperature, but may be flowable liquids or viscous materials at higher temperatures. The fluid 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 acombination of hydrocarbons having different boiling points that exist as a viscous liquid in underground geological formations and at the surface. Petroleum products, for example, may be produced by processing crude oil and other liquids at petroleum refineries, by extracting liquid hydrocarbons at natural gas processing plants, and by producing finished petroleum products at industrial facilities. For example, a petroleum product may include a transportation fuel, among other products. Refinery intermediates, for example, may refer to any refinery hydrocarbon that is not crude oil or a finished petroleum product (such as gasoline), including all refinery output from distillation (for example, distillates or distillation fractions) or from other conversion units. In some non-limiting embodiments of systems and methods, petroleum fluids may include heavy blend crude oil used at a pipeline origination station, natural gas, and / or other types of crude oil, as will be understood by one skilled in the art. Heavy blend crude oil is typically characterized as having an American Petroleum Institute (API) gravity of about 30 degrees or below. In other embodiments, the petroleum fluids may include lighter blend crude oils, for example, having an API gravity of greater than 30 degrees. “Renewable fluids” as used herein, may refer to fluid products containing plant and / or animal derived feedstock. Further, the renewable fluids may be hydrocarbon based. For example, a renewable fluid may be a pyrolysis oil, oleaginous feedstock, biomass derived feedstock, 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.

[0032] In some embodiments, the disclosure is directed to systems, analyzers, controllers, and associated methods for enhancing fluid production from a resid destruction system using one or more machine learning models. The resid destruction unit may be configured to process one or more feedstock resid materials (e.g., vacuum tower bottoms materials (e.g., high viscosity vacuum tower bottoms material), pitch, resid from tankage (e.g., purchased resid), or a resid hydrocracker bottoms material) to destruct the feedstock resid materials and form more valuable hydrocarbons, such as gas oil (e.g., heavy gas oil) that may be used as feed to another unit (e.g., a fluid catalytic cracker), distillate (e.g., diesel), naphtha, and LPGs. The resid destruction unit may include a plurality of resid destruction sub-units or sub-operations, each resid destruction sub-unit configured to receive a feedstock resid material and convert at least a portion of the feedstock resid material into more valuable products, such as gas oil, deasphalted oil, distillate, naphtha, and LPGs. The resid destruction sub-units may include one or more of (e.g., each of) a coker and a solvent11CTdeasphalting unit. In some embodiments, the resid destruction unit comprises at least two coker units.

[0033] The feedstock resid materials may exhibit different properties, depending on the source (e.g., the sub-unit) form which the feedstock resid material originates. The feedstock resid materials may include, for example, a first vacuum tower bottoms material, a second vacuum tower bottoms material, pitch (deasphalted asphalt), resid hydrocracker bottoms, resid from tankage, or purchased resid. Each of the feedstock resid materials may exhibit a different flow rate and one or more different properties. For example, each of the feedstock resid materials may exhibit different properties based on where the respective feedstock resid material originates (e.g., operating conditions of the sub-unit from which the respective feedstock resid material is provided to the resid destruction unit). By way of non-limiting example, the respective feedstock resid materials may exhibit a different coking tendency, which may depend at least in part, on a carbon residue of the respective feedstock material. As used herein, the carbon residue of a material refers to the amount (e.g., the concentration, such as the weight percent) of the carbonaceous material that remains after volatile components have been removed through heating and may be a method of measuring the tendency of the material to form carbon deposits, such as coke. The carbon residue may also be referred to as the carbon residue concentration.

[0034] As described above, the resid destruction unit may include at least two coker units. Each coker unit may include a furnace configured to heat a charge resid material and provide the charge resid material to a coke drum. Each coker unit may include at least two coke drums in fluid communication with the respective furnace and configured to facilitate cracking of the charge resid material to form lighter components and destroy at least a portion of the resid. A portion of the charge resid material may form coke, which may be deposited in the coke drum. Vapors from the coke drum are received by a fractionator, where they are separated based on boiling point to form LPGs, naphtha, distillate, and gas oils.

[0035] One challenge faced in resid destruction units is optimization of the resid destruction unit to increase the throughput of the feedstock resid materials through the resid destruction unit. Some coker units may be limited (rate limited) by a total feed rate to the coker unit (which may be defined by a process safety management (PSM) limit) and other coker units may be limited (rate limited) by a coke drum fill rate (e.g., the time it takes for the coke drum to fill with coke during operation of the coker unit while the resid charge is fed to the coke drum). When two coker units are run in parallel or at the same time, one orboth of the coke drums of the different coker units may not achieve a target level of deposited coke during a cycle time, resulting in a non-optimal destruction of the feedstock resid materials.

[0036] According to embodiments described herein, the resid destruction unit includes a resid destruction controller configured to enhance the destruction of the feedstock resid materials and form more valuable products than the feedstock resid materials. The resid destruction controller may be configured to determine an optimal flow rate and composition of feedstock resid materials to provide to each of the coker units, the SDA unit, and the external resid destruction unit of the resid destruction unit to enhance the destruction of the feedstock resid materials, such as by improving the throughput of the feedstock resid materials through the resid destruction unit. The resid destruction controller may be configured to determine a flow rate of the feedstock resid material from each of a plurality of sources of the feedstock resid materials to provide to each of the resid destruction subunits to enhance the destruction of the resid. Based on the composition of each of the feedstock resid materials from the respective sources of the feedstock resid materials, and the flow rate of the feedstock resid material from each of the feedstock resid sources to the different resid destruction sub-units, the composition of resid provided to each of the resid destruction sub-units may be different. Thus, the feedstock resid materials provided to each of the plurality of sub-units may have a different composition, one or more different properties, and / or a different flow rate than one another. The resid destruction controller may be configured to determine one or more of the flow rate, the composition, or one or more properties of the feedstock resid material to each of the sub-units to enhance operation of the resid destruction unit, the flow rate of resid feed to a first coke drum of a first coker unit (a charge rate of the resid feed to the first coke drum, also referred to as a coke drum charge to the first coke drum), and the flow rate of resid feed to a second coke drum of a second coker unit (a charge rate of the resid feed to the second coke drum, also referred to as a coke drum charge to the second coke drum).

[0037] As a specific non-limiting example, the resid destruction controller may be configured to determine a flow rate of resid material to the resid destruction unit comprising a plurality of coker units (e.g., at least two coker units), a solvent deasphalting unit (SDA), and optionally, an external resid destruction unit, the composition of the resid material to the resid destruction unit, and the coke drum charge to provide to each of at least a first coker unit and a second coker unit to cause the coke drums of the respective two coker units to fill during their respective cycle times such that the coke drum of each of the two coker13CTunits are filled at the end of their respective cycle times such that the unused volume of the coke drums in fluid communication with the resid feed is minimized and the volume of resid that is processed in the resid destruction unit including multiple coker units is optimized. In other words, the resid destruction controller may be configured to determine a flow rate and a composition of the resid material to the resid destruction unit and the coke drum charge to provide to each of the first coke drum of the first coker unit and the second coke drum of the second coker unit such that the first coke drum and the second coke drum are filled at the end of their respective cycle times and the coke drums do not include any unused volume that are not filled with coke (e.g., both coke drums include coke up to a suitable level). The resid destruction controller coordinates the stated parameters to the coker units and the coker units of the resid destruction unit.

[0038] One or more sensors may be disposed throughout the resid destruction unit and / or sub-units that provide the feedstock resid materials to the resid destruction unit. The sensors may be configured to generate sensor data indicative of one or more properties or conditions within the resid destruction unit and / or one or more properties or a composition of the feedstock resid materials and / or other fluids within the resid destruction unit and / or the refinery (such as fluids in sub-units upstream of the resid destruction unit).

[0039] By way of non-limiting example, the sensor data may include one or more of an operating condition (a temperature, a pressure) of one or more components of the resid destruction unit, a flow rate of feedstock resid materials from each of a plurality of resid sources (e.g., a flow rate of a first vacuum tower bottoms, a flow rate of a second vacuum tower bottoms, a flow rate of an atmospheric tower bottoms, a flow rate of pitch, a flow rate of a resid hydrocracker bottoms, a flow rate of resid from tankage), a flow rate of resid to each of a plurality of resid destruction sub-units (to each of the coker units, the SDA unit, and the external resid destruction unit), a composition of resid to each of the plurality of resid destruction sub-units, a carbon residue concentration of each of the plurality of resid sources, a carbon residue concentration of the resid to each of the plurality of resid destruction sub-units, a density and / or a viscosity of each of a plurality of resid source, a density and / or a viscosity of the resid to each of the plurality of resid destruction sub-units, operating conditions within each of the resid destruction sub-units, and operating conditions of units upstream of the resid destruction unit (such as units that provide the various feedstock resid materials to the resid destruction unit).

[0040] The resid destruction unit and / or the refinery may include one or more sample collection and analysis assemblies or systems for obtaining one or more samples of a fluidfrom the resid destruction unit (e.g., a feedstock resid material from each of the resid sources, an intermediate stream, the resid to each of the resid destruction sub-units) and analyzing the samples to measure one or more properties thereof to generate sample data. One or more sample analysis assemblies may be configured to analyze the collected samples to measure, determine, and / or provide properties of the collected samples. In some embodiments, the one or more samples are analyzed in a laboratory. In some embodiments, the one or more samples (or at least one of the one or more samples) are analyzed using an in-line sensor, such as an in-line spectrometer. The one or more sample collection analysis assemblies may be configured to generate one or more properties of one or more of respective sample.

[0041] The resid destruction controller may be in operable (e.g., signal) communication with the one or more sensors and the one or more sample collection and analysis assemblies and configured to receive the sensor data and data from the one or more sample collection and analysis assemblies. The resid destruction controller may include a machine learning model configure to receive input data comprising the sensor data from the one or more sensors and the sample data from the one or more sample collection and analysis assemblies. In some embodiments, the input data includes a target parameter(s) comprising at least one of target coke drum fill time (also simply referred to herein as a “target fill time”), and a target level for each of a first coke drum of the first coker unit and a second coke drum of the second coker unit. The target fill time of a coke drum herein may correspond to and also be referred to herein as a target cycle time. In some embodiments, the input data to the machine learning model includes the data from the sensors and the analysis data (e.g., composition, property) with respect to one or more of the feedstock resid materials, the resid materials to each of the resid destruction sub-units or the intermediate streams,. The machine learning model may be referred to herein as a “resid destruction unit controller model.”

[0042] The resid destruction controller may include a machine learning model configured to, based on the inputs, generate an output comprising one or more predicted parameters, such as one or more of a predicted coke drum fill rate of each of the first coke drum and the second coke drum, a predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit and the second coker unit, at least one of a predicted flow rate or composition of pitch to each of the first coker unit and the second coker unit, a predicted flow rate of a first resid feed to the first coke drum (e.g., a predicted charge rate of the first resid feed to the first coke drum), and a predicted flow rate of a second resid15CTfeed to the second coke drum (e.g., a predicted charge rate of the second resid feed to the second coke drum). It should be noted that the predicted flow rate of each of the plurality of feedstock resid materials to each coker unit determines the total flow rate and composition for that coker unit. In some embodiments, the predicted parameters include a predicted composition of the first resid feed to the first coke drum and a predicted composition of the second resid feed to the second coke drum. The inputs may include one or more properties of each of the plurality of feedstock resid materials, a flow rate of a first resid material to the first coke drum, a flow rate of a second resid material to the second coke drum, or one or more operating conditions of the resid destruction system to achieve the target parameters. In some embodiments, the predetermined parameter comprises a coke drum fill rate for each of the first coke drum and the second coke drum. In some embodiments, based on the predicted coke drum fill rate for each of the first coke drum and the second coke drum, a second machine learning model of the resid destruction controller may generate an output comprising one or more operating conditions of the resid destruction unit to generate the target parameters. Based on the output, the resid destruction system may be configured to adjust one or more operating conditions of the resid destruction unit and / or one or more operating conditions of upstream units providing feedstock resid materials to the resid destruction unit. The machine learning model(s) may be trained to generate the output based on the input. In some embodiments, the machine learning model may be trained using historical data from the resid destruction unit, such as historical data from the one or more sensors and historical data from samples obtained from the one or more sample analysis assemblies.

[0043] In some embodiments, the resid destruction unit controller includes at least two machine learning models. For example, a first machine learning model may be applied to the input data to generate an output comprising the one or more predicted parameters. The first machine learning model may be trained with training data comprising first training inputs comprising the sensor output (also referred to as “sensor data”), data from the sample analysis assemblies during a training period; and first training outputs comprising the fill rate of the first coke drum (the coke drum fill rate of the first coke drum) and the fill rate of the second coke drum (the coke drum fill rate of the second coke drum) during the training period. In some embodiments, the resid destruction unit controller includes a plurality of the first machine learning models, each configured to be applied to the input data to generate the predicted parameters. Each first machine learning model may comprise a part of a predictive control and may be associated with a particular piece of equipment ofthe resid destruction unit and / or a particular portion of the resid destruction unit. For example, each of the resid destruction sub-units may include at least one machine learning model configured to generate an output comprising one or more predicted parameters associate with the respective resid destruction sub-unit.

[0044] A second machine learning model may be configured to receive one or more inputs comprising the output from the first machine learning models; the sensor outputs, the data from the sample analysis assemblies; and the target parameters. Based on the inputs to the second machine learning model, the second machine learning model may be configured to generate an output comprising changes in the one or more operating conditions of the resid destruction unit and / or one or more upstream processes (e.g., to generate a desired feedstock resid composition) to achieve the target parameters and / or relative changes to the fill rate of the first coke drum or the fill rate of the second coke drum caused by the changes in the one or more operating conditions. In some embodiments, the one or more operating conditions include one or more of the coke drum fill rate of each of the first coke drum and the second coke drum, the flow rate of each of the plurality of feedstock resid materials to each of the first coker unit and the second coker unit, at least one of the flow rate or the composition of pitch to each of the first coker unit and the second coker unit, the flow rate of the first resid to the first coke drum (the coke drum charge to the first coke drum), or the flow rate of the second resid to the second coke drum (the coke drum charge to the second coke drum). Thus, the resid destruction unit controller may be configured to facilitate operation of two coker units such that the coke drums of each of the coker units reach a level of about 100 percent or at least about a threshold level (e.g., at least about 90 percent, at least about 95 percent) at the end of a cycle such that respective resid feeds are provided to the coke drums during their cycle time to optimize the amount of resid destroyed during the cycle time. The feedstock resid materials may include a first vacuum tower bottoms, a second vacuum tower bottoms, a resid hydrocracker tower bottoms, pitch, and resid from one or more external sources.

[0045] In some embodiments, the resid destruction controller comprises a local enhancement module includes an algorithm configured to facilitate optimization of the resid destruction unit to achieve the target parameters (e.g., the objective function) based on the outputs from the first machine learning models (e.g., of predictive controls modules of the resid destruction controller), the target parameters, unit constraints, and inputs. The algorithm may be configured to facilitate optimization of the resid destruction unit (e.g., destruction of the resid) based on a machine learning model. The local enhancement17CTmodule may control the refinery operation control devices based on the outputs and constraints of the resid destruction unit (e.g., the resid destruction sub-units). In some embodiments, the local enhancement module includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g. maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules), the target parameters, unit constraints, and inputs. The local enhancement module may control the refinery operation control devices based on the outputs and constraints of the resid destruction unit (e.g., the resid destruction sub-units).

[0046] In an embodiment, the systems and methods may include a computing device, apparatus and / or controller (referred to hereafter as a controller) to obtain various data points (which may form a data set) and / or parameters to train a machine learning model. Such data may include a historical data set and / or a currently generated data set including an outcome. The outcome may include, for example, the fill rate of the first coke drum, the fill rate of the second coke drum, the flow rate of the first resid to the first coke drum, or the flow rate of the second resid to the second coke drum, a flow rate of the feedstock resid materials to the resid destruction unit, a flow rate of resid to each of the resid destruction sub-units, a composition of the resid to each of the resid destruction sub-units, operating conditions of a resid destruction unit, and / or operating conditions of each of the resid destruction sub-units. 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 may be used or generated to simulate a data set. In another embodiment, the data may include a partial data set. In such an embodiment, the partial data set, may be filled in via a first principle model and / or a machine learning model. Each data set may include a series of parameters, properties, spectra, and / or other data points associated with a refining operation or process or sub-operation or sub-process. Each data set may also include target parameters, target properties, and / or an outcome and / or target product. Further, in an embodiment where a supervised machine learning model is utilized, 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 basedon that maximum or minimum, or based on some other factor in other embodiments. In yet another embodiment, a trained machine learning model may learn or be trained based on trends included in the data (in other words, the trained machine learning model may comprise a deep learning model). In another embodiment, any of the trained machine learning models described herein may predict and / or optimize target parameters and / or fluids used within a refinery, refinery operation, or refinery sub-operation.

[0047] Once these data sets have been received by the controller (e.g., the resid destruction unit 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, shutdown, turn-arounds, and / or upsets), remove undesired data, remove invalid measurements, and / or segregate the data set into sequences of contiguous data based on selected time intervals (for example, time intervals of 30 minutes, 1 hour, 2 hours, and / or 3 hours, or more or less than the time intervals listed).

[0048] Once the data has been pre-processed, the controller may begin training a model based on a portion of the data set. For example, the controller may utilize an 80 / 20 training and testing process. Other percentages may be utilized in training, testing, and / or validation. As the model is fed data, the model may compare data received to the outcome (in other words, whether the outcome was desired based on some factor, such as an indicated positive / negative flag or classification, based on some maximum or minimum of a function generated based on the data, or based on a trend within the data). Once the training portion of data has been utilized, the controller may test and / or validate the model using the 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 reached (achieved) that threshold, then the controller may output the trained machine learning model for further use.

[0049] Further, a trained machine learning model may be further refined using new data, as such data is generated. Such a refinement may occur while the trained machine 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 testing19CTand / or error rating meets a selected threshold, in addition to other factors, then a controller may replace the version currently utilized to the refined version.

