Design method and system for generating petrochemical engineering non-standard equipment based on deep learning

Abstract

The invention provides a design method and system for generating petrochemical engineering non-standard equipment based on deep learning, relates to the technical field of non-standard pressure vessel design, and aims to solve the problems that an automatic generation design method is poor in interactivity to users; the design method comprises the following steps: S1, acquiring process parameters and a standard calculation formula in a pressure vessel production process; s2, constructing a process parameter database; s3, training according to the process parameter database to obtain a pressure vessel design model; s4, inputting the product performance requirements of the target pressure vessel into the pressure vessel design model as input parameters to obtain module parameters of the target pressure vessel displayed on the simulation software; and S5, receiving and identifying the content adjustment of the user on the module parameters on the simulation software, and feeding back the content adjustment to the pressure vessel design model to obtain new module parameters, so that the interactivity between the user and the pressure vessel design model can be enhanced, and the experience feeling is good.

Description

Technical Field

[0001] The present invention relates to the technical field of non-standard pressure vessel design, and in particular to a design method and system for generating non-standard equipment for petrochemical industry based on deep learning. Background Art

[0002] Since non-standardized pressure vessels need to meet special working conditions, complex structures or customized requirements, their design often involves coupling optimization between multiple parameters, adaptability between new materials, and high-intensity safety verification. Traditional design methods rely on manual experience and continuous trial and error, which are inefficient and difficult to balance between multiple objectives. Therefore, it is currently proposed to use deep learning methods to learn a large number of parameters of pressure vessels, which can automatically generate non-standardized pressure vessels. However, this automatic generation method requires a lot of training to obtain the non-standardized pressure vessels that users want. After each automatic generation, it is inconvenient for users to adjust the model results, and the user experience is poor and the interactivity is not good. Summary of the invention

[0003] The purpose of the present invention is to provide a design method for generating non-standard petrochemical equipment based on deep learning. Deep learning is used to perform massive analysis of process parameters in the pressure vessel production process, which can quickly integrate pressure vessel standards and real-time process parameters, greatly shorten the design cycle, and reduce dependence on manual labor.

[0004] In order to solve the above technical problems, the present invention adopts the following solutions: A design method for generating non-standard equipment for petrochemical industry based on deep learning, comprising the following steps: S1. Obtain the process parameters and related standard calculation formulas in the pressure vessel production process; S2. Perform structured storage according to the process parameters and their data types, and associate the corresponding process parameters through relevant standard calculation formulas to obtain a process parameter database; S3. Construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and perform iterative training using the relevant standard calculation formula as constraint conditions to obtain a pressure vessel design model; S4. Inputting the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model to obtain module parameters of the target pressure vessel displayed on the simulation software; S5. Receive and identify the content adjustment of the module parameters by the user on the simulation software, feed back the content adjustment to the pressure vessel design model, update the pressure vessel design model, and obtain new module parameters.

[0005] Furthermore, in S1, the process of obtaining the process parameters in the pressure vessel production process and their related standard calculation formulas is specifically: obtaining the process spreadsheet file recorded in the pressure vessel production process, performing content recognition processing on the process spreadsheet file, extracting the process parameters, and obtaining the related standard calculation formulas according to the process rules corresponding to the process parameters.

[0006] Furthermore, the step S2 includes the following steps: S21. According to the data types of the process parameters in the pressure vessel production process, all process parameters are type-marked, and one data type corresponds to a number of process parameters; S22. Construct a process parameter database, use the data type as the library name of the sub-library in the process parameter database, store the corresponding process parameters in the sub-library, and store the process parameters of each process table file in isolation in the sub-library, number the process parameters of the same process table file, and associate the process parameters with the same number through relevant standard calculation formulas to obtain a process parameter database.

[0007] Furthermore, the data types include container geometry, head geometry, interface design, and material type. The container geometry includes cylindrical, spherical, and special-shaped. The head geometry includes elliptical, spherical, and flat. The interface design includes inlet and outlet pipe diameters, flange size, and connection method. The material types include carbon steel, stainless steel, titanium alloy, and composite materials.

