Training method of fuel valve position determination model in cracking furnace raw material switching process

Abstract

The invention provides a training method and a training device of a fuel valve position determination model in a cracking furnace raw material switching process, a raw material switching control method of a cracking furnace, electronic equipment and a readable storage medium. The training method comprises the following steps: acquiring a training data set and a verification data set based on historical data in a cracking furnace raw material switching process, wherein the training data set and the verification data set both comprise corresponding data of multiple groups of fuel valve positions and feed valve positions; constructing a multiple regression model by taking the valve position of the fuel valve as a dependent variable and the valve position of the feed valve as an independent variable; and training the constructed multiple regression model based on corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model. According to the method, the fuel valve position determination model is obtained through machine learning, so that the control process of the fuel valve position does not depend on the skill level of an operator, is not affected by the material flow and is only related to the valve position of the feeding valve, and the temperature stability of a hearth is improved.

Description

Technical Field

[0001] The invention relates to the technical field of petrochemical industry, and in particular to a training method and a training device for a fuel valve position determination model during raw material switching of a cracking furnace, a raw material switching control method of the cracking furnace, an electronic device and a readable storage medium. Background Art

[0002] According to the requirements of the production process, the ethylene cracking furnace needs to be decoked periodically, or when decoking is not required, a certain cracking furnace will also be switched for raw materials for material balance. The above processes all involve the process of withdrawing and feeding materials from the cracking furnace.

[0003] At present, the material withdrawal and feeding process of cracking furnaces in most enterprises are manually operated according to the start and stop cards. Due to different operators, the feeding and withdrawal methods are inconsistent, which will lead to changes in the cracking temperature COT and reduce the production of ethylene and propylene.

[0004] In some enterprises, the material withdrawal and feeding processes of the cracking furnace are automatically controlled according to the detected material flow to replace the traditional manual operation, which can effectively improve the operation efficiency of the device. However, due to the large fluctuations in the material flow during the withdrawal and feeding process, it is very unstable, which makes the automatic control process implemented according to the material flow low in accuracy, and also leads to unstable furnace temperature, which is not conducive to the smooth progress of the production process.

[0005] In view of this, the present invention is proposed. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome at least some of the shortcomings of the prior art, and provide a training method for a fuel valve position determination model during the raw material switching process of a cracking furnace. A multivariate regression model with the fuel valve position of the cracking furnace as the dependent variable and the feed valve position of the cracking furnace as the independent variable is trained based on historical data during the raw material switching process of the cracking furnace, so as to obtain a fuel valve position determination model, which is used to control the fuel valve position during the withdrawal process and the feeding process of the cracking furnace, so that the control process of the fuel valve position is neither dependent on the skill level of the operator nor affected by the material flow rate, but is only related to the feed valve position, thereby improving the furnace temperature stability.

[0007] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0008] A training method for a fuel valve position determination model during a cracking furnace raw material switching process, comprising:

[0009] Acquire a training data set and a verification data set based on historical data during the cracking furnace raw material switching process, wherein the training data set and the verification data set both include corresponding data of multiple groups of fuel valve positions and feed valve positions;

[0010] Constructing a multiple regression model with the fuel valve position as the dependent variable and the feed valve position as the independent variable; and

[0011] The constructed multivariate regression model is trained based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model.

[0012] In some embodiments, the step of constructing a multivariate regression model with the fuel valve position as the dependent variable and the feed valve position as the independent variable includes:

[0013] A plurality of options are set for the highest order terms of the multivariate regression model to obtain a plurality of multivariate regression sub-models, each of which takes the fuel valve position as a dependent variable and the feed valve position as an independent variable.

[0014] In some embodiments, the step of training the constructed multivariate regression model based on the corresponding data of the plurality of groups of fuel valve positions and feed valve positions in the training data set and the validation data set comprises:

[0015] Inputting corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set into multiple multivariate regression sub-models to calculate multiple first cost functions;

[0016] Inputting corresponding data of multiple groups of fuel valve positions and feed valve positions in the verification data set into multiple multivariate regression sub-models to calculate multiple second cost functions;

[0017] A multivariate regression sub-model in which both the plurality of first cost functions and the plurality of second cost functions are minimized is determined as the fuel valve position determination model.

