Method for operating a continuous contact reforming unit

Abstract

A method for operating a continuous catalytic reforming unit is a step of forming one or more product outflow flows through a hydrocarbon reaction logistics to a continuous catalytic reforming unit comprising at least one fluid flow preheater, at least one catalytic reactor, and at least one separation unit; a step of executing a hydrocarbon reforming process control system comprising a hydrocarbon reforming unit variable data memory storing processor executable instructions, a hydrocarbon reforming unit output conversion module, and one or more predictive hydrocarbon reforming unit modeling processors, wherein the one or more predictive hydrocarbon reforming unit modeling processors execute the processor executable instructions and receive into the process control system one or more signals from one or more state variable actuator hardware indicating one or more current state variables, wherein the current state variables are process variables that cannot be directly set in the continuous catalytic reforming unit The process may include: receiving one or more signals, one or more signals indicating one or more current control variables of the continuous contact reforming unit from one or more control variable actuator hardware, wherein the current control variables are process variables that can be set directly in the continuous contact reforming unit; and using a machine learning model to generate improved control variables that increase a selected performance variable based on input of one or more current state variables or one or more current control variables, wherein the machine learning model has been trained using at least historical state variable data, historical control variable data, and historical performance variable data as inputs; and adjusting one or more current control variables of the continuous contact reforming unit based on the improved control variables determined by the machine learning model.

Description

Description of Related Applications

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 509,618, filed on June 22, 2023, and U.S. Patent Application No. 18 / 462,068, filed on September 6, 2023, the entire contents of both of which are hereby incorporated by reference in their entirety.

Technical Field

[0002] Embodiments of the present disclosure broadly relate to chemical processing, and more particularly, to processes and systems for reforming hydrocarbon materials.

Background Art

[0003] Catalytic reforming is a chemical process that can be used to reform refined hydrocarbons, such as naphtha streams distilled from crude oil, into high-octane products (sometimes called reformate). Such reformate can be used as gasoline.

Summary of the Invention

Problems to be Solved by the Invention

[0004] Due to the very high demand for gasoline and other products from reformate, there is a desire in the industry to improve the way reforming units operate.

Means for Solving the Problems

[0005] According to one or more embodiments, a method for operating a continuous catalytic reforming unit is a step of forming one or more product outflow flows through a hydrocarbon reaction logistics to a continuous catalytic reforming unit comprising at least one fluid flow preheater, at least one catalytic reactor, and at least one separation unit; executing a hydrocarbon reforming process control system comprising a hydrocarbon reforming unit variable data memory storing processor executable instructions, a hydrocarbon reforming unit output conversion module, and one or more predictive hydrocarbon reforming unit modeling processors, wherein the one or more predictive hydrocarbon reforming unit modeling processors execute the processor executable instructions and receive one or more signals from one or more state variable actuator hardware indicating one or more current state variables, the current state variables not directly set in the continuous catalytic reforming unit. The process may include: receiving one or more signals which are process variables and one or more signals which indicate one or more current control variables of the continuous contact reforming unit from the actuator hardware, wherein the current control variables are process variables which can be set directly in the continuous contact reforming unit; and using a machine learning model to generate improved control variables which increase a selected performance variable based on the input of one or more current state variables or one or more current control variables, wherein the machine learning model has been trained using at least historical state variable data, historical control variable data, and historical performance variable data as inputs; and adjusting one or more current control variables of the continuous contact reforming unit based on the improved control variables determined by the machine learning model.

[0006] Additional features and advantages of the described embodiments are described in the following detailed description, some of which will be readily apparent to those skilled in the art from that description, or will be recognized by carrying out the described embodiments, including the following detailed description, claims, and accompanying drawings. [Brief explanation of the drawing]

[0007] The following detailed description of specific embodiments of this disclosure will be best understood when read in conjunction with the following drawings, which have similar structures and are numbered similarly. [Figure 1] Generalized schematic diagram of a continuous contact reforming unit according to one or more embodiments described in this disclosure. [Figure 2] A diagram illustrating a machine learning model used for operating a continuous contact reforming unit according to one or more embodiments described in this disclosure. [Figure 3] Figure showing an example model created using a machine learning model used for operating a continuous contact reforming unit according to one or more embodiments described in this disclosure. [Modes for carrying out the invention]

[0008] For the purpose of describing simplified schematic diagrams and descriptions of the chemical processing system, numerous valves, temperature sensors, electronic control devices, etc., which are used and well known to those skilled in the art of a particular chemical processing operation, are not included in the drawings. Furthermore, ancillary components that are usually included in a typical chemical processing operation, such as air supply devices, catalyst hoppers, and flue gas treatment systems, are not illustrated. Also, components associated with the hydrocracking unit, such as bleed streams, spent catalyst discharge subsystems, and catalyst exchange subsystems, are not shown. It should be understood that these components are included in the spirit and scope of the disclosed embodiments. However, operating components such as those described in this disclosure may be added to the embodiments described in this disclosure.

[0009] Furthermore, it should be noted that relevant arrows in the drawings may indicate process flows. However, these arrows may also equally indicate transfer lines that move process flows between two or more system components. In addition, arrows connected to system components define inlets or outlets at each given system component. The direction of the arrows generally coincides with the main direction of movement of the material in the flow contained within the physical transfer line indicated by the arrow. Furthermore, arrows not connected to two or more system components indicate product flows discharged from the illustrated system, or system inlets flowing into the illustrated system. Product flows may be further processed in an accompanying chemical processing system or commercialized as final products. System inlets may be flows transferred from an accompanying chemical processing system or unprocessed raw material flows. Some arrows may represent recirculation flows, which are discharge flows from system components that are recycled back into the system. However, it should be understood that in some embodiments, the indicated recirculation flows may be replaceable by system inlets of the same material, and a portion of the recirculation flow may be discharged from the system as system products.