[0008] Furthermore, in S22, the process of associating process parameters with the same number through relevant standard calculation formulas is specifically as follows: In the process parameter database, the sub-libraries are numbered in sequence, and the position information of each process parameter is obtained through the numbering of the sub-library and the process parameter. The position information refers to the coordinate position formed by the sub-library where the process parameter is located and the position information inside the sub-library. Then, the coordinate position of each process parameter is associated according to the relevant standard calculation formula to form a coordinate association table.

[0009] Furthermore, in S3, an initial deep learning model is constructed, and the process parameters in the process parameter database are input as input parameters into the initial deep learning model according to the association relationship, and the relevant standard calculation formulas are used as constraints for iterative training to obtain the pressure vessel design model. The process is as follows: An initial deep learning model is constructed, and the corresponding process parameters are found according to the association relationship. Multiple training data sets are obtained using the process parameters and the corresponding historical design requirement data. The multiple training data sets are input into the initial deep learning model for iterative training. In each iteration, the generated process parameters are constrained and checked through relevant standard calculation formulas, so that the trained deep learning model can be used as a pressure vessel design model.

[0010] Furthermore, the deep learning model includes a generative adversarial network, a convolutional neural network, and multi-objective optimization. A non-standardized container geometry that meets the design requirements is generated through a generative adversarial network, and material properties are predicted based on a process parameter database through a convolutional neural network. Multiple training data sets are simultaneously input into the initial deep learning model for optimization through multi-objective optimization to obtain the optimal deep learning model as a pressure vessel design model.

[0011] Further, in S4, the product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the process of obtaining the module parameters of the target pressure vessel displayed on the simulation software is as follows: The product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the relevant standard calculation formulas are obtained according to the product performance requirements. The process parameters of the target pressure vessel are obtained through the relevant standard calculation formulas. The process parameters are input into the simulation software for simulation to obtain the target pressure vessel and its module parameters displayed on the simulation software. The module parameters are obtained through the simulation software, including cylinder module parameters, head module parameters, and flange module parameters.

[0012] A design system for generating non-standard equipment for petrochemical industry based on deep learning, using a design method for generating non-standard equipment for petrochemical industry based on deep learning, comprising: Data acquisition module: obtain the process parameters of the pressure vessel production process and its related standard calculation formulas; Model building module: construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and use the relevant standard calculation formulas as constraints for iterative training to obtain the pressure vessel design model; Pressure vessel design module: input the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model, and obtain the module parameters of the target pressure vessel displayed on the simulation software; Pressure vessel feedback module: Receives and identifies the user's adjustment of module parameters on the simulation software, feeds back the content adjustment to the pressure vessel design model, updates the pressure vessel design model, and obtains new module parameters.

[0013] Beneficial effects of the present invention: The present invention provides a design method and system for generating non-standard equipment for petrochemical industry based on deep learning, which changes the traditional design method relying on manual experience into a design method automatically generated by a deep learning model. In this design method, the process parameters and related standard calculation formulas are first stored in a structured manner, and each process parameter is associated with each other through structured storage. The optimal design method of the target pressure vessel can be quickly determined through the association relationship. In addition, based on deep learning, a massive analysis of the process parameters in the pressure vessel production process is performed, which can quickly integrate the pressure vessel standards and real-time process parameters, greatly shorten the design cycle, reduce dependence on manual labor, and at the same time, promote the development of high-performance and high-safety non-standard equipment to meet the stringent requirements of petrochemical, aerospace and other fields.

[0014] The system includes not only a pressure vessel design module, but also a pressure vessel feedback module. The pressure vessel design module is used to perform massive analysis of process parameters in the pressure vessel production process based on deep learning. The pressure vessel feedback module is used to receive and identify the user's content adjustment of the module parameters, and output it to the pressure vessel design module, so that the pressure vessel design module can be re-trained. The pressure vessel design module plus the pressure vessel feedback module can achieve interactivity between the user and the pressure vessel design model, which is convenient for the user to accurately generate the required non-standardized pressure vessels and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the design method in Example 1 of the present invention; Figure 2 This is a structural schematic diagram of the process parameter database in Example 1 of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless specifically stated otherwise.