[0050] It will be understood that such systems and methods described herein may utilize a number of trained machine learning models or classifiers (also referred to as trained models or classifiers). 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, hydrocracker operations, other blending operations, solvent deasphalting (SDA) operation, solvent deasphalting operations wherein a portion of the operation (the deasphalted oil stripper) is performed at supercritical conditions of the solvent (e.g., a supercritical solvent deasphalting 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 subprocess). The use of terms operation and process refers to the steps taken to produce a particular product from a selected feedstock (and, in some embodiments, other inputs). As such, when referring to a particular refining operation or process, the terms “operation” and “process” may be used interchangeably. Further, such models may be trained specifically for equipment at a particular plant or refinery. For example, a resid destruction unit at a first plant may exhibit different characteristics than that of a resid destruction unit at a second plant. Thus, a model trained for one may not work for the other and training a model for either resid destruction unit may include utilization of historical data corresponding to that resid destruction unit. Various aspects of one model may be utilized to train other models for other similar equipment though. In some embodiments, a resid destruction unit at one refinery may exhibit different characteristics than a resid destruction unit or a resid destruction unit at a second refinery. Thus, a model trained for a resid destruction unit at a first refinery may not be the same or work for a resid destruction unit at a second refinery. Accordingly, training a machine learning model for a resid destruction unit may include utilization of historical data and current data corresponding to that particular resid destruction unit. Similarly, a machine learning model for a resid destruction sub-unit mayinclude utilization of historical data and current data corresponding to that particular resid destruction sub-unit.

[0051] Once a model (e.g., a machine learning 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 or 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 devices 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 subcontrollers, and / or laboratory data indicative of properties and / or a composition of one or more feed materials, intermediate materials, and product materials, such as feedstock resid materials. In another embodiment, one or more operation controllers may obtain such data, as well as target products, target parameters, and / or other factors or parameters from a refinery controller or platform.

[0052] As noted, one input to any of the models described herein may include spectra or properties of feedstock, intermediaries, products or outputs, and / or other fluids or materials utilized in a refinery, determined via one or more of a spectrographic analyzer or a chromatographic analyzer. In addition, one input to any of the models described herein may include properties of one or more feedstock resid materials to the resid destruction unit and / or resid materials to a resid destruction sub-unit. For example, the input may include a carbon residue concentration, a density, a viscosity, and / or a composition of each of such resid materials. 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 sample21CTcollection. Once a sample has been obtained, the controller and / or an analyzer may initiate analysis of the sample.

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

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

[0055] In another embodiment, the controller may optimize an operation based on the current demand for selected products. For example, for a particular targeted product (e.g., gas oil, naphtha, LPGs, distillate materials), selected amounts of feed and / or intermediaries may be utilized, increasing the demand for that feed and / or intermediaries. In some embodiments, the controller optimizes the operation of the resid destruction system based on the margin difference between the resid materials and the more valuable products formed from the resid materials. In other embodiments, demand may be a factor utilized in training a model. For example, a selected product may experience increased demand at varying times or a particular feedstock, used to produce a particular product, may be in highdemand. Data indicating such demand may be utilized in the described trained learning models.

[0056] In yet another embodiment, the controller may compare the output of the model to the current properties for a selected operation. Based on that difference of such a comparison, the controller may adjust various aspects of that operation to facilitate formation of a material (e.g., gas oil, naphtha, LPGs, distillate materials) having desired properties.

[0057] By utilizing the trained machine learning models, the systems and methods described herein may determine specific adjustments to a plurality of operations and parameters specific to equipment at a refinery to accurately and more frequently (as compared to typical adjustment times) 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.

[0058] Thus, rather than attempting to adjust operations at a significant delay, a refinery’s operations may be adjusted in real-time or near 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. In addition, since the refinery may include multiple interrelated process units, wherein the operation of an upstream unit affect the subsequent operation of a downstream unit, the use of the controllers including the machine learning models described herein, may facilitate improved (e.g., increased) control of complex interrelated processes. A refinery controller may include multiple layers of individual controls, each individual control configured to control an operation of a particular sub-operation or processing unit. In addition, each individual control may include multiple layers, each including predictive control modules each including a machine learning model configured to receive inputs associated with particular equipment or portions of the particular sub-operation to generate23CTa predicted parameter of the sub-operation. The individual control may further include a local enhancement circuity configured to receive, as an input, at least the outputs (the predicted parameters) from the predictive control modules to generate an output comprising operating conditions and / or feedstock conditions to achieve the target parameter.

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

[0060] As illustrated in FIG. 1A, a refinery 100 may include a refinery controller 101 and / or a plurality of operation controllers 102. As will be illustrated in subsequent drawings, additional components may be included, such as a refining enhancer, other circuitry, various other controllers, and / or other computing devices. In an embodiment, the refinery controller 101 and / or the plurality of operation controllers 102 may include, for example, a trained machine learning model, as well as other instructions to adjust various devices and / or operations or processes within the refinery 100. In some embodiments, the model comprises a classifier including one or more machine learning algorithms used to assign a class label to input data. The refinery controller 101 and / or the plurality of operation controllers 102 may individually connect to or be in signal communication (e.g., operable communication) with: (a) one or more sensors, meters, transducers, and / or other measurement devices positioned throughout the refinery 100; (b) to the equipment (for example, connected to some control aspect or device associated with the equipment) positioned at the refinery 100; and / or (c) user input or data from a server, such as data related to laboratory analysis of one or more samples from the refinery 100 and / or individual operations within the refinery 100, such as from a particular operation. Therefinery 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. In some embodiments, the refinery controller is configured to determine operating parameters to achieve a target product and / or target parameters. 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 of the refinery controller 101.

[0061] With reference to FIG. IB, in some embodiments, each of the operation controllers 102 may include a local enhancement circuitry 184, predictive controls circuitry 194, 191, and / or 195 (also simply referred to as “predictive controls”), and / or equipment and device controls 199. In embodiments, each of the local enhancement circuitry 184, the predictive controls circuitry 194, 191, 195, and the equipment and device controls 199 may be a module or instructions and may include a machine learning model. In embodiments, the operation controller 102 may include one or more varying or different predictive controls. The predictive controls circuitry 191, 194, 195 may be configured to predict at least one of a flow rate or a quality of a material (e.g., feedstock resid materials, resid material to each of a plurality of resid destruction sub-units), a composition of the material, a quality of an intermediate stream, a flow rate of an intermediate stream, or a composition of an intermediate stream based on one or more inputs. The one or more inputs may include, for example, sensor outputs from one or more sensors disposed within the refinery 100, such as within an operating unit of the refinery 100 (e.g., within a resid destruction unit). 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. In some embodiments, the trained machine learning model 193 is a trained machine learning model configured to predict one or more of a flow rate, a composition, or a quality of a product from a piece of equipment (e.g., a coker furnace, a coke drum, an asphalt separator) based on one more inputs including one or more of a flow rate, a quality, or a composition of a feed material to the piece of equipment, an operating temperature of the equipment, or a pressure of the25CTequipment. As such, the operation controller 102 may include a plurality of predictive controls circuitry 191. In some embodiments, the trained machine learning model 193 is configured to predict one or more of a flow rate, a composition, or a quality of feedstock resid materials based on the input data. The operation controls may also include predictive controls circuitry 194, which includes a trained machine learning model 196 and / or a first- principle model 198 (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 102 may also include predictive controls 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. In some embodiments, the first- principles model 197 is an existing first-principles model 197 and may be used to generate data (e.g., a data set) for a refinery rather than sensor and analysis data from the refinery. In some embodiments, the machine learning model of the predicative controls 195 may be used to fit, for example, coefficients of the first-principles model 197 (such as a reaction equation coefficient) to more closely match the first-principles model 197 to the refinery or actual refinery operation.

[0062] In an embodiment, the operation controller 102 may include a local enhancement circuitry 184. The local enhancement circuitry 184 may be in operable communication with each of the predictive controls circuitry 194, 191, and / or 195. The local enhancement circuitry 184 may include a trained machine learning model 190 and target setpoint instructions 192. In some embodiments, the target setpoint instructions 192 are received from a user input or from the refinery controller 101. The trained machine learning model 190 may utilize data associated with a specific refining operation and / or the output from each predictive controls circuitry 194,191, 195 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 (e.g., a target product having one or more of desired specifications, properties, or characteristics, such as a flow rate, a quality, or a composition of feedstock resid materials to a resid destruction unit and / or of resid materials to each of a plurality of resid destruction sub-units). The operation controller 102 may also include an equipment and device controls 199. The equipment and device controls 199 may cause equipment and / or devices to adjust to the target setpoints. The equipment and device controls 199 may include, for example, one or more of temperature controls, flow rate controls, pressure controls, feedstock flow rate controls, product flow ratecontrols, or other controls associated with equipment within the specific refining operation. By way of non-limiting example, and as described with reference to FIG. 6A through FIG. 7, the equipment may include one or more of a coker furnace, a coke drum, a fractionator of a coke unit, another component of a resid destruction unit, an asphaltene separator, a heater of a solvent deasphalting unit, or another component of a solvent deasphalting unit.

[0063] As illustrated in FIG. IB, the operation controllers 102 may further be connected to or in signal communication with a sample collection assembly 186 and / or a sample analysis assembly or sample analyzer 188. The sample analyzers 188 may include spectrographic analyzers, standardized spectrographic analyzers, and / or chromatographic analyzers. As described with reference to FIG. 6A through FIG. 7, the sample analyzers 188 may be configured to determine one or more properties of feedstock resid materials to a resid destruction unit, properties of resid materials to individual resid destruction sub-units and / or a product of the resid destruction unit and / or resid destruction sub-units. 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 near-infrared 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. In some embodiments, the sample analyzers 188 include viscosity analyzers, such as an in-line viscometer. In some embodiments, the sample analyzers 188 include analyzers that may be used to analyze one or more properties of feedstock resid materials and / or resid materials to each of the resid destruction sub-units and / or properties of one or more fluids formed in the one or more resid destruction sub-units, such as a carbon concentration (e.g., one or more of a Conradson carbon residue (CCR), micro residue carbon (MCR), a Ramsbottom carbon residue (RCR)) of the feedstock resid materials, resid materials to the resid destruction sub-units, or the product fluids, a viscosity of such materials, a density of such materials, a composition of such material, or another property of such materials. In some embodiments, some of the sample analyzers 188 are located within the process unit and some of the sample analyzers 188 are located in a laboratory.

[0064] The refinery controller 101 and / or operation controllers 102 may include a processor and a memory or non-transitory machine-readable storage medium storing27CTinstructions executable by the processor (as illustrated in subsequent drawings). In some examples, the refinery controller 101 and / or the operation controller 102 may be a computing device. The term “computing device” is used herein to refer to any one or all of programmable logic controllers (PLCs), distributed control systems (DCSs), a proportional integral derivative (PID) controller, a DCS-PID controller, programmable automation controllers (PACs), industrial computers, servers, virtual computing device or environment, desktop computers, personal data assistants (PDAs), laptop computers, tablet computers, smart books, palm-top computers, personal computers, smartphones, virtual computing devices, cloud based computing devices, and similar electronic devices equipped with at least a processor and any other physical components necessarily to perform the various operations described herein. Devices such as smartphones, laptop computers, and tablet computers are generally collectively referred to as mobile devices.

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

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

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

[0068] Turning to the equipment positioned at the refinery 100, the refinery 100 may include a reactor 104 and, in some embodiments, a regenerator 120. The reactor 104 may be a catalytic reactor and / or a fluid catalytic cracking unit or reactor. The reactor 104 may include one or more sensors or meters positioned within the reactor 104 (such as sensor or meter 108) and / or proximate the reactor (such as sensors or meters 112, 114, 110, and 144). These sensor or meters may measure some aspect of fluid or material where the sensor or meter is positioned. Further, the reactor 104 may receive feedstock and / or an amount of water / steam at one or more locations of the reactor 104. The reactor 104 may be positioned or configured to convert heavy gas oil, deasphalted oil, residua (residue, residuum, resid), 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 (e.g., in a fractionation column 148). 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 the 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, along29CTwith, 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 of predictive controls circuitry 191 may be trained to maximize or be utilized for maximizing operating temperature in relation to feedstock and a threshold temperature that may cause over-cracking. 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 104 in relation to feedstock and process parameters, composition, and / or other aspects to maximize yield or economically optimize yield from the reactor 104. In another embodiment, the trained machine learning model may be trained to adjust or be utilized for adjusting one or more refinery operation control devices to produce a selected yield of a product that also maximizes profit. Such an application of data to such a model may generate a vector that includes parameters or parameter settings corresponding to devices and / or equipment associated with the reactor 104 and, in some embodiments, the regenerator 120. That vector may be utilized by the local enhancement model of local enhancement circuitry 184 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 controls 199. Other models may be trained to determine parameters based on other relationships associated with the reactor 104 and / or other equipment.

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

[0070] The refinery 100 may include a regenerator 120. While a reactor 104 with a side- by-side configuration with the regenerator 120 is illustrated in FIG. 1A, it will be understood that other configurations of the reactor 104 and regenerator 120 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 102may 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 bum coke off of spent catalyst, the coke being deposited onto the catalyst in the reactor 104. The regenerator 120 may provide (e.g., recycle) the regenerated catalyst back to the reactor 104. In embodiments, the refinery controller 101 and / or operation controllers 102 may apply the data from the regenerator 120 to produce parameters or parameter settings to adjust corresponding equipment or devices to enhance operation of the regenerator 120, the reactor 104, and / or the product generated. The trained machine learning models may be trained or utilized to determine parameters to maximize the amount of carbon build up burned from catalyst, to increase temperature within the reactor 104 (for example, via heat from regenerated catalyst), and / or to minimize the amount of resources utilized by the regenerator 120.

[0071] Other equipment may be positioned throughout the refinery 100 and the refinery controller 101 and / or operation controllers 102 may connect to such equipment. The refinery controller 101 and / or operation controllers 102 may obtain or gather data related to that equipment during the refining operation. For example, the refinery controller 101 and / or operation controllers 102 may obtain data from and / or related to a fractionation column 148 or distillation column. Further, the refinery 100 may include the refinery controller 101 and / or operation controllers 102 and / or one or more of the sub-controllers, each of which may obtain data from a hydrotreater (such as hydrotreater 166 and hydrotreater 174) and / or an alkylation unit 158. The refinery controller 101 and / or operation controllers 102 may obtain data from the valve 146, sensors or meters 150, 154, 160, 164, 168, 172, 178, and 180, as well as the properties associated with the products from the fractionation column 148 (for example, off gas 152, LPG 156, alkylate 162, gasoline 170, diesel 176, slurry 182, and / or other products), the hydrotreater 166 (for31CTexample, gasoline or high-octane gasoline), the hydrotreater 174 (for example, diesel, low- sulfur diesel, and / or higher purity diesel), and / or the alkylation unit 158 (for example, alkylate).

[0072] In an embodiment, the refinery controller 101 and / or operation controllers 102 may obtain data in real-time or near real-time and / or continuously or substantially continuously. In another embodiment, the refinery controller 101 and / or operation controllers 102 may obtain data periodically. In another embodiment, the refinery controller 101 and / or operation controllers 102 may apply data to a trained machine learning model at a selected time interval. Such an interval may be based on the time for samples from various positions within the refinery 100 to be collected and analyzed. In yet another embodiment, each of the operation controllers 102 may obtain data related to a selected section (e.g., process, unit operation) 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 modules of predictive controls circuitry 194, 191, 195 to produce parameters to produce a target product. The predictive controls modules of the predictive controls circuitry 194, 191, 195 may each include one or more machine learning models configured to receive the properties and / or spectra and data and generate an output based on the properties and / or spectra and data. The local enhancement module of local enhancement circuitry 184 of the operation controller 102 may then apply, to a trained machine learning model of the local enhancement module of local enhancement circuitry 184, 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 device controls 199.

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

[0074] FIG. 2 is a simplified diagram that illustrates an apparatus for enhanced 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 FIG. 1 A, FIG. IB, and below in connection with FIG. 3 through FIG. 9.

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

[0076] The processing circuitry 202 may be configured to execute software instructions stored in the memory 204 or otherwise accessible to the processing circuitry 202 (e.g., software instructions stored on a separate storage device). In some cases, the processing circuitry 202 may be configured to execute hard-coded functionality. As such, whether configured by hardware or software methods, or by a combination of hardware with software, the processing circuitry 202 represents an entity (for example, physically embodied in circuitry) capable of performing operations according to various embodiments of the present disclosure while configured accordingly. Alternatively, as another example, when the processing circuitry 202 is embodied as an executor of software instructions, the33CTsoftware instructions may specifically configure the processing circuitry 202 to perform the algorithms and / or operations described herein when the software instructions are executed.

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

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

[0079] The apparatus 200 may include a modeling circuitry 208 configured to obtain parameters from one or more components, equipment, devices, sensors, and / or analyzers and / or apply those parameters to a trained machine learning model to obtain parameters that enable equipment to produce a target product. In other embodiments, the modeling circuitry 208 may apply, in addition to the parameters described herein, the output of other similar circuitry (in other words, an additional plurality of modeling circuitry that each correspond to one of a plurality of sub-operations). In some embodiments, the modeling circuitry 208 may be applied to an output from another machine learning model (e.g., additional plurality of modeling circuitry that each correspond to one of a plurality of suboperations). 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.

[0080] In another embodiment, the modeling circuitry 208 may train a machine learning model to generate 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).

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

[0082] The modeling circuitry 208 may utilize processing circuitry 202, memory 204, or any other hardware component included in the apparatus 200 to perform these operations, as described above in connection with FIG. 1 A and FIG. IB and below in connection with FIG. 3 through FIG. 9. The output of the modeling circuitry 208 may be transmitted to other circuitry of the apparatus 200 (such as the fluid adjustment circuitry 210 and / or equipment and device adjustment circuitry 212).

[0083] In addition, the apparatus 200 further comprises the fluid adjustment circuitry 210 that may cause adjustment of feedstock and / or other fluids utilized in a refining operation. It will be noted that “fluids” may include solid materials, resid, slurries, and materials having a relatively high viscosity and / or that are solids at room temperature, such as pitch. In an embodiment the output from the modeling circuitry 208 may be a matrix, a series of35CTparameters, 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; an operation of an SDA process; an operation of a supercritical solvent deasphalting operation process). 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 FIG. 1A and FIG. IB and below in connection with FIG. 3 through FIG. 9. The fluid adjustment circuitry 210 may further utilize communications circuitry 206 to transmit signals to adjust the type and / or amount of feedstock to utilize.

[0084] In addition, the apparatus 200 further comprises the equipment and device adjustment circuitry 212 that may cause adjustment of equipment and / or devices utilized in a refining operation. In an embodiment, the equipment and device adjustment circuitry 212 may utilize that output (from the modeling circuitry 208) 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, 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 FIG. 1 A and FIG. IB and below in connection with FIG. 3 through FIG. 9. The equipment and device adjustment circuitry 212 may further utilize communications circuitry 206 to transmit signals to adjust equipment and / or devices utilized.