[0010] In addition, arrows in the diagram may schematically represent process steps that transport fluid flow from one system component to another. For example, an arrow pointing from one system component to another may represent "passing" the discharge of one system component to the other, which may include the "discharge" or "removal" of the contents of the process flow from one system component and the "introduction" of the contents of the resulting product flow into the other system component.

[0011] In embodiments shown in the related drawings, it should be understood that an arrow between two system components may indicate that the fluid flow is not processed between the two system components. In other embodiments, the fluid flow indicated by the arrow may have substantially the same composition throughout transport between the two system components. In addition, it should be understood that in one or more embodiments, the arrow may indicate that at least 75% by mass, at least 90% by mass, at least 95% by mass, at least 99% by mass, at least 99.9% by mass, and even 100% by mass of the fluid flow are transported between the system components. Therefore, in some embodiments, for example, when a slip stream is present, not all of the fluid flow indicated by the arrow is transported between the system components.

[0012] In the schematic flowcharts of the related drawings, where two or more lines intersect, it should be understood that two or more process flows are “mixed” or “merged.” Mixing or merging may also include mixing by directly introducing both fluid flows into similar reactors, separators, or other system components. For example, where it is shown that two fluid flows are directly merged just before entering a separation unit or reactor, it should be understood that in some embodiments, the fluid flows may be introduced equally into the separation unit or reactor and mixed within the reactor.

[0013] Various embodiments are described in more detail below. Some of these embodiments are shown in the accompanying drawings. Wherever possible, the same reference numerals are used throughout the drawings to refer to the same or similar parts.

[0014] This specification describes a method for operating a continuous catalytic reforming unit (sometimes referred to herein as "CCR") that processes hydrocarbon reaction logistics, such as those that reform naphtha into product effluent flows such as reformed oil. Such a continuous catalytic reforming unit may utilize a hydrocarbon reforming process control system that can predict the system attributes of the continuous catalytic reforming unit using machine learning models, as described in detail herein. Specifically, the continuous catalytic reforming unit described herein may be configured to generate improved control variables based on a machine learning model developed from at least historical control variable data and historical state line number data. The continuous catalytic reforming unit may be adjusted based on this model so that the current control variables are adjusted to the generated improved control variables.

[0015] As described herein, continuous catalytic reforming units generally process hydrocarbon reaction flows into one or more product flows, which may be further processed or collected as products for use. For example, continuous catalytic reforming units can utilize naphtha flows or straight-run gasoline as reaction flows, and generally the octane number of such reaction flows can be improved. Reforming, which is a process carried out in a continuous catalytic reforming unit, can generally be defined as the combined effect of molecular changes (i.e., hydrocarbon reactions) resulting from, for example, the dehydrogenation of cyclohexane, the dehydrogenation isomerization of alkylcyclopentanes, and the formation of aromatic compounds by the dehydrogenation cyclization of paraffins and olefins; the isomerization of n-paraffins; the formation of cyclohexane by the isomerization of alkylcycloparaffins; the isomerization of substituted aromatic compounds; and / or the hydrocracking of paraffins (producing gas and coke). Catalysts are used in catalytic reforming. Some such catalysts may contain a metal hydrogenation-dehydrogenation (hydrogen transition) component, or multiple components, sometimes platinum, substantially atomically dispersed on the surface of a porous inorganic oxide support such as alumina. The support typically contains halides, particularly chlorides, and may provide acidic functional groups that function in isomerization, cyclization, and hydrocracking reactions.

[0016] As used herein, “catalyst” refers to any substance that increases the rate of a particular chemical reaction. The catalysts described herein, but are not limited to, can be used to accelerate a variety of reactions, including, but are not limited to, dehydrogenation, dehydrogenated isomerization, isomerization, and hydrocracking. As used herein, “decomposition” generally refers to a chemical reaction in which a carbon-carbon bond is cleaved. For example, a molecule having a carbon-carbon bond may be decomposed into multiple molecules by the cleavage of one or more carbon-carbon bonds, or compounds containing alkyl or cyclic moieties, such as alkanes, cycloalkanes, naphthalenes, and aromatic compounds, may be converted into compounds that do not contain a cyclic moiety, or have fewer cyclic moieties than before decomposition, and / or olefin compounds.

[0017] Referring here to Figure 1, the continuous catalytic reforming unit 100 is generally shown, comprising a chemical processing unit 104 (including all of the chemical processing units of the continuous catalytic reforming unit 100) and a hydrocarbon reforming unit process control unit 200. Other continuous catalytic reforming units may be used in the processes currently disclosed, and it should be understood that the continuous catalytic reforming unit 100 in Figure 1 is an illustrative continuous catalytic reforming unit.

[0018] The chemical processing unit 104 of the continuous catalytic reforming unit 100 may generally include a fluid flow preheater 110, a catalytic reactor 130, and a separation unit 150. The continuous catalytic reforming unit 100 in Figure 1 comprises two fluid flow preheaters 110 and two catalytic reactors 130 arranged alternately in series. Specifically, the continuous catalytic reforming unit 100 comprises a first set 180 of fluid flow preheaters 110 and catalytic reactors 130, followed by a second set 182 of another fluid flow preheater 110 and catalytic reactor 130. Additional sets of fluid flow preheaters 110 and catalytic reactors 130 may be included in the continuous catalytic reforming unit 100, and thus there may be three, four, five, or even more fluid flow preheaters 110 and catalytic reactors 130 (only two sets are shown in Figure 1).