[0018] In some implementations, the first cost function and the second cost function are mean square error functions.

[0019] The present invention also provides a training device for a fuel valve position determination model during a cracking furnace raw material switching process, comprising:

[0020] An acquisition module, used to acquire a training data set and a verification data set based on historical data in the cracking furnace raw material switching process, wherein the training data set and the verification data set both include a plurality of sets of corresponding data of fuel valve positions and feed valve positions;

[0021] A construction module, for constructing a multivariate regression model with the fuel valve position as a dependent variable and the feed valve position as an independent variable; and

[0022] The training module is used to train the constructed multivariate regression model based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model.

[0023] The present invention also provides a raw material switching control method for a cracking furnace, comprising:

[0024] During the process of raw material withdrawal or feeding, the following steps are continuously performed:

[0025] Retrieve the current valve position of the feed valve of the cracking furnace;

[0026] Inputting the retrieved current valve position of the feed valve into the fuel valve position determination model trained based on the training method of the fuel valve position determination model in the cracking furnace raw material switching process described above to obtain the required valve position of the fuel valve; and

[0027] The valve position of the fuel valve of the cracking furnace is controlled based on the required valve position of the fuel valve so that the furnace temperature reaches a target value.

[0028] In some embodiments, the required valve position of the fuel valve in the fuel valve position determination model and the current valve position of the feed valve satisfy the following relationship:

[0029] CV RL =f(CV JL )

[0030] Among them, CV RL represents the desired valve position of the fuel valve, CV JL represents the current valve position of the feed valve, f is a mapping function, and the function form is a polynomial form.

[0031] In some embodiments, before the step of calling the current valve position of the feed valve of the cracking furnace, the control method further includes:

[0032] Obtaining the type of raw material for the cracking furnace; and

[0033] Based on the type of the raw material of the cracking furnace, a fuel valve position determination model matching the type of the raw material is retrieved.

[0034] The present invention also provides an electronic device, comprising:

[0035] Processor; and

[0036] A memory, communicatively connected to the processor;

[0037] In which, the memory stores a program that can be executed by the processor. When the program is executed by the processor, the processor can execute the training method of the fuel valve position determination model during the cracking furnace raw material switching process described above, or should be based on the raw material switching control method of the cracking furnace described above.

[0038] The present invention also provides a readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, a training method for determining a fuel valve position model during the cracking furnace raw material switching process described above is implemented, or the raw material switching control method for the cracking furnace described above is implemented.

[0039] After adopting the above technical scheme, the present invention has the following beneficial effects compared with the prior art.

[0040] The present invention provides a training method for a fuel valve position determination model during a cracking furnace raw material switching process. The method trains a multiple regression model with the fuel valve position of the cracking furnace as a dependent variable and the feed valve position of the cracking furnace as an independent variable based on historical data during the cracking furnace raw material switching process, thereby obtaining a fuel valve position determination model for controlling the fuel valve position during the material withdrawal process and the material feeding process of the cracking furnace, so that the control process of the fuel valve position is neither dependent on the skill level of the operator nor affected by the material flow rate, but is only related to the feed valve position, thereby improving the furnace temperature stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the accompanying drawings:

[0042] Figure 1 is a flow chart of a method for training a fuel valve position determination model in a cracking furnace raw material switching process according to an exemplary embodiment of the present invention;

[0043] Figure 2 is a flow chart of step S130 according to an exemplary embodiment of the present invention;

[0044] Figure 3 is a schematic diagram of a training device for a fuel valve position determination model in a cracking furnace raw material switching process according to an exemplary embodiment of the present invention;

[0045] Figure 4 is a schematic flow diagram of a raw material switching control method for a cracking furnace according to an exemplary embodiment of the present invention;

[0046] Figure 5 is a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention.