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

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

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

[0088] As will be appreciated based on this disclosure, example embodiments contemplated herein may be implemented by an apparatus 200 (or by a refinery controller). Furthermore, some example embodiments (such as the embodiments described for FIG. 1A and FIG. IB and FIG. 3 through FIG. 9) may be a computer program product comprising software instructions stored on at least one non-transitory computer-readable storage medium (such as memory 204). Any suitable non-transitory computer-readable storage medium may be utilized in such embodiments, some examples of which are non-transitory hard disks, CD-ROMs, flash memory, optical storage devices, and magnetic storage devices. It should be appreciated, with respect to certain devices embodied by apparatus 200 as described in FIG. 2, that loading the software instructions onto a computing deviceor apparatus produces a special-purpose machine comprising the means for implementing various functions described herein.

[0089] FIG. 3 a simplified diagram that illustrates example refining controllers and an example refining enhancer to enhance control of a refining process at a refinery, according to an embodiment of the disclosure. As illustrated in FIG. 3, a refinery may include one or more operation controllers 302, which may correspond to the operation controllers 102 (FIG. 1 A, FIG. IB). 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 (e.g., be in operable communication with) 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 306 A, valve 306B, and up to valve 306N). In such embodiments, each of feed A 303A, 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, a resid destruction unit, a resid destruction sub-unit, a SDA unit, and other equipment as described herein). The first processing unit 308 may convert, process, and / or transform a feed into a unit material (such as unit material A 310A, unit material B 310B, and up to unit material N 310 N. Additional processing units may be positioned throughout the refinery. As illustrated though, the unit materials may flow to a “Nth” processing unit 314. Sensors (such as sensor 312A, 312B, and up to 312N) and valves (such as valve 313 A, valve 313B, and up to valve 313N) may be positioned between the feed and “Nth” processing unit 314. The final processing device (in other words, the “Nth” processing unit 314) may produce one or more end materials (such as end materials A 316A, end materials 316B, and up to end materials 316N). Sensors (such as sensor 318A, 318B, and up to 318N) and valves (such as valve 320A, valve 320B, and up to valve 320N) may be positioned between the end materials and the material destination 322.

[0090] In an embodiment, as each feed is fed to the next processing unit 314, the operation controller 302 may determine various characteristics and / or properties of the feed. For example, the operation controller 302 may determine temperature, pressure, and / or flow rate, in addition to the content of the feed and the spectra or properties determined via spectrographic analysis and / or other analysis of the feed. For example, as illustrated, the operation controller 302 may determine or obtain feed information 324 (including, at leastfeed content 326 (e.g., a feed composition) and / or feed properties 328 (e.g., carbon residue concentration, density, viscosity, other properties), among other data), unit material information 330 (including, at least unit material content 332 and / or unit material properties 334, among other data), and / or end material information 336 (including, at least end material content 338 (e.g., end material composition) and / or end material properties 340, among other data). Thus, the operation controller 302 may obtain data related to each feed / material in real-time or near real-time, during a refinery operation, and / or directly or indirectly (for example, spectra may be obtained via a sample or spectrographic analyzer).

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

[0092] In another embodiment, the output of the machine learning model 360 (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 360, then the parameter settings of the equipment or devices at the refinery may be adjusted.

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

[0094] The machine learning model 360 may include neural networks, supervised learning models, semi -supervised learning models, unsupervised learning models, or some combination thereof, as will be readily understood by one having ordinary skill in the art.39CTIn 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.

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

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

[0097] In another embodiment, a refinery may include a plurality of operation controllers 302. Each operation controller 302 may include a plurality of trained machine learning models 360. Each trained machine learning model 360 may be trained to recognize a specific or selected trend in a set of data. Thus, each operation may be adjusted based on the outputs of a plurality of models, ensuring the operation as a whole produces an accuratetarget product. Further, each operation controller may interact with each other operation controllers. For example, adjusted parameters from all operation controllers may be provided to each operation controller. Thus, as one operation is adjusted, a downstream and / or upstream operation may be further adjusted based on the adjustment of the one operation.

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

[0099] As noted, the machine learning models described herein may be trained using data, which may be referred to as training data or training input 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 (e.g., feed content), feed properties, material composition (e.g., material content), material properties, a target product or products, target composition (e.g., target content), 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 a41CTspecific attribute or parameter. Another model may be trained to utilize the outputs of each of those plurality of models, in addition to data.

[0100] Once the historical equipment specific data 402, and any other current data (e.g., current and marked up equipment specific data 404), is available, that data may be pre- processed 406. In such embodiments, pre-processing of the data includes normalizing the data. 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 start-up, shut-down, turn-arounds, maintenance, and / or upsets. Further, undesired data and invalid measurements may be removed. Finally, data may be segregated or separated into sequences based on time. The sequences may comprise data obtained over a consecutive time period, such as time intervals of 30 minutes, 1 hour, 2 hours, and / or 3 hours, or more or less than the time intervals listed. In another embodiment, other factors may be utilized to segregate or separate the data, such as feed used and / or target product being produced.

[0101] Once the data set has been pre-processed, a model may be trained 408 to form a trained machine learning model or a trained machine learning model 412. 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 (e.g., content) 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 and / or the weight of such parameters. Once the data has been used to train the machine learning model, 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, the trained machine learning model may be re-trained or refined with a different randomized portion of the data set, and the re-training may be repeated as necessary, until the selected error rate is met or achieved. In another embodiment, other training schema may be utilized. In another embodiment, readiness of the trained machine learning model may be determined based on how close the trained machine learning model comes to an expected outcome, based on the test data set.

[0102] Once the trained machine learning model 412 meets a selected error rate, the trained machine learning model 412 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 betrained, then tested. The most accurate models, determined by an error rate for each model, may be utilized.

[0103] FIG. 5 is a schematic diagram of a refinery 500. Crude oil 502 is initially processed by an atmospheric distillation tower 504 to separate different fractions of the crude oil 502 based, at least in part, on boiling point range. Lighter components from the atmospheric distillation tower 504 may include a gas 506, which may be sent to a gas processing unit 508. The gas processing unit 508 separates sour gas 509 from the gas 506 and other gas 507 resulting from different processes throughout the refinery. Gas processing unit 508 diverts the sour gas 509 to an amine treater 516. The remainder of the gas 506 and other gas 507 that is not sour gas may be referred to as a “fuel gas” and may be passed to a mercaptan treater 510 that separates mercaptans from the fuel gas (such as a mercaptan treater) and facilitates formation of liquid petroleum gas 512 (“LPG”) and butane 514. The butane 514 may be kept as a finished product, sent to a gasoline blending pool 543 and / or sent to a C4 isomerization unit 591 to be processed into isobutane 593 as a finished product or further sent to an alkylation unit 594.

[0104] The amine treater 516 may separate hydrogen sulfide from the sour gas 509 to form a refinery fuel 518 substantially free of hydrogen sulfide and a hydrogen sulfide gas 520. The hydrogen sulfide gas 520 and hydrogen sulfide gas collected from processes throughout the refinery 500 is sent to a sulfur plant 522, which may also be referred to as a “sulfur recovery unit” (SRU). The sulfur plant 522 converts the hydrogen sulfide gas 520 into sulfur 530.

[0105] Additionally, sour water 532 collected from processes throughout the refinery 500 may be sent to a sour water steam stripper 528. The sour water steam stripper 528 uses steam 534 to remove hydrogen sulfide gas 526 from the sour water 3532. The hydrogen sulfide gas 526 is also sent to the sulfur plant 522.

[0106] With continued reference to FIG. 5, the atmospheric distillation tower 504 separates light naphtha 536 from the crude oil 502. The light naphtha 536 may be sent to a hydrotreater 538 configured to remove sulfur from the light naphtha 536 and form desulfurized light naphtha. The desulfurized light naphtha may be sent to an isomerization unit 540 to be processed into isomerate 542. The isomerate 542 may include branched isomers of the light naphtha 536 (which may include straight chain hydrocarbons) and may exhibit a higher octane value than the straight chain hydrocarbons of the light naphtha 536. The isomerate 542 may be sent to a gasoline blending pool 543.43CT

[0107] In addition to separating the light naphtha 536 from the crude oil 502, the atmospheric distillation tower 504 separates heavy naphtha 544 from the crude oil 502. Optionally, the heavy naphtha 544 may be provided to a splitter 545. The splitter 545 may include one or more splitter columns that separate heavier C7 or higher naphtha molecules from lighter components of the heavy naphtha 544. The heavier C7 naphtha may be sent to the hydrotreater 554 to be desulfurized and produced as jet fuel 558 or a component of jet fuel 558. The remaining lighter components of the heavy naphtha 544 may be sent to a hydrotreater 546 to remove sulfur therefrom prior to being sent to the catalytic reformer 548 to be converted into reformate 550. 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. The reformate 550 includes high-octane branched and cyclic hydrocarbons, such as benzene, toluene, xylene, and ethylbenzene. The reformate 550 may be sent to the gasoline blending pool 543.

[0108] 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, to form a finished jet fuel 558 meeting desired specifications. Once desulfurized, the finished jet fuel 558 may be sold or sent to a jet fuel blending pool 559 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.

[0109] With continued reference to FIG. 5, 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 (e.g., a diesel hydrotreater) 562 to remove sulfur from the diesel 560 and form desulfurized diesel 564. The desulfurized diesel 564 may then be sold or sent to a diesel blending pool 565.

[0110] 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 a vacuum distillation tower 570 where the atmospheric bottoms 568 may be further separated into low vacuum gas oil 572 (“LVGO”), medium vacuum gas oil 599 (“MVGO), heavy vacuum gas oil 584 (“HVGO”), and vacuum residuum 521. Vacuum distillation tower 570 may also be configured to separate the atmospheric bottoms 568 into more or less components than that described.

[0111] The vacuum residuum 521 may be sent to a solvent deasphalting unit 598 (“SDA”) or used directly in asphalt 519, which may be used in an asphalt binder. In some applications, the vacuum residuum 521 may be sent to an asphalt binder blending pool. The SDA unit 598 may be used to extract lighter components from the vacuum residuum 521 using solvents such as propane, butane (n-butane, isobutane), or combinations thereof to extract deasphalted oil 501 from the vacuum residuum 521 and form the deasphalted oil 501 and a deasphalted asphalt, also referred to as pitch 579. The deasphalted oil 501 may be sent to a hydrocracker 586 or a fluid catalytic cracker 576 for refining into hydrocarbon products including naphtha, jet fuel, diesel, and fuel oil. The pitch 579 includes the vacuum residuum 521 after removal of the lighter components. The pitch 579 may also be used in the asphalt binder blending pool.

[0112] The atmospheric gas oil 566, the LVGO 572, the MVGO 599, and deasphalted oil 501 from the SDA unit 598 may be sent to a hydrotreater 574 to remove sulfur, after which the hydrotreated material is fed into the fluid catalytic cracker 576 and / or the deasphalted oil 501 may be sent to a hydrocracker 586. The fluid catalytic cracker 576 processes the atmospheric gas oil 566, the low vacuum gas oil 572, the medium vacuum gas oil 599, and / or the deasphalted oil 501 into naphtha 578, jet fuel 581, diesel 582, fuel oil 583, and butenes and pentenes 592.

[0113] The butenes and pentenes 592 from the fluid catalytic cracker 576, as well as the isobutane 593 from the C4 isomerizaion unit 591, may be sent to an alkylation unit 594 where the feedstocks are processed into alkylate 596. The alkylate 596 may then be sold or sent to the gasoline blending pool 543. The alkylate 596 may have a higher octane value and a lower Reid vapor pressure than the isobutane 493 and the butenes and pentenes 592.

[0114] The naphtha 578 may be sent to a hydrotreater 580 to further remove sulfur from the naphtha 578 and form desulfurized naphtha. The desulfurized naphtha may be sold or sent to a gasoline blending pool. The jet fuel 581 may also be passed through a hydrotreater 585 to remove sulfur and other contaminants and may be sent to a jet fuel blending pool. The diesel 582 may be passed through a hydrotreater 587 to remove sulfur and other contaminants and sent to a diesel blending pool. The fuel oil 583 may be passed through a hydrotreater 589 to remove sulfur and other contaminants and 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.

[0115] 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. The45CThydrocracker 586 may also produce smaller (lighter) hydrocarbons that may be sent to gas processing unit 508. The hydrocracker 586 process may integrate desulfurization and other contaminant removal such that further hydrotreating may not be necessary for the products.

[0116] The vacuum residuum 521 and pitch 579 may also be sent to a coker 597 to be processed into lighter components and / or other materials, such as naphtha 503, jet fuel 511, diesel 513, fuel oil 515, and petroleum coke 517. The naphtha 503 may be processed through a gasoline desulfurization unit 556 (“GDU”), which may include one or more processes (e.g., hydrotreating processes) to remove contaminants from the naphtha 503 and may include catalysts, particulate catches, clay treaters, salt driers, and mercaptan treaters, separators, and steam strippers. 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 the fuel oil blending pool. The pitch 579 may also be sent to an asphalt blending operation and may form at least a portion of an asphalt binder. In addition, the pitch 579 may be sold without additional processing.

[0117] 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, filters, and particulate catches.

[0118] Many of the connections, feedstocks, and outputs are not shown in FIG. 5, 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, the atmospheric distillation tower 504 and the vacuum distillation tower 570 may each represent multiple units set up in parallel or series, and may separate their respective feedstocks into more or fewer components. Additional equipment may be added to each process to further refine and separate the products of each process. For example, products and components of the products may be passed through isomerization, reformation, and alkylation processes not shown in FIG. 5 to process the products and components to meet regulatory requirements and customer specifications.

[0119] FIG. 6A is a schematic diagram of a resid destruction control system to enhance fluid production at a portion of a refinery, such as a resid destruction unit 600, according to an embodiment of the disclosure. In embodiments, a section of the refinery corresponding to a resid destruction operation may include a resid destruction controller 602. Similar to previously described controllers, the resid destruction controller 602 mayobtain data associated with the refinery equipment of the resid destruction operation, such as from a first vacuum tower 610 (also referred to as a first vacuum distillation tower or a first vacuum fractionation tower), a second vacuum tower 611 (also referred to as a second vacuum distillation tower or a second vacuum fractionation tower), an external resid source 614, a resid hydrocracker tower 615, one or more coker units 616A, 616B, and up to 616N, a solvent deasphalting unit (SDA) 618, and an external resid operation facility 620. The resid destruction controller 602 may also initiate capture of samples of fluids associated with the resid destruction operation, such as feedstock resid materials, as described herein. The sample collection and analysis assembly 608 may then analyze the samples and produce properties and / or a spectra for each sample. The resid destruction controller 602 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 604 and / or predictive controls module 606 to produce an output indicative of adjustment to parameters and / or feed. The resid destruction controller 602 may utilize the output to adjust various parameters and / or feed associated with the resid destruction operation via the local enhancement module 604.

[0120] For a resid destruction operation, such as the resid destruction unit 600, the resid destruction controller 602 may optimize yield of high value products. For example, the resid destruction controller 602 may determine amounts of resid flowing to for example, each coker 616A through 616N, the SDA unit 618, and / or the external resid operation facility 620 and determine what amount at each unit may produce a selected target.

[0121] With reference to FIG. 6A, the resid destruction unit 600 may include a plurality of resid destruction sub-units, each configured to receive at least a portion of a feedstock resid material and form a more valuable product from the feedstock resid material. As illustrated in FIG. 6 A, the resid destruction sub-units may include the coker units 616A through 616N, the SDA unit 618, and the external resid operation facility 620. The resid destruction subunits may include at least two coker units, such as the first coker unit 616A and the second coker unit 616B.

[0122] The first vacuum tower 610 may be configured to provide a first vacuum tower resid material 622 to one or more of the resid destruction sub-units; the second vacuum tower 611 may be configured to provide a second vacuum tower resid material 624 to one or more of the resid destruction sub-units; the resid hydrocracker tower 615 may be configured to provide a resid hydrocracker tower bottoms material 628 to the one or more resid destruction sub-units; and the external resid source 614 may be configured to provide an external resid material 630 (e.g., such as from tankage, a truck, a pipeline) to the one or47CTmore resid destruction sub-units. The first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630 may collectively be referred to as “feedstock resid materials” and be provided to the resid destruction unit 600.

[0123] Each of the feedstock resid materials (e.g., each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630) may exhibit one or more different properties than one another. The properties of each of the feedstock resid materials may depend on, for example, the operating conditions of the sub-units from which the respective feedstock resid materials originate (e.g., the operating conditions of the first vacuum tower 610 and the associated process equipment, the operating conditions of the second vacuum tower 611 and the associated process equipment, the operating conditions of the resid hydrocracker tower 615 and the associated process equipment, and the properties of external resid material 630). In addition, the properties of the different feedstock resid materials may depend, at least in part, on the properties of the crude oil from which the feedstock resid materials are formed, such as the density, the carbon residue concentration, and / or the assay of the crude oil.

[0124] A sensor package 632 may be in fluid communication with the first vacuum tower resid material 622; a sensor package 634 may be in fluid communication with the second vacuum tower resid material 624; a sensor package 638 may be in fluid communication with the resid hydrocracker tower bottoms material 628; and a sensor package 640 may be in fluid communication with the external resid material 630.

[0125] Each of the sensor packages 632, 634, 638, 640 may be configured to measure data with respect to each of the respective fluids (e.g., each of the respective the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630). Each of the sensor packages 632, 634, 638, 640 may be in operable communication with the resid destruction controller 602, such as with one or both of the local enhancement module 604 and / or the predictive controls module 606. The sensor packages 632, 634, 638, 640 may be configured to provide sensor data to the resid destruction controller 602.

[0126] Each of the sensor packages 632, 634, 638, 640 may include one or more of a flow meter, a sensor configured to measure a carbon residue concentration (e.g., one or more of a Conradson carbon residue (CCR), micro residue carbon (MCR), a Ramsbottom carbon residue (RCR)) of the respective materials, an asphaltene concentration, a temperaturesensor, a Coriolis meter, a viscosimeter, a densometer, and / or an in-line sensor configured to measure one or more additional properties of the respective material.