[0019] As used in this disclosure, “reactor,” such as catalytic reactor 130, refers to a vessel capable of producing one or more chemical reactions between one or more reactants in the presence of one or more catalysts, as may be necessary. For example, a reactor may include a batch reactor, a continuous stirred tank reactor (CSTR), or a tank or tubular reactor configured to operate as a plug-flow reactor. Exemplary reactors include packed-bed reactors, such as fixed-bed reactors, and fluidized-bed reactors. One or more “reaction zones” may be located within the reactor. As used in this disclosure, “reaction zone” refers to a zone within the reactor where a particular reaction occurs. For example, a packed-bed reactor with multiple catalyst beds may have multiple reaction zones, each of which is defined by a zone on each catalyst bed.

[0020] As used in this disclosure, “separation unit,” such as separation unit 150, refers to any separation device or system of separation devices that separates one or more chemical substances mixed in a process stream from each other, at least partially. For example, a separation unit can selectively separate different chemical species, phases, or materials of different particle sizes from each other to form one or more chemical fractions. Examples of separation units include, but are not limited to, distillation columns, flash drums, knockout drums, knockout pots, centrifuges, cyclones, filters, traps, scrubbers, expansion devices, membranes, and solvent extractors. It should be understood that the separation processes described in this disclosure may not completely separate all of one chemical component from all of another. It should be understood that the separation processes described in this disclosure separate different chemical components from each other “at least partially,” and that separation may include only partial separation, even if not explicitly stated. As used in this disclosure, one or more chemical components may be “separated” from a process stream to form a new process stream. Generally, a process stream may enter a separation unit and be divided, i.e., separated, into two or more process streams having a desired composition. Furthermore, in some separation processes, a "low-boiling point fraction" (also called the "light fraction") and a "high-boiling point fraction" (also called the "heavy fraction") may be discharged from the separation unit, and on average, the components of the low-boiling point fraction have lower boiling points than those of the high-boiling point fraction. Other fluid flows may fall between the low-boiling point and high-boiling point fractions, such as the "intermediate-boiling point fraction."

[0021] According to the embodiment of FIG. 1 described in this specification, the hydrocarbon reaction stream 102 can be sent to the continuous catalytic reforming unit 100. Before being passed through the continuous catalytic reforming unit 100, hydrogen 190 can be mixed with the hydrocarbon reaction stream 102. Specifically, the hydrocarbon reaction stream 102 can be mixed with hydrogen 190 and sent to the fluid stream preheater 110 of the first set 180. The fluid stream preheater 110 can generally heat the contents of the hydrocarbon reaction stream 102. The heating in the fluid stream preheater 110 can be carried out in various ways, such as the combustion of hydrocarbon fuel or heat exchange with other media. For example, the fuel gas can be burned in the fluid stream preheater 110 to heat the hydrocarbon reaction stream 102.

[0022] The heated hydrocarbon reaction stream 102 can exit the fluid stream preheater 110 as the heated reaction stream 112 and be sent to the catalytic reactor 130. In the catalytic reactor 130, the heated reaction stream 112 contacts the catalyst, and the heated reaction stream 112 undergoes one or more reactions to form the intermediate catalytic reactor effluent stream 132. As shown in FIG. 1, the intermediate catalytic reactor effluent stream 132 can be sent to another fluid stream preheater 110 and then to the catalytic reactor 130 (shown as the second group 182 of the fluid stream preheater 110 and the catalytic reactor 130). That is, the intermediate catalytic reactor effluent stream 132 can be heated in the fluid stream preheater 110 to form the heated catalytic reactor effluent stream 114, and the heated catalytic reactor effluent stream 114 can be passed through the catalytic reactor 130 and reacted by contact with the catalyst in the catalytic reactor 130 of the second group 182 to form the final catalytic reactor effluent stream 134.

[0023] Referring further to Figure 1, the catalyst can generally be passed through one or more catalytic reactors 130, where it comes into contact with a heated reaction flow 112 or a heated catalytic reactor outflow 114. In the embodiments described herein, the catalytic reforming unit is a “continuous” catalytic reforming unit 100, meaning that one or more catalytic reactors 130 are moving bed reactors, in contrast to fixed bed reactors, where the catalyst is continuously added and removed. As shown in Figure 1, the catalyst can be circulated between the catalytic reactors 130 and the catalyst regenerator 170 via catalyst flows 172, 174, and 176. The catalyst regenerator 170 can remove coke that may form from the catalyst during the reaction in the catalytic reactor 130. In addition, since the catalyst may lose chloride over time, it can be rechlorinated by being treated with chloride via a fluid flow 192. The scheme shown in Figure 1 involves a catalyst circulating between two catalytic reactors 130 before being returned to the catalyst regenerator 170. However, other configurations are also possible in which the catalyst circulates through only one catalytic reactor 130 before being returned to the catalyst regenerator 170.

[0024] According to the embodiment, reactions occurring in one or more catalytic reactors 130 may include naphthene dehydrogenation, naphthene isomerization, and / or paraffin dehydrogenation cyclization. Naphthene dehydrogenation generally refers to a rapid, highly endothermic reaction. This reaction can be accelerated by metal catalyst function and preferably proceeds under high temperature and low pressure conditions. Naphthenes are usually the most desirable starting material because they are easy to accelerate and can produce by-products such as hydrogen and aromatic hydrocarbons. Naphthene isomerization generally involves isomerization with ring rearrangement, and the probability of ring opening to form paraffin can be relatively high. Paraffin dehydrogenation cyclization generally refers to a very endothermic reaction, and because the reaction rate is relatively low, this reaction usually requires harsher operating conditions than usual, and coke production may increase. As the molecular weight of the paraffin increases, the paraffin cyclization process may become easier. Dehydrogenation cyclization is promoted under low-pressure, high-temperature conditions, and typically requires the function of both a metal catalyst and an acid catalyst to accelerate this reaction.