[0047] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0049] In the description of the present invention, it should be noted that the directions or positional relationships indicated by terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0050] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, unless explicitly limited or contradictory to the context, the specific steps included in the method described in the present invention are not necessarily limited to the order described, but can be performed in any order or in parallel.

[0052] Figure 1 The flowchart of a method 100 for training a fuel valve position determination model during a cracking furnace feedstock switching process according to an exemplary embodiment of the present invention is shown.

[0053] like Figure 1 As shown, the execution of the training method 100 includes the following steps:

[0054] S110, acquiring a training data set and a verification data set based on historical data during the cracking furnace raw material switching process, wherein the training data set and the verification data set both include a plurality of sets of corresponding data of fuel valve positions and feed valve positions;

[0055] S120, constructing a multivariate regression model with the fuel valve position as a dependent variable and the feed valve position as an independent variable; and

[0056] S130, training the constructed multivariate regression model based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model.

[0057] It should be understood that the steps shown in the training method 100 are not exclusive, and the training method 100 may further include additional steps not shown and / or may omit the steps shown, and the scope of the present invention is not limited in this respect.

[0058] The types of cracking furnace raw materials can be divided into heavy raw materials, light raw materials and gas raw materials. As an example, heavy raw materials refer to heavy liquid raw materials such as hydrocracking tail oil (HTO), light raw materials refer to light liquid raw materials such as light naphtha, hydrogenated coking naphtha, hydrogenated top oil, rich n-pentane, and aromatic raffinate oil, and gas raw materials refer to ethane (C2 mixed gas raw materials) and butane (C4 mixed gas raw materials), wherein the C2 mixed gas raw materials are mixed by refinery rich ethane gas, propane dehydrogenation ethane tail gas, and circulating ethane.

[0059] Depending on the type of raw materials, cracking furnaces can be divided into three types: heavy oil furnaces, light oil furnaces, and gas furnaces.

[0060] It can be understood that for different cracking furnace raw materials, the historical data are different, so the fuel valve position determination model obtained by training is also different. In other words, for each cracking furnace raw material, a corresponding fuel valve position determination model is trained.

[0061] Refer to the following Figure 1 to Figure 2 Taking one of the cracking furnace raw materials as an example, steps S110 to S130 of the training method of the fuel valve position determination model during the cracking furnace raw material switching process are explained.

[0062] S110

[0063] Specifically, by analyzing the operation mechanism of the cracking furnace, factors affecting the fuel valve position are obtained, such as feed valve position, COT temperature, dilution steam and other factors, and these factors are selected as characteristic values.

[0064] The minute-level feed valve position data, fuel valve position data and COT temperature data are extracted from the historical data of the past one to two years during the raw material switching process of the cracking furnace. The start and end time of the material withdrawal is determined by the feed valve position decreasing from full load to 0, the charring time period is determined by the feed valve position maintaining at 0 and the COT temperature increasing in steps, and the feeding time period is determined by the process of the feed valve position increasing from 0 to full load.

[0065] By analyzing historical data, it was found that there were multiple withdrawal and feeding processes in the past one to two years. The corresponding data of the fuel valve position and the feed valve position in the multiple withdrawal and feeding processes were divided into two groups, namely the training data set and the validation data set, which were used to train the model in the subsequent steps.

[0066] As an example, the training data set includes 3 to 5 sets of corresponding data of fuel valve positions and feed valve positions, and the verification data set includes 2 to 4 sets of corresponding data of fuel valve positions and feed valve positions. It should be noted here that there can be multiple feed valves.

[0067] It should be pointed out that, theoretically speaking, increasing the amount of data included in the training data set and the validation data set can help the model better learn the essential laws of the problem, thereby improving the generalization ability of the model. However, if the amount of data included in the training data set and the validation data set is too much, there will be a risk of overfitting. Therefore, the amount of data included in the training data set and the validation data set can be determined according to actual needs to keep it within a reasonable range, so that the performance of the trained model is more reliable.