[0127] The resid destruction unit 600 may further include the sample collection and analysis assembly 608, which may include a plurality of sample collection and analysis assemblies A through H. Each sample collection and analysis assembly A through H may include a sample collection assembly configured to facilitate gathering a fluid sample; and a corresponding analysis assembly configured to provide analysis data comprising information about the sample to the resid destruction controller 602, such as to one or both of the local enhancement module 604 and / or the predictive controls module 606. In some embodiments, the sample collection and analysis assemblies A through H are configured to provide the information about the samples to the predictive controls module 606 associated with a particular portion of the resid destruction unit 600. The analysis data about the samples may include one or more properties (e.g., a composition, a carbon residue concentration, a density, a c7A, an asphaltene concentration) and / or one or more conditions (e.g., a flow rate) of the sample. In some embodiments, each of the sample collection and analysis assemblies A through H is configured to generate analysis data comprising a carbon residue concentration of each of the resid materials. As used herein, the c7A of a material may include the weight percent of carbon atoms originating from C7 aromatics (e.g., toluene, mono-methyl-substituted benzenes) of the material and may correspond to a concentration of asphaltenes in the material and may also be referred to as the c7A content. The c7A of the material may also simply be referred to as the c7A of the material. In some embodiments, the c7A may be measured with a Raman spectrometer, such as an in-line Raman spectrometer.

[0128] Each of the sample collection and analysis assemblies A through H of the sample collection and analysis assembly 608 may be configured to facilitate gathering of a respective sample from the respective sample collection and analysis assembly, analysis of the sample (e.g., at a laboratory), and reporting of the results to the sample collection and analysis assembly 608 and / or to the resid destruction controller 602, such as to the local enhancement module 604 and / or the predictive controls module 606. In some embodiments, each of the sample collection and analysis assemblies A through H are in operable communication with the predictive controls modules 606. Each of the sample collection and analysis assemblies A through H may be in operable communication with the sample collection and analysis assembly 608 and / or with the resid destruction controller 602.49CT

[0129] By way of non-limiting example, a sample collection and analysis assembly A may be in fluid communication with the first vacuum tower resid material 622; a sample collection and analysis assembly B may be in fluid communication with the second vacuum tower resid material 624; a sample collection and analysis assembly C may be in fluid communication with the resid hydrocracker tower bottoms material 628; a sample collection and analysis assembly D may be in fluid communication with the external resid material 630; a sample collection and analysis assembly E may be in fluid communication with a first coker unit product 652; a sample collection and analysis assembly F may be in fluid communication with a second coker unit product 654; a sample collection and analysis assembly G may be in fluid communication with an Nth coker unit product 656; and a sample collection and analysis assembly H may be in fluid communication with pitch 658 from the SDA unit 618. In some embodiments, at least a portion of the pitch 658 or all of the pitch 658 is provided to one or more of the coker units 616A through 616N.

[0130] In addition to the sensor packages 632, 634, 638, 640, the resid destruction unit 600 may include a sensor package 642 in operable communication with the first vacuum tower 610; a sensor package 644 in operable communication with the second vacuum tower 611; a sensor package 648 in operable communication with the resid hydrocracker tower 615; and a sensor package 650 in operable communication with the external resid source 614. Each of the sensor packages 642, 644, 648, may be configured to measure one or more conditions of the respective first vacuum tower 610, second vacuum tower 611, resid hydrocracker tower 615. For example, the sensor packages 642, 644, 648 may be configured to measure one or more of the temperature and pressure of the respective first vacuum tower 610, second vacuum tower 611, resid hydrocracker tower 615. Accordingly, the sensor packages 642, 644, 648 may provide sensor data to the resid destruction unit 600, the sensor data indicative of one or more conditions of the respective first vacuum tower 610, second vacuum tower 611, resid hydrocracker tower 615. The sensor package 650 may be configured to measure one or more properties or conditions of the external resid source 614, such as a temperature, a pressure, and / or a composition of the external resid source 614.

[0131] Each of the sensor packages 642, 644, 648, 650 may be in operable communication with the resid destruction controller 602, such as with one or both of the local enhancement module 604 and / or the predictive controls module 606, as described above with reference to the sensor packages 632, 634, 638, 640. The sensor packages 642, 644, 648, 650 may be configured to provide the respective sensor data to the resid destruction controller 602.

[0132] As described above, each of the feedstock resid materials may exhibit different properties and different flow rates based, at least in part, on the operating conditions of the sub-units from which the different feedstock resid materials originate. For example, the different feedstock resid materials may each exhibit different coking tendencies due, at least in part, to the different composition and different carbon residue concentrations of the different feedstock resid materials. The different coking tendencies, as well as the other differing properties of the feedstock resid materials, increases the complexity of optimizing the throughput of the feedstock resid materials through the resid destruction unit 600. The different properties and flow rates of the different feedstock resid materials presents challenges in determining the optimal resid composition and flow rate to provide to each resid destruction sub-unit (e.g., each of the coker units 616A through 616N, the SDA unit 618, and the external resid operation facility 620) to enhance the products formed from the resid destruction unit 600 and / or to maximize the throughput of the feedstock resid materials through the resid destruction unit and enhance the destruction of the feedstock resid materials.

[0133] Each of the sensor packages 632, 634, 638, 640, 642, 644, 648, 650 and the sample analysis and collection assemblies A through H of the resid destruction unit 600 may be configured to provide sensor data and analysis data to the resid destruction controller 602 to facilitate enhancement of the destruction of the feedstock resid materials in the resid destruction unit 600. The sensor data may be indicative of one or more operating conditions of the resid destruction unit 600, operating conditions of the sub-units upstream of the resid destruction unit 600 (e.g., sub-units that form the resid feedstock materials), and / or properties and conditions of the different feedstock materials. The analysis data from each of the sample analysis and collection assemblies A through H may be indicative of properties of the respective fluids (the respective feedstock resid materials).

[0134] Each of the resid destruction sub-units may be configured to receive a resid feed to process the resid feed to destroy (e.g., convert) the resid feed into more valuable products than the resid feed. As described above the flow rate of the resid feed and the composition of the resid feed to each of the resid destruction sub-units may be different from one another based on the operating conditions of the respective resid destruction sub-units, process constraints of the resid destruction sub-units, and the flow rate and composition of each of the feedstock resid materials. In some embodiments, the resid destruction controller 602 may be configured to determine one or more operating conditions of the resid destruction unit 600, such as the flow rate and composition of resid material to provide to each of the51CTresid destruction sub-units, to enhance the throughput of the feedstock resid materials through the resid destruction unit 600.

[0135] The first coker unit 616A may be configured to form the first coker unit product 652; the second coker unit 616B may be configured to form the second coker unit product 654; the coker unit 616N may be configured to form the Nth coker unit product 656, and the SDA unit 618 may be configured to form the pitch 658. The coker unit products 652, 654, 656 may include a combination of unique materials, such as LPGs, naphtha, distillate, and gas oil. The coker unit products 652, 654, 656 may be in fluid communication with respective sample collection and analysis assemblies E through G. In addition, the pitch 658 may be in fluid communication with a sample collection and analysis assembly H.

[0136] Each of the coker units 616A through 616N, the SDA unit 618, and the external resid operation facility 620 may include one or more sensor packages and / or one or more sample collection and analysis assemblies configured to receive sensor data and / or analysis data corresponding to one or more conditions within the respective coker units 616A through 616N, the SDA unit 618, and the external resid operation facility 620 and / or one or more properties or conditions of one or more of the fluids within the respective resid destruction sub-units.

[0137] As described in more detail herein, the resid destruction controller 602 may be configured to generate an output comprising one or more operating parameters of the resid destruction unit 600 to achieve one or more target parameters. The one or more target parameters may facilitate enhancing (e.g., optimizing) the destruction of the feedstock resid materials and / or the formation of more valuable products from the feedstock resid materials. The target parameters may include, for example, one or more of (e.g., each of) a target coke drum fill time for a first coke drum of the first coker unit 616A, a coke drum fill time for a second coke drum of the second coker unit 616B, a target level for the first coke drum, a target level for the second coke drum. In some embodiments, the target parameters include a target coke drum fill rate and a target level for each of the first coke drum and the second coke drum. The coke drum fill time for each of the coke drums (e.g., the first coke drum and the second coke drum) may correspond to a target time to fill the respective coke drums to a 100 percent level (e.g., 100 percent full of coke, wherein 100 percent full corresponds to a predetermined setpoint for full, such as 15 feet from the top of the respective coke drum). In some embodiments, the resid destruction controller 602 includes a machine learning model configured to receive inputs (e.g., sensor data, analysis data, and the target parameters) to generate an output comprising predicted parameters(e.g., a predicted flow rate and / or composition of resid to each of the resid destruction subunits, a predicted flow rate and / or composition of the feedstock resid materials from each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the hydrocracker vacuum tower bottoms material 628, and the external resid material 630, and the flow rate and composition of pitch 658 to each of the first coker unit 616A, the second coker unit 616B, and the other coker units 616N),; the flow rate of a first resid feed to the first coke drum of the first coker unit 616A (the charge rate to the first coke drum of the first coker unit 616A); and the flow rate of a second resid feed to the second coke drum of the second coker unit 616B (the charge rate to the second coke drum of the second coker unit 616B). The one or more operating parameters may include one or more operating conditions of the resid destruction unit 600, one or more operating conditions of each of the resid destruction sub-units, the flow rate of resid material to each of the resid destruction sub-units, and the composition of the resid material to each of the resid destruction subunits to achieve the target parameters to adjust the flow rate and / or composition of resid to each of the resid destruction sub-units, the flow rate and / or composition of each of feedstock resid materials (each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the hydrocracker vacuum tower bottoms material 628, and the external resid material 630), the flow rate and composition of pitch 658 to each of the first coker unit 616A, the second coker unit 616B, and the other coker units 616N to achieve the target parameters, the flow rate of a first resid feed to the first coke drum of the first coker unit 616A, and the flow rate of a second resid feed to the second coke drum of the second coker unit 616B. In some embodiments, the resid destruction controller 602 applies a machine learning model to an input received from operation controllers 102 of the resid destruction sub-units (e.g., from a coker controller 1002 of the first coker unit 616A, from a coker controller 1002 of the second coker unit 616B, from an SDA controller 802 of an SDA unit 800).

[0138] In some embodiments, each of the resid destruction sub-units includes a unique controller including a local enhancement module and predictive controls module associated with the particular resid destruction sub-unit. FIG. 6B describes a coker controller 1002 that each of the coker units 616A through 616N may include; and FIG. 6C describes an SDA controller 802 that the SDA unit 618 may include.

[0139] FIG. 6B is a simplified schematic diagram of a coker unit 1000, according to at least one embodiment of the disclosure. The coker unit 616 may correspond to the coker units 616A through 616N of the resid destruction unit 600 and may include one of the resid53CTdestruction sub-units of the resid destruction unit 600. The coker unit 616 may include enhanced coker control system to enhance fluid production at a portion of a refinery, such as at the coker units 616A through 616N. A section of the refinery corresponding to a coker unit may include a coker controller 1002, along with various equipment and devices, such as one or more coke drums 1016, 1018, a boiler or heater 1012 (also referred to as a furnace), valves 1014, 1020 and / or a fractionation or distillation column 1010. Similar to previously described controllers, the coker controller 1002 may obtain data associated with each of the equipment and devices of the coker unit 1000. The coker controller 1002 may also initiate capture of samples of fluids associated with the coker unit. The sample collection and analysis assembly 1008 may then analyze the samples and produce properties and / or a spectra for each sample. The coker controller 1002 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 1004 and / or predictive controls module 1006 to produce an output indicative of adjustment to parameters and / or feed to achieve a target parameter, such as a target yield of gas oil 1034 or naphtha 1032 (or gasoline). At least one of the machine learning models may be trained based on the relation between product yield versus heater outlet temperature. Such a model may determine parameters that maximizes yield at the lowest potential heater temperature. In another embodiment, such a model may optimize heater outlet temperature to give the product yield that maximizes profit. The coker controller 1002 may then utilize the output to adjust various parameters and / or feed associated with the coker unit via the local enhancement module 1006.

[0140] The coker unit 1000 may be configured to receive a resid feed 1022, which may also be referred to as a coker feed. The resid feed 1022 may include a combination one or more of the feedstock resid materials described above with reference to the resid destruction unit 600. For example, the resid feed 1022 may include a combination of one or more of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630. Since the resid feed 1022 includes a combination of the different feedstock resid materials, the resid feed 1022 may exhibit a different flow rate and a different composition than each of the feedstock resid materials.

[0141] In some embodiments, the resid feed 1022 is provided to the distillation column 1010. In some embodiments, at least a portion of the resid feed 1022 is provided directly to the heater 1012 rather than to the distillation column 1010. Relatively lighter (e.g., more volatile) components of the resid feed 1022 may be vaporized in the distillation column1010 and heavier portions may exit the bottom of the distillation column 1010 as a heavy distillate 1024. At least a portion of the heavy distillate 1024 may be heated in the heater 1012 to form a coke drum charge 1026 (also referred to as a coke drum feed or a heated resid feed). In some embodiments, the heater 1012 heats the coke drum charge 1026 to a temperature within a range of from about 482.2°C (about 900°F) to about 510°C (about 950°F). The heated coke drum charge 1026 may be provided to one of the coke drums 1016, 1018, which may be controlled via the valve 1014. In some embodiments, a portion of the heavy distillate 1024 may be provided to tankage or another portion of the refinery as stream 1024a.

[0142] In some embodiments, during operation of the coker unit 1000, one of the coke drums 1016, 1018 may receive the coke drum charge 1026 while the other coke drum 1016, 1018 is offline (out of fluid communication with the coke drum change 1026 and the distillation column 1010) to facilitate the removal of deposited coke from the coke drum 1016, 1018. Thus, one of the coke drums 1016, 1018 may be in fluid communication with the coke drum charge 1026 while the other coke drum 1016, 1018 is out of fluid communication with the coke drum charge 1026. Upon entering the coke drum 1016, 1018, the coke drum charge 1026 may continue thermal cracking wherein the coke drum charge 1026 cracks into a coke drum overhead 1028 lighter molecules including gases (LPGs), naphtha, distillates, and gas oils (e.g., light gas oil, heavy gas oil) and solid carbon in the form of coke. The lighter components exit the top of the coke drum 1016, 1018 and the coke deposits in the coke drum 1016, 1018. During operation of the coke drum 1016, 1018, the coke accumulates within the coke drum 1016, 1018. A cycle time of the coker unit 1000 may include a duration over which the coke drum 1016, 1018 is filled with coke, which may depend on, among other things, the flow rate of the coke drum charge 1026 and the carbon residue concentration of the resid feed 1022.

[0143] The coke drum overhead 1028 may flow through a valve 1020 and to the distillation column 1010. The distillation column 1010 may facilitate the separation of the components of the coke drum overhead 1028 from one another based on boiling point to form a distillation column overhead 1030 comprising liquified petroleum gases (LPGs), naphtha 1032, distillate 1033, and a gas oil 1034. The distillation column overhead 1030 may include a mixture of propanes (C3s including a mixture of propane and propylene) and butanes (C4s including a mixture of n-butane, iso-butane, and butylene). The distillation column overhead 1030 may include additional materials, such as methane, ethane, and other relatively light hydrocarbons. The naphtha 1032 may include heavy cat naphtha55CTwhich may used as a reformer feed, a hydrotreater feed, or blended into gasoline; and light cat naphtha which may be blended into gasoline (e.g., after treating in a gasoline hydrotreater). The distillate 1033 may include diesel materials. The gas oil 1034 may include a light gas oil and a heavy gas oil and may be further processed in a hydrocracker and / or a diesel hydrotreater.

[0144] The coker unit 1000 may include a sensor package 1036 in fluid communication with the resid feed 1022; a sensor package 1038 in fluid communication with the heavy distillate 1024; a sensor package 1040 in fluid communication with the coke drum charge 1026; a sensor package 1042 in fluid communication with the coke drum overhead 1028; a sensor package 1044 in fluid communication with the distillation column overhead 1030; a sensor package 1046 in fluid communication with the naphtha 1032; a sensor package 1047 in fluid communication with the distillate 1033; and a sensor package 1048 in fluid communication with the gas oil 1034. Each of the sensor packages 1038, 1038, 1040, 1042, 1044, 1046, 1047, 1048 may be substantially similar to the sensor packages 632, 634, 638, 640 described above and may be configured to obtain sensor data indicative of one or more properties or conditions of the respective fluids. By way of non-limiting example, the sensor package 1038 may be configured to measure the composition, the flow rate, the carbon residue concentration, the density, and / or the viscosity of the resid feed 1022 (or such properties may be inferred from the sensor data from the individual feedstock resid materials obtained from the sensor packages 632, 634, 638, 640 and / or the sample analysis and collection assemblies A through D. The sensor package 1040 may be configured to measure at measure a flow rate, a temperature, density, and / or viscosity of the coke drum charge 1026; the sensor package 1042 may be configured to measure a flow rate, a temperature, density, and / or viscosity of the coke drum overhead 1028; the sensor package 1044 may be configured to measure a measure a flow rate, a temperature, pressure, density, and / or viscosity of the distillation column overhead 1030; the sensor package 1046 may be configured to measure a flow rate, a temperature, density, and / or viscosity of the naphtha 1032; the sensor package 1047 may be configured to measure a flow rate, a temperature, density, and / or viscosity of the naphtha distillate 1033; and the sensor package 1048 may be configured to measure a flow rate, a temperature, , density, and / or viscosity of the gas oil 1034.

[0145] In addition, a sensor package 1050 may be in operable communication with the distillation column 1010; a sensor package 1052 may be in operable communication with the heater 1012; a sensor package 1054 may be in operable communication with the firstcoke drum 1016; and a sensor package 1056 may be in operable communication with the second coke drum 1018. Each of the sensor packages 1050, 1052, 1054, 1056 may be configured to measure one or more properties or conditions of the respective vessel, such as a temperature or a pressure of the respective vessel. In some embodiments, the sensor package 1054 is configured to determine a coke level of the first coke drum 1016 and the sensor package 1056 is configured to determine a coke level of the second coke drum 1018.

[0146] The sample collection and analysis assembly 1008 may include a sample collection and analysis assembly A in fluid communication with the resid feed 1022; a sample collection and analysis assembly B in fluid communication with the heavy distillate 1024; a sample collection and analysis assembly C in fluid communication with the coke drum charge 1026; a sample collection and analysis assembly D in fluid communication with the coke drum overhead 1028; a sample collection and analysis assembly E in fluid communication with the distillation tower overhead 1030; a sample collection and analysis assembly F in fluid communication with the naphtha 1032; a sample collection and analysis assembly G in fluid communication with the distillate 1033; and a sample collection and analysis assembly H in fluid communication with the gas oil 1034.