[0025] According to one or more embodiments, the final catalytic reactor effluent stream 134 can be sent to a separation unit 150, where the contents of the final catalytic reactor effluent stream 134 are separated into a plurality of product effluent streams 160. The product effluent streams 160 may include an offgas 152, a hydrogen-rich gas stream 154, an LPG stream 156, and a reformate stream 158. The various product effluent streams 160 may be passed through downstream processing units and further processed or stored as final products for sale and distribution. For example, the offgas 152 can be sent to a fuel gas supply network for use as fuel gas for heat input, the hydrogen-rich gas stream 154 can be utilized in pressure swing adsorption ("PSA") processing, isomerization processing, and / or naphtha hydrotreating ("NHT"), the LNG stream 156 can be sent to a liquefied natural gas ("LNG") storage facility, and / or the reformate stream 158 can be sent to a gasoline blending pool.

[0026] As detailed with respect to the chemical processing section 104, it should be noted that the operating continuous catalytic reforming unit 100 can include a number of "state variables" and a number of "control variables". As described herein, a control variable is a variable that can be directly set in the continuous catalytic reforming unit 100. On the other hand, a state variable is a variable that cannot be directly set in the continuous catalytic reforming unit 100. For example, an operator of the continuous catalytic reforming unit 100 can directly set control variables such as the amount of catalyst supplied to a particular catalytic reactor 130. However, variables such as the pressure within a particular catalytic reactor 130 cannot be directly set by an operator of the continuous catalytic reforming unit 100 and are, therefore, state variables. This is because there is no direct input to the system for controlling the pressure within the catalytic reactor 130. In other words, a control variable can be understood as an input to the system that can be directly selected by a plant operator or the hydrocarbon reforming unit process control unit 200, and a state variable is a function of the various control variables of the continuous catalytic reforming unit 100 and is a variable that can be indirectly controlled by adjustment of the control variables.

[0027] State variables assumed in the continuous catalytic reforming unit 100 include, but are not limited to, catalyst activity, mean pressure of the catalytic reactor, reactor downtime, reformed oil yield, hydrogen yield, off-gas yield, LPG yield, fuel gas consumption, cooling water consumption, power consumption, separation unit temperature, or heat exchanger load on the combined fluid of feedstock and waste.

[0028] In addition, but not limited to, control variables assumed in the continuous catalytic reforming unit 100 include the weighted average inlet temperature of the reaction flow, the supply rate of the reaction flow, the quality of the reactant feed flow (e.g., N+2A supply value, initial boiling point, and final boiling point), the chloride injection rate, the average carbon combustion rate in the catalyst regenerator 170, the catalyst circulation rate, the hydrogen / hydrocarbon ratio, and the energy consumption in the fluid flow preheater 110.

[0029] According to the embodiment, both state variables and control variables can be monitored by one or more control variable monitoring hardware 232 and / or one or more state variable monitoring hardware 234, respectively. In addition, control variables can be controlled by one or more control variable actuator hardware 236. The exemplary locations of the control variable monitoring hardware 232, state variable monitoring hardware 234, and control variable actuator hardware 236 are shown in Figure 1 as merely an example, and these various sensors and actuators may be located elsewhere in the continuous contact reforming unit 100 in a manner consistent with the description herein. None of the specifically illustrated control variable monitoring hardware 232, state variable monitoring hardware 234, and control variable actuator hardware 236 should be construed as essential to the continuous contact reforming unit 100.

[0030] In some embodiments, the control variable monitoring hardware 232 and / or one or more state variable monitoring hardware 234 can generally be sensors depending on the type of variable being monitored (i.e., measured). For example, the control variable monitoring hardware 232 and / or one or more state variable monitoring hardware 234 related to pressure can be pressure gauges, and the control variable monitoring hardware 232 and / or one or more state variable monitoring hardware 234 related to temperature can be thermometers. Similarly, the control variable actuator hardware can be any device depending on the type of variable being controlled. For example, the control variable actuator hardware 236 controlling flow rate can be an actuator on a valve that controls the flow rate. In some embodiments, the control variable monitoring hardware 232 and the control variable actuator hardware 236 can be integrated for control variables of fluid flow, where a single integrated valve has both the functions of monitoring and controlling the flow rate and performs the functions of both the control variable monitoring hardware 232 and the control variable actuator hardware 236. Specific equipment suitable for use with the control variable monitoring hardware 232, the state variable monitoring hardware 234, and / or the control variable actuator hardware 236 will be recognizable and selectable to a person skilled in the art based on a specific control variable or state variable or its operation.