[0068] S120

[0069] The present invention selects a multiple regression model to predict the relationship between the feed valve position and the fuel valve position.

[0070] Specifically, multiple options are set for the highest-order terms of the multiple regression model to obtain multiple multiple regression sub-models, each of which takes the fuel valve position as a dependent variable and the feed valve position as an independent variable.

[0071] In the specific implementation process, the model training uses Python's Scikit-learn library. PolynomialFeatures is first used to establish a polynomial feature model. The feature data in the polynomial feature model includes the fuel valve position data and the feed valve position data. Then the highest order term of the polynomial feature model is set. For example, if the highest order term of the polynomial feature model is set to 1 to 5, 5 polynomial feature models can be obtained, that is, 5 multivariate regression sub-models.

[0072] It should be pointed out here that, according to different raw materials in the cracking furnace, the range of the highest order item can be set according to actual conditions, and the present invention does not impose any limitation on this.

[0073] In order to facilitate training and make the training process converge faster, the feature data can also be annotated. As an example, the StandardScaler tool is used to process the feature data so that the mean value of each feature data is 0 and the variance is 1.

[0074] S130

[0075] In some embodiments, Figure 2 As shown, the execution of step S130 includes the following steps:

[0076] S131, inputting corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set into multiple multivariate regression sub-models to calculate multiple first cost functions;

[0077] S132, inputting the corresponding data of multiple groups of fuel valve positions and feed valve positions in the verification data set into multiple multivariate regression sub-models to calculate multiple second cost functions;

[0078] S133: Determine the multivariate regression sub-model in which the plurality of first cost functions and the plurality of second cost functions are all minimized as the fuel valve position determination model.

[0079] Specifically, taking the five multivariate regression sub-models constructed in the above step S120 as an example, the mean square error mean_squared_error is selected as the cost function. In step S131, the corresponding data of the multiple groups of fuel valve positions and the feed valve positions in the training data set are respectively input into the five multivariate regression sub-models to calculate five first cost functions. In step S132, the corresponding data of the multiple groups of fuel valve positions and the feed valve positions in the verification data set are respectively input into the five multivariate regression sub-models to calculate five second cost functions.

[0080] In step S133, the first cost function and the second cost function calculated for each multivariate regression sub-model are compared, and the multivariate regression sub-model corresponding to the time when the first cost function and the second cost function are both minimum is determined as the fuel valve position determination model of the cracking furnace.

[0081] Figure 3 A schematic diagram of a training device 200 for a fuel valve position determination model during a cracking furnace feedstock switching process according to an exemplary embodiment of the present invention is shown.

[0082] like Figure 3As shown, the training device 200 includes an acquisition module, a construction module and a training module. The acquisition module is used to acquire a training data set and a verification data set based on historical data in the cracking furnace raw material switching process, wherein the training data set and the verification data set both include a plurality of groups of corresponding data of the fuel valve position and the feed valve position; the construction module is used to construct a multivariate regression model with the fuel valve position as a dependent variable and the feed valve position as an independent variable; the training module is used to train the constructed multivariate regression model based on the plurality of groups of corresponding data of the fuel valve position and the feed valve position in the training data set and the verification data set, and obtain a fuel valve position determination model.

[0083] The training device 200 outputs the trained fuel valve position determination model to the raw material switching control device of the cracking furnace so that it can be retrieved during the process of the cracking furnace executing raw material withdrawal or feeding.

[0084] The process of training the fuel valve position determination model by the training device 200 of the present invention can refer to the above-mentioned method based on Figure 1 to Figure 2 The present invention will not elaborate on the training method 100 of the fuel valve position determination model during the cracking furnace raw material switching process.

[0085] Figure 4 The flowchart of a raw material switching control method 300 for a cracking furnace according to an exemplary embodiment of the present invention is shown.