[0147] Each of the sensor packages 1038, 1038, 1040, 1042, 1044, 1046, 1047, 1048, 1050, 1052, 1054, 1056 and the sample collection and analysis assemblies A through H may be in operable communication with the coker controller 1002, such as one or both of the local enhancement module 1004 or the predictive controls module 1006. The senor packages 1038, 1038, 1040, 1042, 1044, 1046, 1047, 1048, 1050, 1052, 1054, 1056 may be configured to provide the respective sensor data to the coker controller 1002. The sample collection and analysis assemblies A through H may be configured to provide the respective analysis data to the coker controller 1002.

[0148] In some embodiments, the predictive controls module 1006 may receive inputs comprising sensor data and / or analysis data from one or more of the sensor packages and / or one or more of the sample collection and analysis assemblies A through H. The predictive controls module 1006 may be configured to apply a first machine learning model to the inputs and to one or more target parameters to generate an output comprising one or more predicted parameters. The inputs to the first machine learning model may also include a yield of the naphtha 1032, a yield of the distillate 1033, a yield of the gas oil 1034, a rate of formation of coke in the first coke drum 1016 and / or the second coke drum 1018, a weight percent of the resid feed 1022 that forms coke in the first coke drum 1016 and / or the second coke drum 1018 (e.g., the weight ratio of the coke that forms relative to the57CTweight of the resid feed 1022, which may be determined via mass balance over the coker unit 1000), a process safety management (PSM) limit of each of the first coke drum 1016 and the second coke drum 1018 (e.g. a maximum allowable flow rate of the coke drum charge 1026 to each of the first coke drum 1016 and the second coke drum 1018), a target cycle time of the coker operation, or a target level of the first coke drum 1016 and the second coke drum 1018. It will be understood that the inputs described above are nonlimiting, and the inputs to the coker controller 1002 may include inputs from upstream units providing resid feed to the coker unit 1000 or the resid feed provided to the coker unit 1000, the distillation tower 1010, the heater 1012, and / or either of the first coke drum 1016 or the second coke drum 1018.

[0149] The one or more target parameters may include one or more of a target flow rate of the resid feed 1022, a target composition of the resid feed 1022, a target yield of the naphtha 1032, a target yield of the distillate 1033, a target coke drum charge 1026 rate (a target flow rate of the coke drum charge 1026), or a target fill rate of the first coke drum 1016 or the second coke drum 1018 (whichever is online and in fluid communication with the coke drum charge 1026).

[0150] The one or more predicted parameters may include one or more of an operating condition of coker unit 1000, a flow rate of the resid feed 1022, a composition of the resid feed 1022, a temperature of the coke drum charge 1026, a temperature of the distillation tower overhead 1030, a pressure of the distillation column 1010, or a liquid yield (e.g., a yield of each of the distillation tower overhead 1030, the naphtha 1032, and the gas oil 1034) (which may correspond to the rate of coke deposition in the first coke drum 1016 or the second coke drum 1018).

[0151] In some embodiments, based on the one or more predicted parameters, the local enhancement module 1004 may be configured to apply a second machine learning model to the one or more predicted parameters, the one or more target parameters, and the inputs to generate an output comprising one or more operating conditions of the coker unit 1000 to achieve the one or more target parameters. The second machine learning model may be trained for the specific coker unit 1000. In some embodiments, the coker controller 1002 is configured to generate an output comprising a composition and / or a flow rate of the resid feed 1022 to the coker unit 1000 to generate a target coke drum fill rate. In some embodiments, the local enhancement module 1004 includes an algorithm configured to facilitate optimization of the coker unit 1000 to achieve the target parameters (e.g., the objective function) based on the outputs from the first machine learning models (e.g., ofthe predictive controls modules 1006), the target parameters, unit constraints, and inputs. The algorithm may be configured to facilitate optimization of the coker unit 1000 based on a machine learning model. In some embodiments, the local enhancement module 1004 includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g., maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules 1006), the target parameters, unit constraints, and inputs.

[0152] FIG. 6C is a simplified schematic diagram of a solvent deasphalting unit 800 including a control system to enhance fluid production at a portion of a refinery, according to an embodiment of the disclosure. The SDA 800 may correspond to the SDA unit 618. In some embodiments, solvent recover of the solvent deasphalting unit 800 occurs at subcritical conditions. A section of the refinery corresponding to the SDA may include a SDA controller 802, a boiler or furnace 816, one or more heat exchangers 810, 814, 824, 830, 840, 842, and 846, a deasphalting tower 8312, an asphalt flash drum 818, an asphalt stripper 820, a deasphalted oil stripper 826, a jet condenser 832, a compressor 834, a vaporizer 836, a solvent vaporizer 838, and a solvent work drum 844. Similar to previously described controllers, the SDA controller 802 may obtain data associated with the equipment of the SDA unit 800. The SDA controller 802 may also initiate capture of samples of fluids associated with the SDA unit 800. The sample collection and analysis assembly 808 may then analyze the samples and produce properties and / or a spectra for each sample. The SDA controller 802 may apply the data, properties, and / or spectra to one or more machine learning models of the local enhancement module 804 and / or predictive controls module 806 to produce an output indicative of adjustment to parameters and / or feed. The SDA controller 802 may then utilize the output to adjust various parameters and / or feed associated with the SDA unit via the local enhancement module 804. The SDA controller 802 may maximize gas oil from resid by setting parameters and inferring resid compositions from upstream equipment (for example, from an upstream controller, a refinery controller, and / or a supervisory controller). The SDA controller 802 may also coordinate resid from other equipment, such as from a crude tower, a vacuum tower, and / or from a third party. The SDA controller 802 may adjust DAO lift targets and / or DAO properties for further downstream operations.

[0153] The SDA unit 800 may be configured to receive a resid feed 850, which may include a combination of one or more of the feedstock resid materials described above with reference to the resid destruction unit 600. For example, the resid feed 1022 may include a59CTcombination of one or more of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630. Since the resid feed 1022 includes a combination of the different feedstock resid materials, the resid feed 1022 may exhibit a different flow rate and a different composition than each of the feedstock resid materials. In some embodiments, the resid destruction controller 602 is configured to control the resid feed 1022 to the coker unit 1000 to have a lower density and viscosity than the resid feed 850 to the SDA unit 800.

[0154] The resid feed 850 may be mixed with a solvent 856 (e.g., propane, n-butane, isobutane) in the deasphalting tower 812 to from a deasphalting tower overhead 858 and a deasphalting tower bottoms 860. The deasphalting tower overhead 858 may flow to the solvent vaporizer 838 where it is mixed with steam 871 to recover solvent recycle 869 and a solvent vaporizer bottoms 870. The solvent vaporizer bottoms 870 may be provided to the vaporizer 836 where it is mixed with steam 875 to form recycle solvent 876 and a vaporizer bottoms 877. The vaporizer bottoms 877 may be received by the deasphalted oil stripper 826. Stripping steam 827 may be provided to the deasphalted oil stripper 826 to strip the solvent from the vaporizer bottoms 877 and form a deasphalted oil stripper overhead 881 and deasphalted oil 880. The deasphalted oil stripper overhead 881 flows to a jet condenser 832 which receives water 837 and facilitates the removal of water from the deasphalted oil stripper overhead 881 to form sour water 833 and recycle solvent 835. The deasphalted oil 880 may be cooled in heat exchangers 828, 830.

[0155] The deasphalting tower bottoms 860 may be heated in heat exchange 814 and the furnace 816 and provided to the asphalt flash drum 818. The asphalt flash drum 818 may facilitate separation of the solvent from the deasphalting tower bottoms 860 to form a solvent recycle 885 and an asphalt flash drum bottoms 886. The asphalt flash drum bottoms 886 may be received by the asphalt stripper 820, which may receive stripping steam 890 to strip the solvent from asphalt flash drum bottoms 886 and form pitch 891 and solvent recycle 892. The pitch 891 may be cooled in heat exchangers 822, 824.

[0156] Each of the deasphalting tower 812, the deasphalted oil stripper 826, the asphalt flash drum 818, and the asphalt stripper 820 may include a respective sensor package 864, 883, 888, 821 each configured to measure one or more of a temperature, a pressure, or another property within the respective deasphalting tower 812, the deasphalted oil stripper 826, the asphalt flash drum 818, and the asphalt stripper 820.

[0157] A sensor package 852 may be in fluid communication with the resid feed 850 and configured to measure one or more conditions and / or properties, such as a flow rate, atemperature, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the resid feed 850. A sample collection and analysis assembly A may be in fluid communication with the resid feed 850 and configured to measure a composition, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the resid feed 850.

[0158] A sensor package 868 may be in fluid communication with the deasphalting tower bottoms 860 and configured to measure one or more conditions and / or properties, such as a flow rate, temperature, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalting tower bottoms 860. In addition, a sample collection and analysis assembly C may be in fluid communication with the deasphalting tower bottoms 860 and configured to measure a composition, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalting tower bottoms 860.

[0159] A sensor package 896 may be in fluid communication with the pitch 891 and configured to measure one or more conditions and / or properties, such as a flow rate, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the pitch 891. In addition, a sample collection and analysis assembly E may be in fluid communication with the pitch 891 and configured to measure a composition, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the pitch 891.

[0160] A sensor package 866 may be in fluid communication with the deasphalting tower overhead 858 and configured to measure one or more conditions and / or properties, such as a flow rate, temperature, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalting tower overhead 858. In addition, a sample collection and analysis assembly P may be in fluid communication with the deasphalting tower overhead 858 and configured to measure a composition, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalting tower overhead 858.

[0161] A sensor package 884 may be in fluid communication with the deasphalted oil 880 and configure to measure one or more conditions and / or properties, such as a flow rate, temperature, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalted oil 880. In addition, a sample collection and analysis assembly G may be in fluid communication with the deasphalted oil 880 and configured to61CTmeasure a composition, a density, a viscosity, a carbon residue concentration, a metals concentration, or another property of the deasphalted oil 880.

[0162] A sensor package 894 may be in fluid communication with the stripping steam 890 and configured to measure one or more of the flow rate of the stripping steam 890 and / or one or more properties of the stripping steam 890, such as one or more of the temperature, the pressure, and the quality of the stripping steam 890.

[0163] Each of the sensor packages 821, 852, 864, 866, 868, 883, 884, 888, 821, 894, 896 may be in operable communication with the SDA controller 802, such as one or both of the local enhancement module 804 or the predictive controls module 806 and configured to provide sensor data to the SDA controller 802. Each of the sample collection and analysis assemblies A through P may be in operable communication with the SDA controller 802, such as one or both of the local enhancement module 804 or the predictive controls module 806 and configured to provide analysis data to the SDA controller 802.

[0164] The SDA controller 802 may be configured to one or more of optimize an deasphalted oil lift within the deasphalting tower 812, optimize an amount of deasphalted oil 880 formed while maintaining a deasphalted oil quality above a predetermined threshold, optimize an amount of deasphalted oil lift in the deasphalting tower 812 while maintaining rheological properties of the deasphalting tower bottoms 860 above a predetermined threshold (e.g., such that the deasphalting tower bottoms 860 remains substantially pumpable), optimize the amount of the deasphalted oil 880 formed while maintaining a metals concentration (e.g., nickel, vanadium) of the deasphalted oil 800 below a predetermined threshold, optimize the amount of the deasphalted oil 880 formed while maintaining an asphaltene concentration of the deasphalted oil 880 lower than a predetermined threshold asphaltene concentration, or optimize an amount of deasphalted oil 880 while maintaining a utility consumption (e.g., steam consumption, such as of stripping steam 890, steam 871, steam 875, stripping steam 827) less than a predetermined amount.

[0165] The SDA controller 802 may be configured to apply a solvent deasphalting controller model to one or more inputs to generate one or more predicted parameters; and, based on the one or more predicted parameters, generate an output comprising one or more operating conditions of the SDA unit 800 to achieve a target parameter. In some embodiments, the SDA controller 802 includes one or more first machine learning models configured to apply the first machine learning model to input data to generate output data based on the input data, the output data comprising one or more predicted properties of oneor more product streams (e.g., the deasphalted oil 880, the pitch 891) and / or predicted operating conditions of the SDA unit 800. The inputs may include data from one or more of the sensor packages described herein (one or more of the sensor packages 821, 852, 864, 866, 868, 883, 884, 888, 821, 894, 896), one or more of the sample collection and analysis assemblies (e.g., sample collection and analysis assemblies A through P) of the sample collection and analysis assembly 808, and the target parameters. The target parameter may include one or more of a target deasphalted oil yield, a target deasphalted oil quality, a target pitch quality, a target solvent recovery, or a target profit or target profit margin.

[0166] The SDA controller 802 may include a second machine learning model configured to receive, as inputs, the outputs from the first machine learning models (e.g., the predicted properties) and, apply the second machine learning model to the inputs of the second machine learning model to generate an output comprising one or more operating conditions of the SDA unit 800 to achieve the target parameter. The inputs to the second machine learning model may include, for example, the outputs of the first machine learning models, the inputs to the first machine learning models (e.g., data from the sensor package 821, 852, 864, 866, 868, 883, 884, 888, 821, 894, 896; data from the one or more of the sample collection and analysis assemblies (e.g., sample collection and analysis assemblies A through P), and the target parameters.

[0167] In some embodiments, the inputs to the SDA controller 802 include one or more target parameters and one or more additional inputs. As described above, the SDA controller 802 may be configured to apply the SDA controller 802 to one or more inputs to generate one or more predicted parameters. The one or more predicted parameters may include one or more operating conditions of the SDA unit 800 to achieve the target parameters. For example, the one or more predicted parameters may include one or more of a predicted flow rate of the deasphalted oil 880, a predicted quality of the deasphalted oil 880 (e.g., a predicted concentration of nickel and / or vanadium in the deasphalted oil 880; a predicted asphaltene concentration, a predicted c7A, a predicted carbon residue concentration of the deasphalted oil 880), a predicted flow rate of the pitch 891, a predicted quality of the pitch 891 (e.g., a predicted softness, hardness, composition, density, or viscosity of the pitch 891), a predicted solvent recovery, a predicted flow rate of the resid feed 850 to achieve the target parameter, or a predicted property of the resid feed 850 (e.g., flow rate, composition, density, viscosity, temperature, nickel concentration, vanadium concentration, sulfur concentration, nitrogen concentration, carbon residue concentration, c7A, asphaltene concentration) to achieve the target parameter. The predicted solvent63CTrecovery may include, for example, a predicted flowrate of the solvent and / or a predicted mass balance of the solvent, such as whether the mass flow of the solvent flowing into the solvent work drum 844 balances or is within a predetermined range of the total mass of the solvent flowing out of the solvent work drum 844. The one or more predicted parameters may be one or more parameters that can be measured by the sensor packages 821, 852, 864, 866, 868, 883, 884, 888, 821, 894, 896 and / or by the sample collection and analysis assemblies A through P.

[0168] In some embodiments, the one or more predicted parameters includes a predicted flow rate of the resid feed 850, a predicted composition of the resid feed 850, and / or a predicted metal concentration of the resid feed 850. The predicted parameters may include a predicted operating condition of the deasphalting tower 812, a predicted overhead temperature of the deasphalting tower 812, and / or a predicted temperature of the deasphalting tower overhead 858. In some embodiments, the predicted parameters include a predicted flow rate of stripping steam (e.g., stripping steam 827, stripping steam 890) to the deasphalted oil stripper 826 and / or the asphalt stripper 820, a predicted temperature of the deasphalted oil stripper 826, a predicted pressure of the deasphalted oil stripper 826, a predicted temperature of the asphalt stripper 820, or a predicted pressure of the asphalt stripper 820.

[0169] The SDA controller 802 may be configured to, based on the one or more predicted parameters, generate an output comprising one or more operating conditions of the SDA unit 800 to achieve the target parameters. In some embodiments, the SDA controller 802 includes a second machine learning model configured to receive the outputs from the first machine learning models (e.g., the one or more predicted parameters) and apply the second machine learning model to such outputs and to the one or more additional inputs to generate the output. In some embodiments, the inputs provided to the second machine learning model are different than the inputs provided to the first machine learning model. In some embodiments, at least some of the inputs to which the second machine learning model is applied are the same and at least some of the inputs to which the second machine learning model is applied are different than the inputs to which the first machine learning model is applied. The machine learning model(s) of the SDA controller 802 may be configured to generate the output based on the input(s).

[0170] In some embodiments, the local enhancement module 804 includes an algorithm configured to facilitate optimization of the SDA unit 800 to achieve the target parameters based on the inputs, the target parameters, unit constraints, and the outputs from the firstmachine learning models (e.g., of the predictive controls modules 806), the target parameters, unit constraints, and inputs). The algorithm may facilitate optimization of the SDA unit 800 based on a machine learning model. In some embodiments, the local enhancement module 804 includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g., maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules 806), the target parameters, unit constraints, and inputs.