[0031] Referring further to Figure 1, according to the embodiments described herein, the continuous catalytic reforming unit 100 further comprises a hydrocarbon reformer process control unit 200. The hydrocarbon reformer process control unit 200 generally interacts with the chemical processing unit 104 and, in some embodiments, can directly control it. In one or more embodiments, the hydrocarbon reformer process control unit 200 may comprise a hydrocarbon reformer variable data memory 210 capable of storing processor-executable instructions, and a predictive hydrocarbon reformer modeling processor 220 that can be configured to execute processor-executable instructions. The hydrocarbon reformer process control unit 200 may further include a hydrocarbon reformer output conversion module 230. The hydrocarbon reformer variable data memory 210, the predictive hydrocarbon reformer modeling processor 220, and the hydrocarbon reformer output conversion module 230 can communicate as shown in Figure 1, where at least the hydrocarbon reformer variable data memory 210 communicates directly with the hydrocarbon reformer process control unit 200, and the predictive hydrocarbon reformer modeling processor 220 communicates directly with the hydrocarbon reformer output conversion module 230.

[0032] As described herein, “current” control variables or “current” state variables are variables currently present in the continuous catalytic reforming unit 100, in contrast to “past” state or control variable data, which are historical data related to the control and / or state variables. Generally, past state or control variable data can be collected over a considerably long period of time, and therefore, such data is suitable for use in machine learning models, as described herein.

[0033] In one embodiment, the control variable monitoring hardware 232 can monitor any control variable of the chemical processing unit 104, and such data collected from the control variable monitoring hardware 232 can be sent to the process control unit 200, so that the process control unit 200 can receive one or more signals indicating one or more current control variables in real time or near real time. In such an embodiment, one or more control variable monitoring hardware 232 can communicate with the hydrocarbon reformer variable data memory 210 via a communication module, such as a receiver module 272, as shown in Figure 1. Similarly, the state variable monitoring hardware 234 can monitor any state variable of the chemical processing unit 104, and such data collected from the state variable monitoring hardware 234 can be sent to the process control unit 200, so that the process control unit 200 can receive one or more signals indicating one or more current state variables in real time or near real time. In such embodiments, one or more state variable monitoring hardware 234 can communicate with the hydrocarbon reformer variable data memory 210 via a communication module, such as a receiver module 274, as shown in Figure 1. As described herein, for example, the control variable monitoring hardware 232 and / or state variable monitoring hardware 234 may be thermometers, pressure gauges, valves, etc.

[0034] According to one or more embodiments, the hydrocarbon reformer variable data memory 210 stores data related to a machine learning model and / or one or more of the current state variables and current control variables. The hydrocarbon reformer variable data memory 210 may be configured as any conventional or yet-to-be-developed structure for storing sensor data, such as random access memory (RAM), read-only memory (ROM), data registers, databases, and / or other hardware for storing sensor data. The predictive hydrocarbon reformer modeling processor 220 represents hardware and software suitable for performing operations on data acquired from at least the hydrocarbon reformer variable data memory 210 of the process control unit 200. More specifically, in one or more embodiments, the predictive hydrocarbon reformer modeling processor 220 may include certain software-based logic modules, such as data acquisition logic and modeling logic, for generating one or more improved control variables in real time or near real time based on inputs of the current state variables, current control variables, and selected performance variables. For example, the predictive hydrocarbon reformer modeling processor 220 can receive one or more current state variables and one or more current control variables from the hydrocarbon reformer variable data memory 210, and select performance variables from the hydrocarbon reformer output conversion module 230, and can calculate the model and determine the improved control variables, as described below.

[0035] Referring further to Figure 1, the hydrocarbon reformer process control system 200 may include a hydrocarbon reformer output conversion module 230. The hydrocarbon reformer output conversion module 230 is any hardware configured to convert the output of the predictive hydrocarbon reformer modeling processor 220 into a form usable for controlling the technical operations related to the chemical processing unit 104, and may include, for example, any hardware that generates operational outputs usable in the chemical processing unit 104 to modify, improve, or otherwise control the technical operations within the continuous catalytic reformer unit 100, or to produce technical effects.

[0036] In some embodiments, the hydrocarbon reformer output conversion module 230 can communicate with the predictive hydrocarbon reformer modeling processor 220 to receive at least improved control variables. In addition, the hydrocarbon reformer output conversion module 230 can communicate with one or more control variable actuator hardware 236 of the continuous catalytic reformer unit 100. The control variable actuator hardware 236 can operate to change the current control variables. For example, the control variable actuator hardware 236 may include a control device or valve actuator that can change the control variables in the continuous catalytic reformer unit 100. For example, in some embodiments, the hydrocarbon reformer output conversion module 230 can send messages to one or more of the control variable actuator hardware 236 to make adjustments to the continuous catalytic reformer unit 100. The hydrocarbon reformer output conversion module 230 may further include an interface for inputting control variables selected by a plant operator, and generate improved control variables based on such input. The hydrocarbon reformer output conversion module 230 can also provide an interface for plant operators to view proposed improved control variables, machine learning models, etc., and then command the hydrocarbon reformer output conversion module 230 to send a message to one or more of the control variable actuator hardware 236 to adjust the control variables of the continuous catalytic reformer unit 100.

[0037] According to one embodiment, the method for operating the continuous catalytic reforming unit 100 may further include the step of adjusting one or more current control variables of the continuous catalytic reforming unit based on one or more improved control variables determined by a machine learning model. In some embodiments, this adjustment can be performed automatically by the process control unit 200 by having the hydrocarbon reformer output conversion module 230 send a signal to the control variable actuator hardware 236 to adjust the control variables. For example, the process control unit 200 can automatically command the control variable actuator hardware 236 to change the current control variables. In another embodiment, the hydrocarbon reformer output conversion module 230 can display improved control variables to the plant operator, who can then adjust one or more current control variables based on the improved performance variables after considering the changes.

[0038] Here, a more detailed description of the operation of a hydrocarbon reformer process control system 200 utilizing a machine learning model is presented in one or more embodiments. Such a description may, in some cases, be limited to selected control variables, selected state variables, and selected performance variables. However, this description is intended to provide an extended interpretation of the subject matter currently described, including embodiments with different selected control variables, selected state variables, and selected performance variables.