[0086] like Figure 4 As shown, the following steps are continuously performed during the process of raw material withdrawal or feeding:

[0087] S310, calling the current valve position of the feed valve of the cracking furnace;

[0088] S320, inputting the retrieved current valve position of the feed valve into a trained fuel valve position determination model to obtain a required valve position of the fuel valve; and

[0089] S330. Control the valve position of the fuel valve of the cracking furnace based on the required valve position of the fuel valve so that the furnace temperature reaches a target value.

[0090] It should be understood that the steps shown in the cracking furnace raw material switching control method 300 are not exclusive. The cracking furnace raw material switching control method 300 may also include additional steps not shown and / or may omit the steps shown. The scope of the present invention is not limited in this respect.

[0091] The following describes steps S310 to S330 of the raw material switching control method 300 for the cracking furnace by taking the raw material feeding process as an example.

[0092] It should be pointed out that before feeding raw materials, various inspections should be carried out, such as the feed valve position, feed flow meter manual valve, cracking raw material type, feed flow pressure, cracking gas large valve position, COT deviation, and the operator should confirm each condition or bypass through the interface to ensure that the system meets the necessary conditions for automatic feeding.

[0093] After the raw material feeding starts, the valve position of the cracking furnace feed valve is opened at a preset speed to allow the raw material to enter the cracking furnace through the feed valve. While opening the valve position of the feed valve, in step S310, the raw material switching control device can retrieve the current valve position of the feed valve of the cracking furnace. In step S320, the raw material switching control device inputs the retrieved current valve position of the feed valve into the trained fuel valve position determination model. The fuel valve position determination model obtains the required valve position of the fuel valve by calculation. In step S330, the raw material switching control device forms a control signal for controlling the fuel valve based on the obtained required valve position of the fuel valve, and sends the control signal to the fuel valve, so that the valve position of the fuel valve is opened to the required valve position, thereby making the furnace temperature of the cracking furnace reach the target value.

[0094] The following describes steps S310 to S330 of the raw material switching control method 300 for a cracking furnace by taking the raw material withdrawal process as an example.

[0095] It should be pointed out that before the raw materials are returned, various inspections should be carried out, such as checking all pipelines of the cracking furnace and related equipment, ensuring that the inside of the cracking furnace has been depressurized at low temperature, and ensuring that the return pipeline is not blocked or leaking, so that the system meets the necessary conditions for automatic feeding.

[0096] After the raw material withdrawal begins, the feed valve position of the cracking furnace is opened at a preset speed so that the raw material can be discharged from the cracking furnace through the feed valve. While opening the feed valve position, in step S310, the raw material switching control device can retrieve the current valve position of the feed valve of the cracking furnace. In step S320, the raw material switching control device inputs the retrieved current valve position of the feed valve into the trained fuel valve position determination model. The fuel valve position determination model obtains the required valve position of the fuel valve by calculation. In step S330, the raw material switching control device forms a control signal for controlling the fuel valve based on the obtained required valve position of the fuel valve, and sends the control signal to the fuel valve, so that the valve position of the fuel valve is opened to the required valve position, thereby making the furnace temperature of the cracking furnace reach the target value.

[0097] In some embodiments, in step S320, the required valve position of the fuel valve in the fuel valve position determination model and the current valve position of the feed valve satisfy the following relationship:

[0098] CV RL =f(CVJL ) ①

[0099] Among them, CV RL represents the desired valve position of the fuel valve, CV JL represents the current valve position of the feed valve, f is a mapping function, and the function form is a polynomial form.

[0100] In some embodiments, the cracking furnace may have multiple feed valves. Then, the required valve position of the fuel valve in the fuel valve position determination model and the current valve position of the feed valve satisfy the following relationship:

[0101] CV RL =f(CV JL1、 CV JL2、 CV JL3、 ……CV JLn ) ②

[0102] Among them, CV RL represents the desired valve position of the fuel valve, CV JL1 ~CV JLn represents the current valve position of each feed valve, f is the mapping function, and the function form is a polynomial form.