[0171] With reference back to FIG. 6A, the resid destruction controller 602 may be configured to enhance operation of the resid destruction unit 600. As described above, one challenge faced in resid destruction units is optimization of the resid destruction unit 600 to increase the throughput of the feedstock resid materials through the resid destruction unit 600 by optimizing the flow rate and composition of the resid feed to the different resid destruction sub-units (including the flow rate of the first resid feed to the first coke drum of the first coker unit 616A, and the flow rate of the second resid feed to the second coke drum of the second coker unit 616B) and minimize (e.g., reduce) unused capacity of the resid destruction sub-units. Some coker units may be limited (rate limited) by a process safety management (PSM) limit comprising a total allowable charge (feed) rate of the coke drum charge 1026 to the coker drums 1016, 1018 and other coker units may be limited (rate limited) by a coke drum fill time (e.g., the time it takes for the coke drum to fill with coke during operation of the coker unit while the resid charge is fed to the coke drum (i.e., reach a level of 100 percent)). The PSM limit may comprise the maximum allowed flow rate of the coke drum charge 1026 to each of the first coke drum 1016 and the second coke drum 1018 and may depend, at least in part, on the design specifications of the particular first coke drum 1016 and the second coke drum 1018. When two coker units are run in parallel or concurrently within the resid destruction unit 600, it is desired to optimize the composition and the flow rate of the resid feed 1022 to each of the coker units 616A, 616B and the charge rate to the coke drums 1016, 1018 in the coker units 616A, 616B such that each of the coke drums 1016, 1018 being filled with coke of each of the coker units 616A, 616B reaches (achieves) a target level (e.g., about 100 percent full) of coke at the end of the cycle time. In some instances, changing these inputs to achieve a 100 percent coke full drums at the end of a cycle time may be prevented by operational limits in the coke drum, such as the PSM charge rate limit to the drum or other operational limit. For these cases, the total volume left in both coke drums under their target levels during their drum cycle times will be minimized by the resid destruction controller 602. As one example if the65CToperating coke drum 1016, 1018 of one of the coker units 616A, 616B is filled to a maximum coke level before the cycle time, the destruction of the feedstock resid materials may not be optimized (i.e., there unused volume in the coke drum 1016, 1018). Similarly, if a coke drum of one of the coker units 616A through 616N is operating at a PSM limit but does not completely fill with coke during the cycle time whereas the coke drum of another coker unit fills with coke during the cycle time, the unfilled coke drum is operating less than optimally. For these cases, the total volume left in both coke drums under their target levels during their drum cycle times is minimized by the resid destruction controller 602.

[0172] According to embodiments described herein, the resid destruction controller 602 may be configured to optimize the destruction of the feedstock resid materials such that the coke drums of concurrently operating coker units 616A through 616N are filled with coke during their respective cycle times while reducing the amount of unused cycle time (such as by changing the flow rate of resid feed to the coker units 616A through 616N and / or changing the coking tendency of the resid feed to the individual coker units 616A through 616N based on the particular conditions and design considerations of each particular coker unit 616A through 616N, or by changing the flow rate of the resid feed to each of the coke drums of the different coker units 616A through 616N (i.e., changing the charge rates to the coke drums of the different coker units 616A through 616N)). In some embodiments, the resid destruction controller 602 is configured to control one or more operating conditions, feedstock resid flow rates, feedstock resid material compositions, resid feed compositions provided to the resid destruction sub-units, or resid feed flow rates provided to the resid destruction sub-units to control the coke drum fill rates of the coke drums of the different coker units 616A through 616N such that the coke drums of the different coker units 616A through 616N are filled with coke to at least a predetermined threshold coke level during the respective cycle time of the coker units 616A through 616N. The predetermined threshold level of the coke drums may be a level of at least about 85 percent, such as at least about 90 percent, at least about 92 percent, at least about 94 percent, at least about 96 percent, at least about 97 percent, at least about 98 percent, or at least about 99 percent of full when the coke drum of at least another of the coker units 616 A through 616N is full.

[0173] In some embodiments, the resid destruction controller 602 is configured to apply input data comprising the sensor data (from the sensor packages 632, 634, 638, 640, 642, 644, 648, 650) and the analysis data (e.g., from the sample analysis and collection assemblies A through H) to a first machine learning model to generate a first outputcomprising one or more predicted parameters of the resid destruction unit 600. The input to the first machine learning model may further include one or more target parameters of the resid destruction unit 600, which may include one or more target parameters of one or more of the resid destruction sub-units, such as at least two of the coker units 616A through 616N (e.g., at least the first coker unit 616A and the second coker unit 616B), the SDA unit 618, and the external resid operation facility 620. In some embodiments, the one or more target parameters comprises one or more of a target coke drum fill time one of the first coke drum 1016 or the second coke drum 1018 of the first coker unit 616A, a target coke drum fill time of one of the first coker drum 1016 or the second coke drum 1018 of the second coker unit 616B, a target coke drum level of one of the first coke drum 1016 or the second coke drum 1018 of the first coker unit 616A, or a target coke drum level of one of the first coke drum 1016 or the second coke drum 1018 of the second coker unit 616A.

[0174] The inputs may include one or more parameters that may affect the coke deposition rate in each of the coke drums, whether directly or indirectly. For example, the inputs may include one or more parameters or operating conditions that may affect the coke drum fill rate, which also affects the coke drum fill time. In some embodiments, the inputs include one or more of (e.g., each of) the feedstock resid flow rate, the flow rate and composition of the first vacuum tower resid material 622; the flow rate and composition of the second vacuum tower resid material 624; the flow rate and composition of the resid hydrocracker tower bottoms material 628; the flow rate and composition of the external resid material 630; the carbon residue concentration of each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630; the density of each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630; one or more operating conditions (e.g., a temperature) of the first vacuum tower 610; one or more operating conditions (e.g., a temperature) of the second vacuum tower 611; one or more operating conditions (e.g., a temperature) of the resid hydrocracker tower 615; the pressure (e.g., the overhead pressure) of the vacuum tower 610; the pressure (e.g., the overhead pressure) of the second vacuum tower 611; the pressure (e.g., the overhead pressure) of the resid hydrocracker tower 615; the level the coke drum 1016, 1018 being filled during a given cycle of a first coker unit 616A; the level of the coke drum 1016, 1018 being filled during the given cycle of a second coker unit 616B; the temperature of the coke drum charge 1026; the flow rate and composition of the coke drum charge 1026 of the first coker unit 616A; the coke drum67CTcharge 1026 of the second coker unit 616B; the flow rate and composition of the resid feed 1022 to the first coker unit 616A; the flow rate and composition of the resid feed 1022 to the second coker unit 616B; the weight percent of the resid feed 1022 that is converted to coke (e.g., which may be determined by subtracting the sum of the mass flow rate of the distillation column overhead 1030, the naphtha 1032, distillate 1033, and the gas oil 1034 from the mass flow rate of the resid feed 1022) in the first coker unit 616A; the weight percent of the resid feed 1022 that is converted to coke in the second coker unit 616B; the ratio of the resid feed 850 to the SDA unit 618 to the resid feed 1022 to each of the coker units 616A through 616N (since the SDA unit 618 forms pitch 658 that is provided as feed to the coker units 616A through 616N); the flow rate of pitch 658 as resid feed 1022 to each of the coker units 616A through 616N; and the bottoms temperature of the distillation column 1010 and / or the temperature of the heavy distillate 1024 of each of the coker units 616 A through 616N.

[0175] In some embodiments, the inputs include a flow rate, composition, and carbon residue concentration of each of the resid materials in the feedstock resid materials (e.g., each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630). In some embodiments, the inputs include the flow rate and the carbon residue concentration of such feedstock resid materials. The inputs may include one or more additional properties, conditions, or compositions associated with the sources of the feedstock resid materials, such as one or more of the density of each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630; the operating conditions (e.g., temperature) of the first vacuum tower 610; the operating conditions (e.g., temperature) of the second vacuum tower 611; the operating conditions (e.g., temperature) of the resid hydrocracker tower 615; the pressure (e.g., the overhead pressure) of the vacuum tower 610; the pressure (e.g., the overhead pressure) of the second vacuum tower 611; or the pressure (e.g., the overhead pressure) of the resid hydrocracker tower 615. In some embodiments, the input includes one or more of the flow rate, the composition, the density, the carbon residue concentration, or the assay of the crude unit feeds to each of the upstream sub-units (e.g., to each of the first vacuum tower 610, the second vacuum tower 611, and the resid hydrocracker tower 615).

[0176] In some embodiments, the inputs include data (e.g., sensor data, analysis data) associated with the resid destruction sub-units and may include the flow rate andcomposition of the resid feed 1022 to the first coker unit 616A; the flow rate and composition of the resid feed 1022 to the second coker unit 616B; the density and / or carbon residue concentration of the resid feed 1022 to the first coker unit 616A; the density and / or carbon residue concentration of the resid feed 1022 to the second coker unit 616B; the density and / or carbon residue concentration of pitch 658 provided to a coker unit 616A through 616N; the weight percent of the resid feed 1022 that is converted to coke (e.g., which may be determined by a mass balance) in the first coker unit 616A; the weight percent of the resid feed 1022 that is converted to coke in the second coker unit 616B; a pumparound rate of the gas oil 1034 of the distillation column 1010 of the first coker unit 616A and the second coker unit 616B; the ratio of the resid feed 850 to the SDA unit 618 to the resid feed 1022 to each of the coker units 616A through 616N (since the SDA unit 618 forms pitch 658 that is provided as feed to the coker units 616A through 616N); the flow rate of pitch 658 as resid feed 1022 to each of the coker units 616A through 616N; and / or the bottoms temperature of the distillation column 1010 and / or the temperature of the heavy distillate 1024 of each of the coker units 616A through 616N.

[0177] In some embodiments, the inputs may include outputs from the coker controller 1002 of each of the coker units 616 A through 616N and the outputs from the SDA controller 802. The outputs of the coker controller 1002 and the SDA controller 802 are described above. In some embodiments, the output of the coker controller 1002 includes one or more of a composition and / or a flow rate of the resid feed 1022 to the coker unit 1000 to generate a target coke drum fill rate and / or to fill the coke drum to about 100 percent full (e.g., at least about 95 percent full, at least about 97 percent full, at least about 98 percent full) within the cycle time for the coker unit 1000. The outputs of the SDA controller 802 may include operating conditions of the SDA unit 800 to achieve one or more target parameters. The outputs may include, for example, a flow rate of the pitch 891, a composition of the pitch 891, a density of the pitch 891, a viscosity of the pitch 891, and / or a carbon residue concentration of the pitch 891.

[0178] In some embodiments, the inputs further include one or more constraints of the resid destruction unit 600, such as one or more of the first coker unit 616A, the second coker unit 616B, the other coker units 616N, the SDA unit 618, or the external resid operation facility 620. By way of non-limiting example, the inputs may further include the flow rate limit of coke drum charge 1026 for the first coke drum 1016 and the second coke drum 1018 of each of the coker units 616A through 616N. In some embodiments, the inputs further include design information with respect to each of the components (e.g., vessels, process69CTequipment) of each of the resid destruction sub-units, such as the size (e.g., the height of the coker drums, the diameter of the coker drums, the available volume of the coker drums for coke deposition), the cycle time for the coker units 616A through 616N. The inputs may further include economic information, such as the current price of products formed from the resid feed 1022 in each of the coker units 616A through 616N (e.g., LPGs, naphtha, distillate, gas oil or products formed from further refinement thereof), the current price of products formed from the resid feed 850 in the SDA unit 618 (e.g., deasphalted oil 880), and the current price of resid.

[0179] The predicted parameters may include one or more of operating conditions of the resid destruction unit 600 and the resid destruction sub-units to achieve the target parameters, such as to fill each of the coke drums of the different coker units 616A through 616N to a coke level of at least about 90 percent, such as at least about 95 percent, at least about 97 percent, at least about 98 percent, or even at least about 99 percent during the cycle times of the respective coker units 616A through 616N.

[0180] By way of non-limiting example, the predicted parameters may include a predicted flow rate of the different feedstock resid materials to the resid destruction unit 600. In some embodiments, the predicted parameters include a predicted composition of the different feedstock resid materials. The predicted parameters may include one or more of a predicted coke drum fill rate of each of the first coke drum of a first coker unit 616A and a second coke drum of a second coker unit 616B, a predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit 616A and the second coker unit 616B, a predicted composition of each of the plurality of feedstock resid materials to each of the first coker unit 616A and the second coker unit 616B, at least one of a predicted flow rate or composition of pitch 658 to each of the first coker unit 616A and the second coker unit 616B, at least one of predicted flow rate of a first resid feed 1022 to the first coke drum (e.g., the predicted charge rate to the first coke drum), or at least one of a predicted flow rate of a second resid feed 1022 to the second coke drum (e.g., the predicted charge rate to the second coke drum). The predicted flow rate of each of the plurality of feedstock resid materials may include a predicted flow rate of each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630. The predicted composition of each of the plurality of feedstock resid materials may include a predicted composition of each of the first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630.

[0181] In some embodiments, the predicted parameters further include one or more predicted operating conditions of the sub-units upstream of the resid destruction unit 600 that form the feedstock resid materials having the predicted compositions and / or flow rates. The predicted parameters may include a predicted fill rate of the first coke drum of a first coker unit 616A and a predicated fill rate of a first coke drum of a second coker unit 616B. The predicted parameters may include a predicted coke drum charge 1026 of the first coke drum of the first coker unit 616A and the predicted coke drum charge 1026 of the first coke drum of the second coker unit 616B.

[0182] In some embodiments, the resid destruction controller 602 includes a first machine learning model configured to apply the first machine learning model to the inputs and the target parameters to generate one or more outputs comprising one or more predicted parameters. In some embodiments, the predicted parameters include a predicted coke drum fill rate of a first coke drum of the first coker unit 616A and a predicted coke drum fill rate of the second coker unit 616B. The predicted parameters may further include a coke drum fill rate between a current time and the end of the cycle time for each of the first coker unit 616A and the second coker unit 616B to fill the respective first coke drum and the second coke drum. The predicted parameters may further include a difference between the predicted coke drum fill rate of the first coke drum and the coke drum fill rate required to fill the first coke drum by the end of the cycle time; and a difference between the predicted coke drum fill rate of the second coke drum and the coke drum fill rate required to fill the second coke drum by the end of the cycle time. In some such embodiments, the resid destruction controller 602 includes a second machine learning model configured to be applied to at least the outputs from the first machine learning models to generate an output comprising one or more predicted parameters, which may include one or more of the predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit 616A and the second coker unit 616B, the predicted flow rate of pitch 658 to each of the first coker unit 616A and the second coker unit 616B, the predicted flow rate of the first resid feed 1022 to the first coke drum of the first coker unit 616A, or the predicted flow rate of the second resid feed 1022 to the second coke drum of the second coker unit 616B.

[0183] In some embodiments, based on the predicted parameters, the resid destruction controller 602 determines one or more operating conditions of the resid destruction unit to achieve the target parameters. In some embodiments, the resid destruction controller 602 includes a machine learning model that applies the output from the first machine learning71CTmodel (e.g., the one or more predicted parameters) and / or the second machine learning model and the target parameters to the machine learning model to generate an output comprising one or more operating conditions of the resid destruction unit 600 to achieve the target parameters. The one or more operating conditions may include a flow rate of one or more of the feedstock resid materials to the resid destruction unit 600, the composition one or more of the feedstock resid materials, the operating conditions of the sub-units upstream of the resid destruction unit (operating conditions of the first vacuum tower 610, the second vacuum tower 611, and the resid hydrocracker tower 615), the flow rate of the resid feed to each of the resid destruction sub-units (each of the coker units 616A through 616N, the SDA unit 616, and the external resid operation facility 620), the composition of the resid feed to each of the resid destruction sub-units, or the operating conditions within each of the resid destruction sub-units. In some embodiments, the one or more operating conditions include one or more of (e.g., each of) the flow rate of the coke drum charge 1026 within each of a plurality of the coker units 616A through 616N, the flow rate of the resid feed 1022 to each of the coker units 616A through 616N, the composition of the resid feed 1022 to each of the coker units 616A through 616N, and the flow rate of pitch 658 to each of the coker units 616A through 616N. In some embodiments, the operating conditions include the flow rate and composition of the first vacuum tower resid material 622, the flow rate and composition of the second vacuum tower resid material 624, the flow rate and composition of the resid hydrocracker tower bottoms material 628, and the flow rate and composition of the external resid material 630.

[0184] In some embodiments, the local enhancement module 604 includes an algorithm configured to facilitate optimization of the resid destruction unit 600 to achieve the target parameters based on the inputs, the target parameters, unit constraints, and the outputs from the first machine learning models. In some embodiments, the local enhancement module 604 includes an optimizer comprising an algorithm configured to optimize (e.g., maximize) and the target parameters (e.g., the objective function) based on the outputs from the first machine learning models (e.g., of the predictive controls modules 606), the target parameters, unit constraints, and inputs. In some embodiments, the local enhancement module 604 includes an algorithm configured to facilitate optimization of the resid destruction unit 600 to achieve the target parameters based on the inputs, the target parameters, unit constraints, and the outputs from the first machine learning models (e.g., of the predictive controls modules 606), and inputs. The algorithm may be configured to facilitate optimization of the resid destruction unit 600 based on a machine learning model.In some embodiments, the local enhancement module 604 includes an optimizer comprising an algorithm configured to achieve target parameters based on the optimization (e.g. maximization) of an objective function based on the outputs from the first machine learning models (e.g., of the predictive controls modules 606), the target parameters, unit constraints, and inputs.

[0185] The resid destruction unit 600 may include a plurality of refinery operation control devices positioned proximate and downstream or upstream of the resid destruction unit 600 configured to control aspects of the feedstock resid materials flowing to the resid destruction unit 600, the resid feeds to each of the resid destruction sub-units, and the product flowing from each of the resid destruction sub-units. In some embodiments, the resid destruction controller 602 is in operable communication with valving and process equipment associated with each of the first vacuum tower 610, the second vacuum tower 611, the resid hydrocracker tower 615, and the external resid source 614 to control the flow rate and composition of the respective first vacuum tower resid material 622, the second vacuum tower resid material 624, the resid hydrocracker tower bottoms material 628, and the external resid material 630. In addition, the resid destruction unit 600 may include valves and circuitry configured to control the flow rate of each of the first vacuum tower 610, the second vacuum tower 611, the resid hydrocracker tower 615, and the external resid source 614 to each of the resid destruction sub-units, including each of the coker units 616A through 616N, the SDA unit 618, and the external resid operation facility 620 responsive to instructions from the resid destruction controller 602. Further, the resid destruction controller 602 may be configured to control one or more operating conditions of the resid destruction sub-units, such as via operable communication with the coker controllers 1002 of the coker units 616A through 616N and operable communication with the SDA controller 802 of the SDA unit 618.