[0018] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0019] Additionally, descriptions of well-known structures, functions, and configurations may be omitted for clarity and conciseness.One of ordinary skill in the art will recognize that various changes and modifications may be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0020] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0021] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0022] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments: Example 1 Since non-standardized pressure vessels need to meet special working conditions, complex structures or customized requirements, their design often involves coupling optimization between multiple parameters, adaptability between new materials and high-intensity safety verification. At present, the design method of non-standardized pressure vessels generally relies on manual experience and continuous trial and error, which is inefficient and difficult to balance between multiple objectives. Therefore, it is currently proposed to use deep learning methods to learn a large number of parameters of pressure vessels, which can automatically generate non-standardized pressure vessels. However, this automatic generation method requires a lot of training to get the non-standardized pressure vessels that users want. After each automatic generation, it is inconvenient for users to adjust the model results, and the user experience is poor and the interactivity is not good.

[0023] In order to solve the above problems, a design method for generating non-standard equipment for petrochemical industry based on deep learning is proposed in this embodiment. Figure 1 As shown, the following steps are included: S1. Obtain the process parameters and related standard calculation formulas in the pressure vessel production process; S2. Perform structured storage according to the process parameters and their data types, and associate the corresponding process parameters through relevant standard calculation formulas to obtain a process parameter database; S3. Construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and perform iterative training using the relevant standard calculation formula as constraint conditions to obtain a pressure vessel design model; S4. Inputting the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model to obtain module parameters of the target pressure vessel displayed on the simulation software; S5. Receive and identify the content adjustment of the module parameters by the user on the simulation software, feed back the content adjustment to the pressure vessel design model, update the pressure vessel design model, and obtain new module parameters.

[0024] In actual situations, the design of non-standardized pressure vessels often involves multiple parameters such as geometric parameters, operating parameters, material parameters, load parameters, safety parameters, manufacturing parameters, economic parameters, environmental parameters, and functional parameters. In addition, during the pressure vessel production process, relevant parameters are usually recorded in the form of tables to form process table files for storage.

[0025] The above parameters have different parameters for different parts of non-standardized pressure vessels. For example, the geometric parameters include not only parameters representing the geometric shape of the container, but also parameters representing the geometric shape of the head, and each parameter contains multiple content values. For example, the geometric shapes of the container include cylindrical, spherical, and special-shaped.

[0026] Therefore, in this embodiment, parameters are grouped into data types for different components and different types of non-standardized pressure vessels to obtain multiple data types. Specifically, the data types include container geometry, head geometry, interface design, the maximum temperature that the container needs to withstand, the minimum temperature that the container needs to withstand, working pressure, working temperature, material type, etc.

[0027] Then, the process parameters in the data type are isolated according to the parameter content. Specifically, the container geometry includes cylindrical, spherical, and special-shaped, the head geometry includes elliptical, spherical, and flat, the interface design includes inlet and outlet pipe diameters, flange sizes, and connection methods, and the material types include carbon steel, stainless steel, titanium alloy, and composite materials. Then, one data type corresponds to several process parameters, and the process parameters of the same data type recorded in different pressure vessel production processes will be different. Therefore, in this embodiment, the process parameters in different pressure vessel production processes can be obtained according to their historical design requirements, increasing the diversity of the database and facilitating deep learning to conduct massive analysis of process parameters in the pressure vessel production process.

[0028] In one embodiment, in S1, the process of obtaining process parameters in the pressure vessel production process and their related standard calculation formulas is specifically: obtaining a process table file recorded in the pressure vessel production process, performing content recognition processing on the process table file, extracting the process parameters, and obtaining the related standard calculation formulas according to the process rules corresponding to the process parameters.