[0039] In some embodiments, the machine learning model is trained using at least historical state variable data, historical control variable data, and historical performance variable data as input. Historical performance variable data may include any historical data related to performance variables, as described herein. Performance variables refer to variables of the chemical processing unit 104 related to the performance of the system, as described herein. For example, one or more performance variables may be selected from the reformed oil production yield, hydrogen production yield, off-gas production yield, LPG production yield, reaction fluid input rate, chloride injection rate, or catalyst residence time in the system. Performance variables may be state variables, but it should be understood that, in general, the machine learning model may use state variables as input while using different performance variables as output. The machine learning model can be trained using historical data collected from the chemical processing unit 104 over a past period. Such data may be collected by the control variable monitoring hardware 232 and / or the state variable monitoring hardware 234.

[0040] Referring to Figure 2, without limitation, a neural network model can be used as a machine learning model. Another possible machine learning model that can be used is a random forest classification model. As shown in Figure 2, in a neural network model, state variable data and control variable data can be used as inputs, and performance variable data can be used as outputs. Historical values ​​of this data can be used to train the neural network model, and by assigning improvement values ​​to each node to match the data regression, the model accuracy for use in predictions based on historical data can be improved.

[0041] A neural network model generally includes one or more layers, each having one or more nodes connected by node connections. The one or more layers may include an input layer, one or more hidden layers, and an output layer. The neural network model can be a deep neural network, a convolutional neural network, or other types of neural networks. The neural network model may include one or more convolutional layers and one or more fully connected layers. The input layer represents the raw information supplied to the neural network model. In some embodiments, state variable data and control variable data may be input to the neural network model in the input layer, and performance variables in the output layer. For example, during training, past state variable data and past control variable data may be input to the neural network model in the input layer, and past performance variable data may be input to the output layer. In training mode, the neural network model can train neural network paths using one or more feedback or backpropagation techniques.

[0042] A neural network model processes raw information received in the input layer through nodes and node connections. One or more hidden layers, dependent on the input from the input layer and the weights on the node connections, perform computational processing. In other words, the hidden layers perform calculations and transmit information from the input layer to the output layer through the relevant nodes and node connections.

[0043] Generally, when a neural network model is learning, it identifies and determines patterns in the raw information received in the input layer (i.e., past control variable data and past state variable data). Accordingly, one or more parameters, such as weights related to node connections between nodes, can be adjusted through a process known as backpropagation. While there are various processes that can be used for learning, it's important to understand that two common learning processes are associative mapping and regularity detection. Associative mapping refers to the learning process in which a neural network model learns to generate another specific pattern corresponding to the input set each time a given specific pattern is applied to the input set. Regularity detection refers to the learning process in which a neural network learns to respond to specific characteristics of the input pattern. In associative mapping, the neural network stores relationships between patterns, whereas in regularity detection, the response of each unit has a specific "meaning." This type of learning mechanism can be used for feature discovery and knowledge representation.

[0044] A neural network possesses knowledge contained within the values ​​of its node connection weights. Modifying the knowledge stored in the network as a function of experience means learning rules for changing the weight values. The information is stored in the neural network's weight matrix. Learning is the determination of weights.

[0045] To train a neural network model to perform a certain task, weights are adjusted in a way that reduces the error between the desired output and the actual output. In this process, the neural network model may need to calculate the derivative of the error with respect to the weights. In other words, it must calculate how the error changes when each weight is slightly increased or decreased. One method used to determine weights is the backpropagation algorithm.

[0046] Data related to the trained machine learning model (such as node weights) is stored in the hydrocarbon reformer variable data memory 210 and can be used by the predictive hydrocarbon reformer modeling processor 220 to generate improved control variable outputs. As described herein, improved control variables refer to control variables that have been improved compared to the current control variables based on information obtained from the machine learning model, and this is described herein. Such improved control variable outputs can be used to adjust the settings of the chemical processing unit 104. It should be understood that the improved control variables do not need to be fully optimized, but only need to be improved compared to the current control variables.

[0047] In some embodiments, current control variables (received from control variable monitoring hardware 232 to hydrocarbon reformer variable data memory 210) and / or current state variables (received from state variable monitoring hardware 234 to hydrocarbon reformer variable data memory 210) can be transmitted to the predictive hydrocarbon reformer modeling processor 220. In some embodiments, the predictive hydrocarbon reformer modeling processor 220 can use a trained machine learning model to generate a machine learning model of performance variables as a function of current state variables and current control variables, such as the one shown in Figure 3. In other embodiments, the predictive hydrocarbon reformer modeling processor 220 can retrieve data for a machine learning model relating to specific control variables and state variables from the hydrocarbon reformer variable data memory 210 (for example, by using past calculations of such a model without calculating the model in real time).

[0048] Referring to Figure 3, a simplified diagram of an example of a model formed using a machine learning model is shown. As shown in the figure, performance variables can be calculated based on the inputs of the current state variables and the current control variables. The performance variable model (shown as a hemisphere in Figure 3) is a function of the current control variables and the current state variables. Based on the results of this model, such as the model shown in Figure 3, the process control unit 200 can determine and generate improved control variables based at least on the current state variables and the current control variables of the continuous catalytic reforming unit 100. For example, the input to the machine learning model is the current control variables and the current state variables, which are used to compute the performance variable model. Then, it is possible to determine and generate improved control variables using this model. For example, in Figure 3, c * This can show the current control variables, and the model can be used to generate improved control variables along the dotted line to enhance performance. These improved control variables do not need to be optimized; they just need to improve performance to some extent within the range predicted by the machine learning model.