[0186] In some embodiments, the resid destruction controller 602 may continuously control the operation of the resid destruction unit 600. For example, in some embodiments, responsive to adjusting one or more operating conditions of the resid destruction unit 600, the resid destruction controller 602 may apply the first machine learning models to the input data and the target parameters every predetermined duration (e.g., every minute, every five minutes, every ten minutes, every 15 minutes) to generate an output comprising the predicted coke drum fill rate of coke drums of the first coker unit 616A and the second coker unit 616B. After determined the predicted coke drum fill rates, the resid destruction controller 602 may apply another machine learning model to the predicted coke drum fill73CTrate of coke drums of the first coker unit 616A and the second coker unit 616B and the target parameters to generate an output comprising one or more predicted parameters to achieve the target parameters. The predicted parameters may include a predicted flow rate and composition of the each of the feedstock resid materials, a predicted flow rate and composition of the resid feed to the first coker unit 616A, a predicted flow rate and composition of the resid feed to the second coker unit 616B, a predicted flow rate and composition of pitch 656 to each of the first coker unit 616A and the second coker unit 616B, a predicted flow rate of resid feed to the first coker unit 616A, a predicted flow rate and composition of the resid feed to the second coker unit 616B, the predicted flow rate of the first resid feed the first coke drum of the first coker unit 616A (the predicted charge rate to the first coke drum of the first coker unit 616A), and the predicted flow rate of the second resid feed to the second coke drum of the second coker unit 616B (the predicted charge rate to the second coke drum of the second coker unit 616A). Based on the predicted parameters, resid destruction controller 602 may change one or more of the flow rate and composition of the each of the feedstock resid materials, the flow rate and composition of the resid feed to the first coker unit 616A, the flow rate and composition of the resid feed to the second coker unit 616B, the flow rate and composition of pitch 656 to each of the first coker unit 616A and the second coker unit 616B, the flow rate and composition of resid feed to the first coker unit 616A, the flow rate and composition of the resid feed to the second coker unit 616B to achieve the target parameters, predicted flow rate of the first resid feed the first coke drum of the first coker unit 616A (the charge rate to the first coke drum of the first coker unit 616A), and the flow rate of the second resid feed to the second coke drum of the second coker unit 616B (the charge rate to the second coke drum of the second coker unit 616A).

[0187] Accordingly, the resid destruction controller 602 may be configured to enhance the destruction of resid, such as by minimizing or reducing the utilized volume of the coke drums of the coker units 616A through 616N that are unfilled with coke during a cycle time. In some embodiments, the resid destruction controller 602 facilitates maximizing the amount of coke formed in the operating coke drum of each coker unit 616A through 616N up to PSM limits for the respective coke drums and associated coker units 616A through 616N. The resid destruction controller 602 may be configured to facilitate controlling the flow rate of resid feed and the composition of the resid feed to each of a plurality of coker units to facilitate filling the coke drums of the respective coker units at or proximate the end of their respective cycle times such that each coker unit continues in operationthroughout the cycle time while reducing or minimizing volume within a coke drum that is utilized (e.g., not filled with coke) during the cycle time.

[0188] FIG. 7 is a schematic diagram of a machine learning model that may be used with the various processes within a refinery, according to at least one embodiment of the disclosure. The machine learning model may be used to optimize operation of the resid destruction controller 602 (including the coker controllers 1002 and the SDA controller 802). In some embodiments, the machine learning model obtains attribute data for each component in the resid destruction unit 600 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 desired specifications. The machine learning model may create multiple combinations that may be compared against a specification and regulatory requirements. Each combination that meets the specifications and regulatory requirements is retained in a potential formulation list, and those that do not meet the specifications are discarded. The machine learning model may 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.

[0189] The machine learning model 700 may include 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. A machine learning model may refer to a neural network or other machine learning algorithm or architecture that learns and approximates complex functions and generates outputs based on a plurality of inputs provided to the machine learning model. Further, a “machine learning model” may refer to one or multiple machine learning models that cooperatively generate one or more outputs based on corresponding inputs. For example, a machine learning model may refer to any system architecture having multiple discrete machine learning components that consider different kinds of information or inputs.

[0190] The machine learning model 700 is connected to a historical database 702. The historical database 702 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 training75CTinformation. In some embodiments, the historical database 702 may be a copy of historical databases that are maintained and updated by the machine learning model 700. The machine learning model 700 is also connected to a training database including training data 703 that includes the process data and historical information that is used for training of the machine learning model 700.

[0191] As discussed in relation to other figures of this disclosure, the machine learning model 700 may be used to analyze and improve specific processes, process controllers, and process interactions.

[0192] In some embodiments, the machine learning model 700 may receive feedstock sensor, sample, and process data 704 from a feedstock process controller 714 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 700. For example, the feedstock process may include one or more of processes of the resid destruction unit 600, the sub-units upstream of the resid destruction unit 600, and or the resid destruction sub-units.

[0193] Further, the machine learning model 700 may receive targeted process sensor, sample, and process data 706 from a targeted process controller 716 or directly from the sensors, sampling and testing systems, and labs obtaining data from the targeted process. The machine learning model 700 may also receive subsequent process sensor, sample, and process data 708 from a targeted process controller 716 or directly from the sensors, sampling and testing systems, and labs obtaining data from the targeted process. The feedstock sensor, sample, and process data 704, the targeted process sensor, sample, and process data 706, and the subsequent process sensor, sample, and process data 708 may include operating pressure and temperature data for the components of each process and sub-process. The feedstock sensor, sample, and process data 704, the targeted process sensor, sample, and process data 706, and the subsequent process sensor, sample, and process data 708 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.

[0194] The machine learning model 700 may also access business data 710. Business data 710 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 710 may also include information regarding the costs, availability, and location of storage, disposal, andin some cases carbon capture options of waste products from each process. Business data 710 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 710 may also include the inventory levels of various products that may be stored in the storage tanks. The business data 710 may also include costs associated with utilities, such as stripping steam and fuels (e.g., natural gas) associated with forming steam.

[0195] The machine learning model 700 may also access lab data 711 that is obtained from laboratory analysis of samples collected from processes within a refinery and products produced by those processes. Lab data 711 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.

[0196] The machine learning model 700 may also access weather data 713. 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 asphalt binder blending pool.

[0197] The machine learning model 700 may publish to and receive instructions from an engineering gateway 712. As used herein, “publish” means one or more of to simulate, send, or display. The engineering gateway 712 may act as a user interface for engineers to review process and machine learning model 700 data, recommendations, warnings, and requests. A user may use the engineering gateway 712 to assist the machine learning model 700 in refining and tuning its algorithms to better predict and adjust each process in response to changes in ambient weather, demand seasonality, and composition (e.g., content) and quality of feedstocks including the crude oil received for processing into refined hydrocarbons. The engineering gateway 712 may also be used to facilitate active learning by the machine learning model 700. A user may use the engineering gateway 712 to provide feedback to the machine learning model 700 to facilitate active learning. Feedback may include user approvals, user corrections, and user instructions of analysis, predictions, interpolations, data, and other recommendations that is provided by the machine learning model 700 through the engineering gateway 712.

[0198] The machine learning model 700 may access, receive data from, and publish instructions to the feedstock process controller 714, the targeted process controller 716, and a subsequent process controller 718. In some embodiments, the machine learning model 700 may adjust an algorithm used by the targeted process controller 716 based on changes77CTidentified from the adjusted algorithm. Once an adjusted algorithm has been selected by the machine learning model 700, the machine learning model 700 requests approval to implement the adjusted algorithm through the engineering gateway 712. Upon receipt of approval, the machine learning model 700 saves a copy of the approval information in the historical database 702 and sends the approved adjusted algorithm to the targeted process controller 716.

[0199] The machine learning model 700 includes a data analysis module 722. The data analysis module 722 may be used by the machine learning model 700 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 700 may save the revised data set in either the training data 703 or the historical database 702 for retraining of the machine learning model 700, for analysis and adjustment of a targeted process, or for later use.

[0200] The machine learning model 700 may include an interpolation module 724. The interpolation module 724 may be used to analyze a data set and interpolate missing data from the data set. The interpolation module 724 may use the historical database 702 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 724 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.

[0201] The interpolation module 724 may compare the historical database 702 with feedstock sensor, sample, and process data 704, the targeted process sensor, sample, and process data 706, and the subsequent process sensor, sample, and process data 708 to identify process changes. The interpolation module 724 may be configured to identify and flag small deviations for further investigation. The interpolation module 724 works with a communication module 728 to publish the data regarding the flagged deviation for user review and guidance. For example, the interpolation module 724 may be used to identifymiscalibrated sensors, misprocessed test data, or misprocessed samples that may be inaccurate. The machine learning model 700 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 712. The machine learning model 700 may use the interpolation module 724 to identify components that may be wearing out and catalysts that may be deactivated or poisoned. The interpolation module 724 may also communicate through the communication module 728 with the engineering gateway 712 to identify these potential process concerns to a user for further investigation.

[0202] The machine learning model 700 may include a prediction module 726. The prediction module 726 may analyze the feedstock sensor, sample, and process data 704, the targeted process sensor, sample, and process data 706, and the subsequent process sensor, sample, and process data 708 to predict changes in a process. For example, temperature excursions (e.g., within the upstream sub-units and / or the resid destruction sub-units) may be predicted in a process and the prediction module 726 may communicate through the communication module 728 with the engineering gateway 712 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 712 or provide different instructions for the machine learning model 700 to implement. In predicting changes in a process, prediction module 726 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 726 may provide a recommended maintenance window for the maintenance procedure to occur.

[0203] The machine learning model 700 may include the communication module 728. The communication module 728 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 728 may send or publish communications to the engineering gateway 712. The communication module 728 may also communicate with the various process controllers and other models used by the refinery.

[0204] In some embodiments, the machine learning model 700 includes a confidence module 730. The confidence module 730 may review the analysis and recommendations of the different modules of the machine learning model 700 and assign a confidence level to the analysis and recommendations. For example, the confidence module 730 may produce Shapley Additive Explanations (SHAP) values, use local interpretable model-agnostic79CTexplanations (LIME), or anchors to analyze each algorithm that is adjusted or created by the machine learning model 700. The confidence module 730 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 728 to the engineering gateway 712 to assist a user in reviewing each algorithm and the changes recommended by the machine learning model 700.

[0205] In some embodiments, the machine learning model 700 includes a regulatory module 732. The regulatory module 732 may access the business data 710 to assist in regulatory compliance. For example, the regulatory module 732 may flag a process that may be predicted by the prediction module 726 to move out of compliance (i.e., non- compliant). The regulatory module 732 may issue a warning through the communication module 728 to the engineering gateway 712. The regulatory module 732 may also identify windows of time when regulations are lessened and recommend changes to process parameters to reduce process costs. The regulatory module 732 may make recommendations to, for example, asphalt binder blending controllers to adjust blending parameters in line with upcoming regulatory changes. Further, the regulatory module 732 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.

[0206] In some embodiments, the machine learning model 700 includes a forecasting module 734. The forecasting module 734 may analyze business data 710 and recommend that various process controllers adjust their parameters to promote the production of components that may be in greater demand. The forecasting module 734 may also recommend maintenance windows for equipment producing products that may be in low demand during specific time frames. The forecasting module 734 may forecast pricing for various components based on the historical database 702 and the business data 710 and publish the forecasted pricing via the communication module 728 to the engineering gateway 712

[0207] In some embodiments, the machine learning model 700 may include an authority module 736 and additional modules 738, 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 736 may track user instructions and approvals for various instructions and changes to be made by the machine learningmodel 700 to the various controllers throughout the refinery. In some applications, the authority module 736 may consider a recommendation from a module of the machine learning model 700 and automatically authorize the machine learning model 700 to issue an instruction to the targeted process controller 716 to make a change to an associated process. In other applications, the authority module 736 may direct the communication module 728 to request approval through the engineering gateway 712 before a recommendation may be implemented and instructions sent to the targeted process controller 716.

[0208] In some embodiments, the authority module 736 directs the communication module 728 to request approval through the engineering gateway 712 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 resid destruction operation where multiple variables, including pressure, temperature, and initial composition of the feedstock affect the product (e.g., LPGs, naphtha, distillate, gas oil, deasphalted oil, pitch) and throughput (e.g., limited by, for example, coke deposition) of the process. In addition, the non-linear process may include multiple variables including the interrelation of the sub-units from which the feedstock resid materials are formed and the interrelation of the resid destruction sub-units as they relate to the throughput of the resid material through the resid destruction sub-units. A nonlinear process may include a process whose product attributes or composition are determined by a plurality of variables and thus may not be readily predictable. As one example, increasing the amount or ratio of pitch 658 with respect to the other resid materials in the resid feed to the coker units 616A through 616N may not be linear with respect to coke deposition. For example, an increase in the amount of pitch 658 to the coker units 616A through 616N may increase the coke deposition in a non-linear fashion with respect to the increase in the pitch 658 in the resid feed.

[0209] In an embodiment, each controller illustrated in FIG. 1 A through FIG. 7 may utilize various data points and properties to predict parameters and fluids to achieve a target product. Those controllers may each connect to a supervisory controller or refinery controller. In embodiments, the supervisory controller or refinery controller may utilize the output of each machine learning model of each of the controllers. In yet another embodiment, each of the controllers may utilize some output from the machine learning model of the supervisory controller or refinery controller. Further, each of the controllers81CTand / or the supervisory controller may adjust a refining operation control device, and thus adjust a process, in real-time or near real-time and / or continuously.

[0210] FIG. 8 A and FIG. 8B are simplified diagrams of control systems 1100 to enhance to enhance fluid production at refinery, according to an embodiment of the disclosure. The control system 1100 may include an operation controller 1101. Further, the operation controller 1101 may connect to one or more sensors 1116A, 1116B, and up to 1116N, one or more devices 1118A, 1118B, and up to 1118N (such as flow control devices and / or temperature control devices), one or more equipment 1120 A, 1120B, and up to 1120N, one or more analyzers 1122A, 1122B, and up to 1122N, and one or more predictive controls 1114 A, 1114B, and up to 1114N. The operation controller 1101 may include memory 1104 and one or more processors 1102. The memory 1104 may store instructions executable by one or more processors 1102. In an example, the memory 1104 may be a non-transitory machine-readable storage medium. As noted, the memory 1104 may store or include instructions executable by the processor 1102.

[0211] 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.

[0212] The memory 1104 may include or store sample and data collection and instructions 1106. Upon execution of such instructions, the operation controller 1101 may obtain samples associated with each equipment 1120A, 1120B, and up to 1120N. Further, the operation controller 1101 may obtain data from the one or more sensors 1116A, 1116B, and up to 1116N and / or one or more flow control devices 1118 A, 1118B, and up to 1118N. Upon collection of the samples, the operation controller 1101 may send the sample to one of the one or more analyzers 1122 A, 1122B, and up to 1122N. The one of the one or more analyzers 1122A, 1122B, and up to 1122N may then analyze the sample and generate properties and / or a spectra.

[0213] The operation controller 1101 may connect to and receive data from the one or more predictive controls 1114A, 1114B, and up to 1114N. In an embodiment, the operation controller 1101 may receive the output from each trained machine learning model of each the predictive controls 1114A, 1114B, and up to 1114N. In an embodiment, the output may comprise a vector or, in other embodiments, a value indicative of a parameter adjustment.

[0214] The memory 1104 may include or store trained machine learning models 1108. The trained machine learning models 1108 may include at least one trained machine learning model to generate an output indicative of parameter and / or feed adjustment. The operation controller 1101 may apply the data, properties, spectra, and / or the output of each trained machine learning model from one or more predictive controls 1114A, 1114B, and up to 1114N to the trained machine learning models 1108 to generate an output indicative of parameter adjustments and / or feed adjustment.

[0215] The memory 1104 may include or store parameter adjustment instructions 1110. Upon generation of the output, the operation controller 1101 may adjust parameters associated with equipment at the refinery. Further the memory 1104 may include or store feed adjustment instructions 1112 to adjust feed based on the output.

[0216] In FIG. 8B, predictive controls 1114 may connect to subsets of each of the components described in FIG. 8A. For example, the predictive controls 1114 may connect to a subset of the sensors 1136A, 1136B, and up to 1136N, a subset of the devices 1138A, 1138B, and up to 1138N, a subset of the equipment 1140A, 1140B, and up to 1140N, and / or a subset of the analyzers 1142A, 1142B, and up to 1142N. The predictive controls 1114 may include a trained machine learning model 1128 and instructions stored in a memory 1126 and executable by a processor 1124.

[0217] The instructions may include sample and data collection instructions 1130, which when executed cause the predictive controls 1114 to collect various data points and / or properties. Based on application of the data received to the trained machine learning model 1128 and, in some embodiments, an output from the operation controller 1101, the predictive controls 1114 may supply or provide the output to the operation controller 1101.

[0218] FIG. 9 is a flow chart illustrating a method 900 for enhanced fluid production at a refinery, according to an embodiment of the disclosure. Unless otherwise specified, the actions of the method 900 may be completed within operation controller 901 and / or predictive controls 1114. Specifically, method 900 may be included in one or more programs, protocols, or instructions loaded into the memory 1104 of operation controller 1101 (FIG. 8 A) and executed on the processor or one or more processors of the operation controller 1101. In other embodiments, method 900 may be implemented in or included in components of FIG. 1 A through FIG. 8B. 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.83CT

[0219] At block 902, the one or more predictive controls may each obtain data from corresponding sources. In such an example, each of the predictive controls may poll corresponding sensors, flow control devices, equipment, temperature control devices, and / or other data generating sources related to a corresponding refinery unit to obtain data therefrom. Further, the predictive controls may initiate sample collection for inputs or feedstock, as well as intermediaries and / or products produced by the corresponding source. Such a sample may be analyzed by one or more spectroscopic analyzers, which may subsequently produce properties and / or spectra of the samples.

[0220] At block 904, 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.

[0221] At block 906, each of the predictive controls may determine whether the subparameters are different than currently set parameters. If the parameters are different, then, in some embodiments and at block 908, each of the predictive controls may adjust one or more of the equipment, devices, or fluids. In such embodiments, the predictive controls may provide the adjustments to an equipment and device controller. In another embodiment, the predictive controls may provide the adjustments to the operation controller 801 (FIG. 8 A) 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 1101.

[0222] At block 910, the operation controller 1101 (FIG. 8 A) may determine updated parameters and / or fluid compositions (e.g., contents) and / or ratios based on application of the obtained data, as well as the outputs from each of the predictive controls, to a trained machine learning model. The operation controller 1101, in some embodiments, may first obtain data from 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 / orother devices, as well as properties and / or spectra obtained from samples taken from each unit within the refinery. Further, the operation controller 1101 may obtain the output of each model of each predictive controls. Once the operation controller 1101 obtains all relevant data, the operation controller 1101 may determine the updated parameters based on application of that data to a machine learning model.

[0223] At block 912, the operation controller 801 (FIG. 8 A) 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 1101 may compare the values of the updated parameters to the currently set parameters. If the parameters are different, then at block 914, the operation controller 1101 may adjust the devices, equipment, or fluid within the refinery to the updated parameters.

[0224] In an embodiment, refinery operations may occur continuously or substantially continuously. As such, method 900 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 900.