[0029] In one embodiment, step S2 includes the following steps: S21. According to the data types of the process parameters in the pressure vessel production process, all process parameters are type-marked, and one data type corresponds to a number of process parameters; S22. Construct a process parameter database, use the data type as the library name of the sub-library in the process parameter database, store the corresponding process parameters in the sub-library, and store the process parameters of each process table file in isolation in the sub-library, number the process parameters of the same process table file, and associate the process parameters with the same number through relevant standard calculation formulas to obtain a process parameter database.

[0030] In one embodiment, in S22, the process of associating process parameters with the same number through relevant standard calculation formulas is specifically as follows: In the process parameter database, the sub-libraries are numbered in sequence, and the position information of each process parameter is obtained through the numbering of the sub-library and the process parameter. The position information refers to the coordinate position formed by the sub-library where the process parameter is located and the position information inside the sub-library. Then, the coordinate position of each process parameter is associated according to the relevant standard calculation formula to form a coordinate association table.

[0031] Specifically, Figure 2 As shown, the process parameter database uses the data type as the library name of the sub-library in the process parameter database, and each sub-library is isolated from each other, including two sub-libraries such as data type 1 and data type 2. Each sub-library contains its own corresponding process parameters, and the process parameters are stored in isolation in the sub-library according to the process table file. For example, data type 1 contains process parameter 1 numbered A and process parameter 2 numbered B. At this time, if data type 2 contains process parameter 1 numbered A and process parameter 2 numbered C, and the relevant standard calculation formula contains data type 1 and data type 2, then the process parameters 1 and data type 2 in data type 1 with the same number will be stored separately. The sub-libraries can be associated with the process parameter 1 in data type 2. Before the association, the sub-libraries can be numbered. For example, data type 1 is numbered a and data type 2 is numbered b. The position information of each process parameter can be located through the sub-libraries and the numbers of the process parameters inside the sub-libraries. For example, the position information of process parameter 1 in data type 1 is (a, A) and the position information of process parameter 1 in data type 2 is (b, A). By associating with the position information, a coordinate association table can be formed and stored in the process parameter database. Through structured storage, each process parameter has an association relationship. Through the association relationship, the optimal design method of the target pressure vessel can be quickly determined.

[0032] In one embodiment, in S3, an initial deep learning model is constructed, process parameters in the process parameter database are input as input parameters into the initial deep learning model according to the association relationship, and relevant standard calculation formulas are used as constraints for iterative training to obtain the pressure vessel design model in the following process: An initial deep learning model is constructed, and the corresponding process parameters are found according to the association relationship. Multiple training data sets are obtained using the process parameters and the corresponding historical design requirement data. The multiple training data sets are input into the initial deep learning model for iterative training. In each iteration, the generated process parameters are constrained and checked through relevant standard calculation formulas, so that the trained deep learning model can be used as a pressure vessel design model.

[0033] In one embodiment, the deep learning model includes a generative adversarial network, a convolutional neural network, and multi-objective optimization. A non-standardized container geometry that meets the design requirements is generated by a generative adversarial network, and material properties are predicted based on a process parameter database by a convolutional neural network. Multiple training data sets are simultaneously input into the initial deep learning model for optimization through multi-objective optimization to obtain the optimal deep learning model as a pressure vessel design model.

[0034] Specifically, when constructing the initial deep learning model, the initial deep learning model includes an input layer, a hidden layer, an output layer, and a loss function. The input layer uses the process parameters in the process parameter database as input parameters, the hidden layer uses a multi-layer neural network to extract features, and the output layer can output the process parameters of the target pressure vessel. The loss function can ensure that the output process parameters of the target pressure vessel comply with relevant standard calculation formulas, thereby realizing constraint verification of the generated process parameters.

[0035] In one embodiment, the product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the process of obtaining the module parameters of the target pressure vessel displayed on the simulation software is as follows: The product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the relevant standard calculation formulas are obtained according to the product performance requirements. The process parameters of the target pressure vessel are obtained through the relevant standard calculation formulas. The process parameters are input into the simulation software for simulation to obtain the target pressure vessel and its module parameters displayed on the simulation software. The module parameters are obtained through the simulation software, including cylinder module parameters, head module parameters, and flange module parameters.