[0049] As described herein, machine learning models can be trained using inputs that include both past state variables and past control variables, and predictive modeling of the process control unit 200 can generate improved control variables based on both current control variables and current state variables. In at least some embodiments, using a machine learning model that utilizes at least state variables and control variables can result in a more accurate modeling of improved control variables compared to using only control variables or only state variables.

[0050] In some embodiments, the control operation of the chemical processing unit 104 by the process control unit 200 can be performed in real time or near real time. That is, for example, one or more signals indicating one or more current state variables are received by the process control system in real time or near real time, one or more signals indicating one or more current control variables are received by the process control system in real time or near real time, and / or the process control system generates one or more improved control variables in real time or near real time. As described herein, real time means that the process control unit 200 performs calculations and / or actions on the chemical processing unit 104 as soon as it receives inputs such as current control variable data and current state variable data. Near real time means that the delay between the process control unit 200 and such calculations and / or actions on the chemical processing unit 104 is longer, such as a delay of several seconds, several minutes, or several hours. In some embodiments, control and adjustment of the continuous catalytic reforming unit 100 via one or more control variable actuator hardware 236 can be performed in real time. In other embodiments, the control and / or adjustment of the control variable actuator hardware 236 can be performed by input from a plant operator who has reviewed the data displayed on the hydrocarbon reformer output conversion module 230.

[0051] In some embodiments, multiple machine learning models utilizing different state and control variables are combined to determine improved control variables. In other embodiments, multiple control variables and / or multiple state variables are used within the same machine learning model.

[0052] This specification will now describe a number of non-limiting technical embodiments, which are listed below as embodiments 1 to 15.

[0053] Embodiment 1. A method for operating a continuous catalytic reforming unit, comprising the steps of: forming one or more product outflow flows through a hydrocarbon reaction logistics to a continuous catalytic reforming unit comprising at least one fluid flow preheater, at least one catalytic reactor, and at least one separation unit; executing a hydrocarbon reforming process control system comprising a hydrocarbon reforming device variable data memory storing processor executable instructions, a hydrocarbon reforming device output conversion module, and one or more predictive hydrocarbon reforming device modeling processors, wherein the one or more predictive hydrocarbon reforming device modeling processors execute the processor executable instructions and receive the process control system: one or more signals from one or more state variable actuator hardware indicating one or more current state variables, the current state variables being those that cannot be directly set in the continuous catalytic reforming unit A method comprising the steps of: receiving one or more signals which are process variables and one or more signals which indicate one or more current control variables of the continuous contact reforming unit from one or more control variable actuator hardware, wherein the current control variables are process variables which can be set directly in the continuous contact reforming unit; and causing a machine learning model to perform the steps of: generating improved control variables which increase a selected performance variable based on the input of one or more current state variables or one or more current control variables, wherein the machine learning model has been trained using at least historical state variable data, historical control variable data, and historical performance variable data as inputs; and adjusting one or more current control variables of the continuous contact reforming unit based on the improved control variables determined by the machine learning model.

[0054] Embodiment 2. The method of Embodiment 1, wherein the hydrocarbon reformer output conversion module transmits one or more signals to one or more control variable actuator hardware to adjust the one or more current control variables.

[0055] Embodiment 3. A method according to any of the preceding embodiments, wherein the process control system automatically adjusts the one or more current control variables in real time based on the one or more improved control variables determined by the machine learning model.

[0056] Embodiment 4. A method according to any of the preceding embodiments, wherein one or more signals indicating the one or more current state variables are received by the process control system in real time or near real time, and one or more signals indicating the one or more current control variables are received by the process control system in real time or near real time.

[0057] Embodiment 5. A method according to any of the preceding embodiments, wherein the process control system generates the one or more improved control variables in real time or near real time.

[0058] Embodiment 6. A method according to any of the preceding embodiments, wherein one or more state variables are selected from the rate of reaction fluid input, catalyst activity, average pressure of the catalytic reactor, reactor downtime, reformed oil production yield, hydrogen production yield, off-gas production yield, LPG production yield, fuel gas consumption, cooling water consumption, power consumption, temperature of the separation unit, heat exchanger load on the composite fluid of raw materials and emissions, or catalyst residence time in the system.

[0059] Embodiment 7. A method according to any of the preceding embodiments, wherein one or more control variables are selected from the weighted average inlet temperature of the reaction flow, the supply rate of the reaction flow, the quality of the reactant feed flow, the chloride injection rate, the average carbon combustion rate in the catalyst regenerator, the catalyst circulation rate, the hydrogen / hydrocarbon ratio, and the energy consumption in the fluid flow preheater.

[0060] Embodiment 8. A method according to any of the preceding embodiments, wherein the selected performance variable is selected from the reformed oil yield, hydrogen yield, off-gas yield, LPG yield, reaction logistics supply rate, chloride injection rate, or catalyst residence time in the system.

[0061] Embodiment 9. The method of Embodiment 8, wherein the selected performance variable is the yield of the reformed oil.

[0062] Embodiment 10. The method of Embodiment 8, wherein the selected performance variable is the hydrogen production yield.

[0063] Embodiment 11. The method of Embodiment 8, wherein the selected performance variable is selected from the off-gas production yield or the LPG production yield.

[0064] Embodiment 12. The method of Embodiment 8, wherein the selected performance variable is selected from the supply rate of reaction fluid, the chloride injection rate, or the residence time of the catalyst in the system.