[0225] FIG. 10 is a simplified flow diagram of a method 1200 of operating a resid destruction unit, according to at least one embodiment of the disclosure. The method 1200 includes receiving one or more sensor outputs and / or one or more properties of each of a plurality of material streams, as shown in act 1202. The one or more sensor outputs may be from one or more sensors disposed throughout the refinery, such as throughout one or more of the resid destruction unit 600, the coker units 1000, or the SDA unit 800, as described above. The materials streams may include, for example, one or more of a plurality of feedstock resid materials, one or more intermediate streams, or one or more product streams. As described above, in some embodiments, one or more properties of each of the plurality of feedstock resid materials, one or more intermediate streams, and / or one or more product streams may be may be measured, such as with one or more sample analysis and collection assemblies.

[0226] Responsive to receiving the one or more sensor outputs and / or the one or more properties of the plurality of material streams, the method 1200 may further include receiving an input comprising at least one of the one or more sensor outputs or the one or more properties of the plurality of material streams, as shown in act 1204. In some embodiments, the input is received by a machine learning model, such as a machine learning model of a predictive controls module. The inputs may be the same as those85CTdescribed above with reference to the resid destruction controller 602, the coker controller 1002, and / or the SDA controller 802.

[0227] With continued reference to FIG. 10, the method 1200 may further include receiving a target parameter comprising at least one of a target coke drum fill time and a target level for each of a first coke drum of a first coker unit and a second coke drum of a second coker unit, as shown in act 1206. The target parameters may be substantially the same as those described above.

[0228] The method 1200 may further include applying a resid destruction controller model to the input to generate an output, as shown in act 1208. The output may comprise one or more predicted parameters, the one or more predicted parameters comprising one or more of a predicted coke drum fill rate of each of the first coke drum and the second coke drum, a predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit and the second coker unit, at least one of a predicted flow rate or composition of pitch to each of the first coker unit and the second coker unit, at least one of predicted flow rate of a first resid feed to the first coke drum, or at least one of a predicted flow rate of a second resid feed to the second coke drum. The resid destruction controller model may comprise a machine learning model trained to generate the output based on the input.

[0229] Based on the output, the method 1200 may further include adjusting one or more operating conditions of the resid destruction unit to achieve the target parameter, as shown in act 1210. Adjusting the one or more operating conditions of the resid destruction unit may include adjusting at least one of the flow rate to the first coke drum and the flow rate of to the second coke drum to achieve the target parameter.

[0230] 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 of operating resid destruction unit, the method comprising: generating an input comprising one or more of: one or more sensor outputs from one or more sensors associated with at least one of a resid destruction unit or upstream sub-units that generate a plurality of feedstock resid materials to the resid destruction unit, the resid destruction unit comprising at least a first coker unit and a second coker unit, each of the first coker unit and the second coker unit configured to convert at least a portion of the feedstock resid materials to one or more products; or one or more properties of each of the plurality of feedstock resid materials, one or more intermediate streams, or the one or more products; receiving a target parameter comprising at least one of a target coke drum fill time and a target level for each of a first coke drum of the first coker unit and a second coke drum of the second coker unit; applying a resid destruction controller model to the input to: generate an output comprising one or more predicted parameters, the one or more predicted parameters comprising one or more of a predicted coke drum fill rate of each of the first coke drum and the second coke drum, a predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit and the second coker unit, at least one of a predicted flow rate or composition of pitch to each of the first coker unit and the second coker unit, at least one of predicted flow rate of a first resid feed to the first coke drum, or at least one of a predicted flow rate of a second resid feed to the second coke drum; and based on the output, adjusting one or more operating conditions of the resid destruction unit to achieve the target parameter; wherein the resid destruction controller model comprises a machine learning model trained to generate the output based on the input.

2. The method of claim 1, wherein generating an output comprises generating an output comprising the at least one of the predicted flow rate of the first resid feed to the first coke drum or the at least one of the predicted flow rate of the second resid feed to the87CTsecond coke drum comprises minimizing a volume left in the first coke drum and the second coke drum after the target coke drum fill time.

3. The method of claim 1 or claim 2, wherein applying a resid destruction controller model to the input comprises applying a first machine learning model to the input to generate an output comprising the predicted coke drum fill rate of the first coke drum and the second coke drum.

4. The method of claim 3, wherein applying a resid destruction controller model to the input comprises applying a second machine learning model to the first coke drum fill rate of the first coke drum and the second coke drum to generate an output comprising one or more of the predicted flow rate of each of the plurality of feedstock resid materials to each of the first coker unit and the second coker unit, the predicted flow rate of pitch to each of the first coker unit and the second coker unit, the predicted flow rate of the first resid feed to the first coke drum, or the predicted flow rate of the second resid feed to the second coke drum.

5. The method of any one of claims 1 through 4, wherein adjusting one or more operating conditions of the resid destruction unit to achieve the target parameter comprises adjusting at least one of the flow rate to the first coke drum and the flow rate of to the second coke drum to achieve the target parameter.

6. The method of any one of claims 1 through 5, wherein the one or more sensor outputs comprises one or more of a density, a carbon residue concentration, or a viscosity of at least one of an atmospheric tower bottoms material or a vacuum tower bottoms material.

7. The method of any one of claims 1 through 6, wherein generating an input comprising one or more properties of one or more feedstock resid materials comprises generating an input comprising a Conradson carbon, a Ramsbottoms carbon, or a micro residue carbon of the one or more feedstock resid materials.

8. The method of any one of claims 1 through 7 wherein generating an output comprising one or more predicted parameters comprises generating an output comprisingone or more of a predicted flow rate and composition of the first resid feed to the first coker unit, a predicted flow rate and composition of the second resid feed to the second coker unit, the predicted flow rate of the first resid feed, or the predicted flow rate of the second resid feed.

9. The method of any one of claims 1 through 8, wherein generating an output comprising one or more predicted parameters comprises generating an output comprising a predicted flow rate and composition of pitch to each of the first coker unit and the second coker unit.

10. The method of any one of claims 1 through 9, wherein generating an output comprising one or more predicted parameters comprises generating an output comprising: a predicted flow rate and composition of a first vacuum tower resid material to the resid destruction unit; and a predicted flow rate and composition of a second vacuum tower resid material to the resid destruction unit.

110. The method of claim 10, wherein generating an output comprising one or more predicted parameters comprises generating an output comprising a predicted flow rate and composition of a resid hydrocracker tower bottoms material to the resid destruction unit.

12. The method of claim 10 or claim 11, wherein generating an output comprising one or more predicted parameters comprises generating an output comprising a predicted flow rate and composition of an external resid material to the resid destruction unit.

13. The method of any one of claims 1 through 12, further comprising applying the resid destruction controller model to the input after adjusting the one or more operating conditions to continuously generate the output and adjust the one or more operating conditions.

14. The method of any one of claims 1 through 13, wherein generating an input comprising one or more properties of one or more feedstock resid materials comprises89CTgenerating an input comprising a carbon residue concentration of the one or more feedstock resid materials.

15. The method of any one of claims 1 through 14, wherein the one or more sensor outputs comprises at least one of a temperature or a pressure of a vacuum tower or a resid hydrocracker tower.

16. The method of any one of claims 1 through 15, wherein the one or more sensor outputs comprises an operating condition of a vacuum tower or a resid hydrocracker tower.

17. The method of any one of claims 1 through 16, wherein the one or more sensor outputs comprises one or more sensor outputs indicative of an operating condition of the first coker unit and the second coker unit.

18. The method of claim 17, wherein the one or more sensor outputs indicative of the operating condition of the first coker unit and the second coker unit comprises one or more of: a predicted flow rate of the first resid feed to the first coker unit; a predicted flow rate of the second resid feed to the second coker unit; the predicted flow rate of the first resid feed; the predicted flow rate of the second resid feed; a predicted temperature of the first resid feed to the first coke drum; a predicted temperature of the second resid feed to the second coke drum; a predicted pressure of a distillation tower of the first coker unit; a predicted flow rate of naphtha from the distillation tower; a predicted flow rate of distillate from the distillation tower; or a predicted flow rate of gas oil from the distillation tower.

19. The method of any one of claims 1 through 18, wherein the input comprises each of: a flow rate of the first resid feed to the first coker unit; a flow rate of the second resid feed to the second coker unit; a temperature of the first resid feed;a flow rate of the first resid feed to the first coke drum; a flow rate of the second resid feed to the second coke drum; a temperature of the second resid feed; a pressure of a coker unit distillation tower; a flow rate of naphtha from the distillation tower; a flow rate of distillate from the distillation tower; or a flow rate of gas oil from the distillation tower.

20. The method of claim 19, wherein the input further comprises one or more operating conditions of: a first vacuum tower in fluid communication with the resid destruction unit; and a resid hydrocracker tower in fluid communication with the resid destruction unit.

21. The method of claim 19, wherein the input further comprises a flow rate of pitch from a solvent deasphalting unit to each of the first coker unit and the second coker unit.

22. The method of claim 1, wherein: the plurality of resid destruction units comprises a solvent deasphalting unit; and the one or more sensor outputs comprises one or more sensor outputs indicative of an operating condition of the solvent deasphalting unit.

23. The method of claim 22, wherein the one or more sensor outputs indicative of an operating condition of the solvent deasphalting unit comprises one or more of: a predicted asphaltene separator overhead temperature; a predicted flow rate of pitch; a predicted flow rate of stripping steam to a deasphalted oil stripper; or a predicted flow rate of stripping steam of an asphalt separator.

24. The method of claim 23, wherein generating an input comprises generating an input comprising at least one of a composition, a flow rate, or one or more properties of the pitch.91CT25. The method of claim 1, wherein the one or more sensor outputs comprises one or more sensor outputs indicative of an operating condition of the first coker unit, one or more sensor outputs indicative of an operating condition of the second coker unit, and one or more sensor outputs indicative of an operating condition of a solvent deasphalting unit.

26. The method of claim 25, wherein generating an output comprises generating an output comprising a predicted flow rate of the first resid feed to the first coker unit, a predicted flow rate of the second resid feed to the second coker unit, and a predicted flow rate of a resid feed to the solvent deasphalting unit.

27. The method of claim 1, wherein generating an output further comprises generating an output comprising: one or more predicted properties of the first resid feed; and one or more predicted properties of the second resid feed.

28. The method of claim 27, wherein adjusting the one or more operating conditions of the resid destruction unit to achieve the target parameter comprises: adjusting at least one of the flow rate and the one or more properties of the first resid feed; and adjusting the flow rate, the composition, and the one or more properties of the second resid feed to cause the first drum to reach the target level at the target coke drum fill time for the first coke drum and the second coke drum to reach the target level at the target coke drum fill time for the second coke drum.

29. The method of any one of claims 1 through 28, wherein: generating an input comprises: generating a first input comprising operating conditions of the first coker unit; and generating a second output comprising operating conditions of the second coker unit; and generating an output comprising one or more predicted parameters comprises: generating an output comprising a predicted coke drum fill rate of the first coke drum; andgenerating an output comprising a predicted coke drum fill rate of the second coke drum.

30. The method of claim 29, wherein generating an input further comprises: generating a carbon residue concentration of the first resid feed; and generating a carbon residue concentration of the second resid feed.

31. The method of any one of claims 1 through 30, wherein: generating an input comprises: receiving sensor data from sensors associated with a first vacuum tower configured to form a first vacuum tower resid material; receiving sensor data from sensors associated with a second vacuum tower configured to form a second vacuum tower resid material; receiving sensor data from sensors associated with a resid hydrocracker tower configured to form a hydrocracker tower bottoms resid material; and receiving data associated with properties of a feedstock resid material in tankage or in a pipeline; and generating an output comprises generating a predicted flow rate of each of the first vacuum tower resid material, the second vacuum tower resid material, the hydrocracker tower bottoms resid material, and the feedstock resid material from tankage to provide to each of the first coker unit and the second coker unit.

32. The method of any one of claims 1 through 31, wherein generating an input comprises applying a machine learning model to input data comprising data associated with a solvent deasphalting unit to generate the input comprising one or more predicted properties or a predicted flow rate of pitch to the resid destruction unit.

33. The method of claim 32, wherein generating an output comprises predicting a coke drum fill rate of the first coke drum based on the predicted at least one of the flow rate and composition of pitch to the first coke drum.

34. The method of any one of claims 1 through 33, wherein applying a resid destruction controller model to the input comprises:93CTapplying a first machine learning model of a first predictive controls model to the input to generate a first output comprising the predicted coke drum fill rate of the first coke drum; and applying a second machine learning model of a second predictive controls model to the input to generate a second output comprising the predicted coke drum fill rate of the second coke drum.

35. The method of claim 34, further comprising, applying a third machine learning model of a local enhancement module to the first input and the second input to generate a third output comprising the flow rate and the composition of first resid feed and the second resid feed.

36. The method of claim 35, further comprising changing the flow rate and composition of the first resid feed and the second resid feed based on the third output.

37. The method of claim 35, further comprising changing an operating condition of a vacuum distillation tower, a solvent deasphalting unit, or a resid hydrocracker tower based on the third output to change at least one of a flow rate and a composition of the first resid feed and the second resid feed.

38. The method of any one of claims 1 through 37, wherein generating an output to achieve the target parameter is based on one or more of demand, price, or cost of the one or more feedstock resid materials, naphtha, or gas oil.

39. The method of any of claims 1 through 38, wherein the machine learning model is trained on a historical data of the resid destruction unit, the historical data including: the one or more sensor outputs from the one or more sensors; and the one or more properties of each of the plurality of feedstock resid materials, the one or more intermediate streams, or the one or more products.

40. The method of claim 39, wherein the historical data includes demand, price, or cost of the plurality of feedstock resid materials, the one or more intermediate streams, or the one or more products.

41. 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 40.

42. 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 40.

43. A system for operating a resid destruction unit, the system comprising: one or more sensors disposed throughout at least one of the resid destruction unit or upstream of the resid destruction unit, the resid destruction unit configured to convert at least a portion of a feedstock resid material to one or more products; one or more sample collection assemblies to collect samples of fluid associated with the resid destruction unit; one or more sample collection assemblies for collecting samples of the feedstock resid material, one or more intermediate streams, or the one or more products; one or more sample analysis assemblies to analyze each collected sample to provide properties of collected samples from the one or more sample collection assemblies; a plurality of refinery operation control devices configured to control aspects of fluid flowing to or from the resid destruction unit or one or more operating conditions of the resid destruction unit; and a resid destruction unit controller in operable communication with the one or more sensors, the one or more sample collection assemblies, the one or more sample analysis assemblies, and the plurality of refinery operation control devices, the resid destruction unit comprising: at least one processor; and a computer memory including instructions that, when executed by the processor, cause the resid destruction unit to carry out operations comprising: apply a machine learning model to input data comprising sensor data from the one or more sensors, analysis data from the one or more sample analysis assemblies, and target parameters to generate an output comprising one or more predicted parameters, the target parameters comprising a target coke drum fill time and a target level for each of a first coke drum of a first coker unit and a second coke drum of a second coker unit; and95CTbased on the output, adjust one or more operating parameters of the resid destruction unit with the plurality of refinery operation control devices.

44. The system of claim 43, wherein the one or more predicted parameters comprises one or more of a predicted coke drum fill rate of each of the first coke drum and the second coke drum, a predicted flow rate of each of a plurality of feedstock resid materials to each of the first coker unit and the second coker unit, at least one of a predicted flow rate or composition of pitch to each of the first coker unit and the second coker unit, a predicted coke drum charge to the first coke drum, or a predicted coke drum charge to the second coke drum.

45. The system of claim 43 or claim 44, wherein the input data comprises a carbon residue concentration of the feedstock resid material.

46. The system of any one of claims 43 through 45, wherein the feedstock resid material comprises at least a first vacuum tower resid material and a second vacuum tower resid material.

47. The system of any one of claims 43 through 46, wherein the input data comprises one or more of: a flow rate of a first resid feed to the first coker unit; a flow rate of a second resid feed to the second coker unit; a temperature of the first resid feed; a temperature of the second resid feed; a composition of the first resid feed; a composition of the second resid feed; a pressure of a coker unit distillation tower of each of the first coker unit and the second coker unit; a flow rate of naphtha from the distillation tower; a flow rate of distillate from the distillation tower; or a flow rate of gas oil from the distillation tower.

48. The system of any one of claims 43 through 47, wherein the input comprises an operating condition of:a first vacuum tower in fluid communication with the resid destruction unit; and resid hydrocracker tower in fluid communication with the resid destruction unit.

49. The system of any one of claims 43 through 48, wherein the input comprises a flow rate of pitch from a solvent deasphalting unit to each of the first coker unit and the second coker unit.

50. The system of any one of claims 43 through 49, wherein the instructions are configured to cause the plurality of refinery operation control devices to adjust one or more operating parameters of the resid destruction unit to cause the first coke drum to reach the target level at the target coke drum fill time of the first coke drum and the second coke drum to reach the target level at the target coke drum fill time of the second coke drum.

51. The system of claim 50, wherein the instructions are configured to change a flow rate and a composition of a first resid to the first coker unit to achieve the target level at the target coke drum fill time of the first coke drum.

52. A system for enhancing fluid production for a resid destruction operation, the system comprising: a resid destruction unit to receive a feed and produce one or more fluids; a plurality of sensors to measure a parameter associated with the resid destruction unit, the plurality of sensors positioned at one or both of proximate the resid destruction unit or within the resid destruction unit; a plurality of refinery operation control devices positioned proximate and downstream or upstream of the resid destruction unit and to control aspects of fluid flowing to or from the resid destruction unit; one or more sample collection assemblies to collect samples of the fluid associated with the resid destruction unit; one or more sample analysis assemblies to analyze each collected sample to provide properties of the collected samples; and a resid destruction controller in signal communication with the resid destruction unit, the plurality of sensors, the plurality of refinery operation control97CTdevices, and the one or more sample analysis assemblies, and storing a trained machine learning model, the resid destruction controller configured to: determine an output including predicted properties of feedstock and parameter settings of the plurality of refinery operation control devices and the resid destruction unit based on application of one or more of data measured by the plurality of sensors; data corresponding to analysis from the one or more sample analysis assemblies; or a target product and corresponding target properties to the trained machine learning model, and adjust one or more of an amount of feed or type of feed and parameters associated with the plurality of refinery operation control devices and resid destruction unit based on the output to enhance fluid production.

Images