[0036] Specifically, the module parameters refer to the parameters available on the simulation software. For example, the cylinder module parameters include the length and width of the cylinder. The user can obtain the corresponding cylinder module parameters by clicking on the cylinder on the simulation software. Since the parameters automatically generated by the model have strong randomness, the cylinder module parameters obtained by the pressure vessel design model may not be what the user really needs. Therefore, in this embodiment, a pressure feedback module is designed based on the pressure vessel design model. There can be multiple pressure feedback modules, each of which corresponds to a module parameter. The pressure feedback module refers to the cylinder simulation model, the head simulation model, the flange simulation model, etc. on the simulation software. The user can accurately obtain the parameters on the cylinder simulation model, the head simulation model, and the flange simulation model by clicking, and adjust the parameters. When the pressure feedback module receives and recognizes the user's adjustment of the module parameters, it will output it to the pressure vessel design module, so that the pressure vessel design model is re-trained. The pressure vessel design model plus the pressure vessel feedback module can achieve interactivity between the user and the pressure vessel design model, which is convenient for the user to accurately generate the required non-standardized pressure vessels and improve the user experience.

[0037] In summary, the present invention provides a design method and system for generating non-standard equipment for petrochemical industry based on deep learning, which changes the traditional design method relying on manual experience into a design method automatically generated by a deep learning model. In this design method, the process parameters and the relevant standard calculation formulas are first stored in a structured manner, and each process parameter is associated with each other through structured storage. The optimal design method of the target pressure vessel can be quickly determined through the association relationship. In addition, based on deep learning, a massive analysis of the process parameters in the pressure vessel production process is performed, which can quickly integrate the pressure vessel standards and real-time process parameters, greatly shorten the design cycle, and reduce dependence on manual labor. At the same time, it promotes the development of high-performance and high-safety non-standard equipment to meet the stringent requirements of petrochemical, aerospace and other fields.

[0038] Example 2 A design system for generating non-standard equipment for petrochemical industry based on deep learning, using a design method for generating non-standard equipment for petrochemical industry based on deep learning, comprising: Data acquisition module: obtain the process parameters of the pressure vessel production process and its related standard calculation formulas; Model building module: construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and use the relevant standard calculation formulas as constraints for iterative training to obtain the pressure vessel design model; Pressure vessel design module: input the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model, and obtain the module parameters of the target pressure vessel displayed on the simulation software; Pressure vessel feedback module: Receives and identifies the user's adjustment of module parameters on the simulation software, feeds back the content adjustment to the pressure vessel design model, updates the pressure vessel design model, and obtains new module parameters.

[0039] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. According to the technical essence of the present invention, within the spirit and principles of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.

Claims

1. A design method for generating non-standard equipment for petrochemical industry based on deep learning, characterized in that: The design method comprises the following steps: S1. Obtain the process parameters and related standard calculation formulas in the pressure vessel production process; S2. Perform structured storage according to the process parameters and their data types, and associate the corresponding process parameters through relevant standard calculation formulas to obtain a process parameter database; S3. Construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and perform iterative training using the relevant standard calculation formula as constraint conditions to obtain a pressure vessel design model; S4. Inputting the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model to obtain module parameters of the target pressure vessel displayed on the simulation software; S5. Receive and identify the content adjustment of the module parameters by the user on the simulation software, feed back the content adjustment to the pressure vessel design model, update the pressure vessel design model, and obtain new module parameters.

2. According to claim 1, a design method for generating non-standard equipment for petrochemical industry based on deep learning is characterized in that: In S1, the process of obtaining the process parameters in the pressure vessel production process and their related standard calculation formulas is specifically: obtaining the process table file recorded in the pressure vessel production process, performing content recognition processing on the process table file, extracting the process parameters, and obtaining the related standard calculation formulas according to the process rules corresponding to the process parameters.