[0103] In addition, before calling the current valve position of the feed valve of the cracking furnace in step S310, the execution body also needs to obtain the type of raw material of the cracking furnace; and based on the type of raw material of the cracking furnace, call the fuel valve position determination model that matches the type of raw material.

[0104] In the above scheme, the present invention controls the position of the fuel valve during the withdrawal process and the feeding process of the cracking furnace, which is neither dependent on the skill level of the operator nor affected by the material flow rate, and is only related to the position of the feed valve, thereby improving the stability of the furnace temperature.

[0105] In some embodiments, the feed valve position can be controlled to open at different preset speeds. As an example, the feed valve position can be opened from 0 to X1 at a faster preset speed, and then opened from X1 to X2 at a relatively slower preset speed. This is because in the early stage of feeding the cracking furnace, the furnace tube is full of steam. When the first batch of materials enters, the hydrocarbon partial pressure is very low, which leads to a long residence time, causing the first batch of materials to easily react in the furnace tube to form coke. Therefore, the first batch of materials fed is relatively large, which is also one of the reasons why the amount of steam in the early stage is maintained large and the amount of equilibrium steam is greater than the amount of mixed steam. During the withdrawal of materials, the oxygen content will definitely increase, because the amount of material in the furnace tube is getting less and less, and the required fuel gas will also be reduced accordingly. If the bottom damper and the furnace negative pressure fan baffle are not adjusted, it is easy for the NOx value caused by the oxygen content to exceed the index.

[0106] The above scheme of the present invention can reduce the material input and output time and improve production efficiency under the premise of ensuring safe production.

[0107] It should be pointed out that the process of controlling the position of the feed valve to open at a preset speed during the material withdrawal process may adopt the commonly used technology in the art, and the present invention will not be described in detail here.

[0108] In some embodiments, the material switching control device can be packaged as a separate control module, which can transmit signals with the original control system, so as to improve the convenience of upgrading the original system without replacing the original control system.

[0109] Figure 5 The structure of an electronic device provided according to an exemplary embodiment of the present invention is shown.

[0110] like Figure 5 As shown, the electronic device 400 includes a processor 401 and a memory 402. The memory 402 is in communication connection with the processor 401. The memory 402 stores a program executable by the processor. When the program is executed by the processor, the processor 401 can execute the training method of the fuel valve position determination model in the cracking furnace raw material switching process and the raw material switching control method of the cracking furnace.

[0111] Figure 5 The electronic device shown further includes a bus 403 and a communication interface 404 , and the processor 401 , the communication interface 404 and the memory 402 are connected via the bus 403 .

[0112] The memory 402 may include a high-speed random access memory (RAM), and may also include a non-volatile memory 402 (Non-Volatile Memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 404 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 403 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus 403 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus 403 or only one type of bus 403 .

[0113] The processor 401 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 401 or the instruction in the form of software. The above processor 401 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor 401 can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 402, and the processor 401 reads the information in the memory 402 and completes the steps of the method of the above embodiment in combination with its hardware.

[0114] An exemplary embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is called and executed by the processor 401, the computer executable instructions prompt the processor 401 to implement the training method of the fuel valve position determination model during the raw material switching process of the cracking furnace and the raw material switching control method of the cracking furnace. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0115] The computer program product of the training method of the fuel valve position determination model during the raw material switching of the cracking furnace and the raw material switching control method of the cracking furnace provided in the embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method in the previous method embodiment. The specific implementation can be found in the method embodiment, which will not be repeated here.

[0116] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and / or electronic device can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0117] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not used to limit the present invention. Any technician familiar with this patent can make some changes or modify the technical contents suggested above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the solution of the present invention.

Claims

1. A training method for a fuel valve position determination model in a cracking furnace raw material switching process, characterized in that: include: Acquire a training data set and a verification data set based on historical data during the cracking furnace raw material switching process, wherein the training data set and the verification data set both include a plurality of sets of corresponding data of fuel valve positions and feed valve positions; Constructing a multiple regression model with the fuel valve position as the dependent variable and the feed valve position as the independent variable; and The constructed multivariate regression model is trained based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model.