[0065] Embodiment 13. The method according to any of the preceding embodiments, wherein the reaction logistics include refinery naphtha.

[0066] Embodiment 14. A method according to any of the preceding embodiments, wherein the machine learning model includes a neural network model.

[0067] Embodiment 15. A continuous contact reforming unit operated by the method of any earlier embodiment.

[0068] The subject matter of this disclosure has been described in detail with reference to specific embodiments. It should be understood that no detailed description of any component or feature of an embodiment necessarily implies that such component or feature is essential to that particular embodiment or any other embodiment. Furthermore, it should be apparent to those skilled in the art that various modifications and changes can be made to the described embodiments without departing from the spirit and scope of the subject matter described in the claims. [Explanation of Symbols]

[0069] 100 Continuous Contact Modification Units 102 Hydrocarbon Reaction Logistics 104 Chemical Processing Department 110 Fluid flow preheater 112 Heated reaction logistics 114 Catalytic reactor effluent 130 Catalytic reactor 132 Intermediate catalyst reactor effluent 134 Final catalytic reactor effluent 150 Separation Units 152 Off-gas 154 Hydrogen-rich gas stream 156 LPG flow 158 Modified oil flow 160 Product effluent 170 Catalyst regenerator 172, 174, 176 catalyst flow 180 First set of catalytic reactor 182 Second set of catalytic reactor 190 Hydrogen 200 Hydrocarbon Reforming Plant Process Control Unit 210 Hydrocarbon Reformer Variable Data Memory 220 Predictive Hydrocarbon Reforming Unit Modeling Processor 230 Hydrocarbon Reformer Output Conversion Module 232 Control Variable Monitoring Hardware 234 State Variable Monitoring Hardware 236 Control Variable Actuator Hardware 272, 274 Receiver Modules

Claims

1. In a method for operating a continuous contact reforming unit, A step of forming one or more product outflow streams through a hydrocarbon reaction flow to a continuous catalytic reforming unit comprising at least one fluid flow preheater, at least one catalytic reactor, and at least one separation unit; A step of executing a hydrocarbon reformer process control system comprising a hydrocarbon reformer variable data memory storing processor-executable instructions, a hydrocarbon reformer output conversion module, and one or more predictive hydrocarbon reformer modeling processors, wherein the one or more predictive hydrocarbon reformer modeling processors execute the processor-executable instructions and provide the process control system: The receiving step, One or more signals from one or more state variable actuator hardware indicating one or more current state variables, wherein the current state variables are process variables that cannot be directly set in the continuous contact reforming unit, and One or more signals from one or more control variable actuator hardware indicating one or more current control variables of the continuous contact reforming unit, wherein the current control variables are process variables that can be directly set in the continuous contact reforming unit, and The step of receiving, A step of using a machine learning model to generate improved control variables that increase a selected performance variable based on one or both of one or more current state variables or one or more current control variables as input, wherein the machine learning model is trained using at least past state variable data, past control variable data, and past performance variable data as input. The process of causing to execute; and A step of adjusting one or more current control variables of the continuous catalytic reforming unit based on the improved control variables determined by the machine learning model; A method that includes this.

2. The method according to claim 1, wherein the hydrocarbon reformer output conversion module transmits one or more signals to one or more control variable actuator hardware to adjust the one or more current control variables.

3. The method according to claim 1 or 2, wherein the process control system automatically adjusts the one or more current control variables in real time based on the one or more improved control variables determined by the machine learning model.

4. One or more signals indicating one or more of the aforementioned current state variables are received by the process control system in real time or near real time. The method according to any one of claims 1 to 3, wherein one or more signals indicating the one or more current control variables are received by the process control system in real time or near real time.

5. The method according to any one of claims 1 to 4, wherein the process control system generates the one or more improved control variables in real time or near real time.

6. The method according to any one of claims 1 to 5, wherein the one or more state variables are selected from the rate of reaction fluid input, catalyst activity, average pressure of the catalytic reactor, reactor downtime, reformed oil production yield, hydrogen production yield, off-gas production yield, LPG production yield, fuel gas consumption, cooling water consumption, power consumption, temperature of the separation unit, heat exchanger load on the composite fluid of raw materials and emissions, or catalyst residence time in the system.

7. The method according to any one of claims 1 to 6, wherein the one or more control variables are selected from the weighted average inlet temperature of the reaction flow, the supply rate of the reaction flow, the quality of the reactant supply flow, the chloride injection rate, the average carbon combustion rate in the catalyst regenerator, the catalyst circulation rate, the hydrogen / hydrocarbon ratio, and the energy consumption in the fluid flow preheater.

8. The method according to any one of claims 1 to 7, wherein the selected performance variable is selected from the reformed oil production yield, hydrogen production yield, off-gas production yield, LPG production yield, reaction logistics supply rate, chloride injection rate, or catalyst residence time in the system.

9. The method according to claim 8, wherein the selected performance variable is the yield of the reformed oil.

10. The method according to claim 8, wherein the selected performance variable is the hydrogen production yield.

11. The method according to claim 8, wherein the selected performance variable is selected from the off-gas production yield or the LPG production yield.

12. The method according to claim 8, wherein the selected performance variable is selected from the supply rate of reaction fluid, the chloride injection rate, or the residence time of the catalyst in the system.

13. The method according to any one of claims 1 to 12, wherein the reaction logistics include refinery naphtha.

14. The method according to any one of claims 1 to 13, wherein the machine learning model includes a neural network model.

15. A continuous contact reforming unit operated by the method described in any one of claims 1 to 14.

Images