3. A design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 2, characterized in that: The step S2 includes the following steps: S21. According to the data types of the process parameters in the pressure vessel production process, all process parameters are type-marked, and one data type corresponds to a number of process parameters; S22. Construct a process parameter database, use the data type as the library name of the sub-library in the process parameter database, store the corresponding process parameters in the sub-library, and store the process parameters of each process table file in isolation in the sub-library, number the process parameters of the same process table file, and associate the process parameters with the same number through relevant standard calculation formulas to obtain a process parameter database.

4. The design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 3 is characterized in that: The data types include container geometry, head geometry, interface design, and material type. The container geometry includes cylindrical, spherical, and special-shaped. The head geometry includes elliptical, spherical, and flat. The interface design includes inlet and outlet pipe diameters, flange size, and connection method. The material types include carbon steel, stainless steel, titanium alloy, and composite materials.

5. The design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 3 is characterized in that: In S22, the process of associating process parameters with the same number through relevant standard calculation formulas is specifically as follows: In the process parameter database, the sub-libraries are numbered in sequence, and the position information of each process parameter is obtained through the numbering of the sub-library and the process parameter. The position information refers to the coordinate position formed by the sub-library where the process parameter is located and the position information inside the sub-library. Then, the coordinate position of each process parameter is associated according to the relevant standard calculation formula to form a coordinate association table.

6. The design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 1 is characterized in that: In S3, an initial deep learning model is constructed. The process parameters in the process parameter database are input into the initial deep learning model as input parameters according to the association relationship, and the relevant standard calculation formulas are used as constraints for iterative training. The process of obtaining the pressure vessel design model is as follows: An initial deep learning model is constructed, and the corresponding process parameters are found according to the association relationship. Multiple training data sets are obtained using the process parameters and the corresponding historical design requirement data. The multiple training data sets are input into the initial deep learning model for iterative training. In each iteration, the generated process parameters are constrained and checked through relevant standard calculation formulas, so that the trained deep learning model can be used as a pressure vessel design model.

7. The design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 6 is characterized in that: The deep learning model includes a generative adversarial network, a convolutional neural network, and multi-objective optimization. A non-standardized container geometry that meets the design requirements is generated through a generative adversarial network, and material properties are predicted based on a process parameter database through a convolutional neural network. Multiple training data sets are simultaneously input into the initial deep learning model for optimization through multi-objective optimization to obtain the optimal deep learning model as a pressure vessel design model.

8. The design method for generating non-standard equipment for petrochemical industry based on deep learning according to claim 1 is characterized in that: In S4, the product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the process of obtaining the module parameters of the target pressure vessel displayed on the simulation software is as follows: The product performance requirements of the target pressure vessel are input as input parameters into the pressure vessel design model, and the relevant standard calculation formulas are obtained according to the product performance requirements. The process parameters of the target pressure vessel are obtained through the relevant standard calculation formulas. The process parameters are input into the simulation software for simulation to obtain the target pressure vessel and its module parameters displayed on the simulation software. The module parameters are obtained through the simulation software, including cylinder module parameters, head module parameters, and flange module parameters.

9. A design system for generating non-standard equipment for petrochemical industry based on deep learning, characterized in that: A design method for generating non-standard equipment for petrochemical industry based on deep learning as described in any one of claims 1 to 8 is applied, comprising: Data acquisition module: obtain the process parameters of the pressure vessel production process and its related standard calculation formulas; Process parameter database construction module: structured storage is performed according to the process parameters and their data types, and the corresponding process parameters are associated through relevant standard calculation formulas to obtain a process parameter database; Model building module: construct an initial deep learning model, input the process parameters in the process parameter database as input parameters into the initial deep learning model according to the association relationship, and use the relevant standard calculation formulas as constraints for iterative training to obtain the pressure vessel design model; Pressure vessel design module: input the product performance requirements of the target pressure vessel as input parameters into the pressure vessel design model, and obtain the module parameters of the target pressure vessel displayed on the simulation software; Pressure vessel feedback module: Receives and identifies the user's adjustment of module parameters on the simulation software, feeds back the content adjustment to the pressure vessel design model, updates the pressure vessel design model, and obtains new module parameters.

Images