2. The training method for the fuel valve position determination model in the cracking furnace raw material switching process according to claim 1 is characterized in that: The steps of constructing a multivariate regression model with the fuel valve position as the dependent variable and the feed valve position as the independent variable include: A plurality of options are set for the highest order terms of the multivariate regression model to obtain a plurality of multivariate regression sub-models, each of which takes the fuel valve position as a dependent variable and the feed valve position as an independent variable.

3. The training method for determining the fuel valve position model in the cracking furnace raw material switching process according to claim 2, characterized in that: The step of training the constructed multivariate regression model based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set includes: Inputting corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set into multiple multivariate regression sub-models to calculate multiple first cost functions; Inputting corresponding data of multiple groups of fuel valve positions and feed valve positions in the verification data set into multiple multivariate regression sub-models to calculate multiple second cost functions; A multivariate regression sub-model in which both the plurality of first cost functions and the plurality of second cost functions are minimized is determined as the fuel valve position determination model.

4. The training method for the fuel valve position determination model in the cracking furnace raw material switching process according to claim 3 is characterized in that: The first cost function and the second cost function are mean square error functions.

5. A training device for a fuel valve position determination model during a cracking furnace raw material switching process, characterized in that: include: An acquisition module, used to acquire a training data set and a verification data set based on historical data in the cracking furnace raw material switching process, wherein the training data set and the verification data set both include a plurality of sets of corresponding data of fuel valve positions and feed valve positions; A construction module, for constructing a multivariate regression model with the fuel valve position as a dependent variable and the feed valve position as an independent variable; and The training module is used to train the constructed multivariate regression model based on the corresponding data of multiple groups of fuel valve positions and feed valve positions in the training data set and the verification data set to obtain a fuel valve position determination model.

6. A method for controlling raw material switching of a cracking furnace, characterized in that: include: During the process of raw material withdrawal or feeding, the following steps are continuously performed: Retrieve the current valve position of the feed valve of the cracking furnace; Inputting the retrieved current valve position of the feed valve into a fuel valve position determination model trained by the training method for a fuel valve position determination model in a cracking furnace raw material switching process according to any one of claims 1 to 4 to obtain a desired valve position of the fuel valve; and The valve position of the fuel valve of the cracking furnace is controlled based on the required valve position of the fuel valve so that the furnace temperature reaches a target value.

7. The method for controlling raw material switching of a cracking furnace according to claim 6, characterized in that: The required valve position of the fuel valve in the fuel valve position determination model and the current valve position of the feed valve satisfy the following relationship: CV RL =f(CV JL ) Among them, CV RL represents the desired valve position of the fuel valve, CV JL represents the current valve position of the feed valve, f is a mapping function, and the function form is a polynomial form.

8. The raw material switching control method for a cracking furnace according to claim 6 or 7, characterized in that: Before the step of adjusting the current valve position of the feed valve of the cracking furnace, the control method further comprises: Obtaining the type of raw material for the cracking furnace; and Based on the type of the raw material of the cracking furnace, a fuel valve position determination model matching the type of the raw material is retrieved.

9. An electronic device, characterized in that: include: processor; as well as A memory, communicatively connected to the processor; In which, the memory stores a program that can be executed by the processor. When the program is executed by the processor, the processor can execute the training method of the fuel valve position determination model during the raw material switching of the cracking furnace according to any one of claims 1 to 4, or the raw material switching control method of the cracking furnace according to any one of claims 6 to 8.

10. A readable storage medium, characterized in that: A computer program is stored on the readable storage medium, and when the computer program is executed by the processor, a training method for a fuel valve position determination model during a cracking furnace raw material switching process according to any one of claims 1 to 4 is implemented, or a raw material switching control method for a cracking furnace according to any one of claims 6 to 8 is implemented.

Images