Method and device for optimizing process parameters of heat treatment, electronic equipment and storage medium

Abstract

Embodiments of the present application relate to the technical field of manufacturing, and disclose a heat treatment process parameter optimization method and device, electronic equipment and a storage medium. The heat treatment process parameter optimization method comprises the following steps: obtaining a factor graph model of a workpiece, the factor graph model being trained based on historical heat treatment process parameters of the workpiece and a knowledge graph of the workpiece, and being used to represent influence parameters affecting the total length of the workpiece; inputting the influence parameters into a pre-trained graph attention model to obtain a predicted total length of the workpiece after heat treatment; the graph attention model is trained based on influence parameters of a plurality of workpieces, and is used to predict the total length of the workpiece after heat treatment; and based on the predicted total length and a pre-stored expected total length of the workpiece after heat treatment, the influence parameters are optimized, the process parameters affecting the product workpiece in the heat treatment process can be determined and parameter optimization is performed, so that the final total length of the workpiece is qualified.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of manufacturing, in particular to a heat treatment process parameter optimization method and device, electronic equipment and storage medium. BACKGROUND

[0002] The heat treatment process is a process of heating a workpiece to a certain temperature in a medium and keeping it for a certain time, and then cooling it at a certain speed, so as to change the organizational structure of the metal and thus change its performance, including physical, chemical and mechanical properties. For example, increasing or decreasing the hardness, strength, elasticity, toughness, plasticity, etc. of the metal material, the purpose is to eliminate defects in the blank, such as castings, forgings, so as to improve its process performance and prepare for subsequent machining.

[0003] Four fires, namely annealing, normalizing, quenching and tempering, are adopted to heat treat the metal. However, after the workpiece passes through the heat treatment process each time, the total length of the workpiece often fails to meet the requirements. The judgment of physical phenomena by artificial means cannot determine the parameters affecting the total length of the workpiece in the heat treatment process. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a heat treatment process parameter optimization method and device, electronic equipment and storage medium, which can determine the process parameters affecting the product workpiece in the heat treatment process and optimize the parameters, so that the final total length of the workpiece is qualified.

[0005] To solve the above technical problems, the embodiments of the present application provide a heat treatment process parameter optimization method, which comprises the following steps: obtaining a factor graph model of a workpiece, the factor graph model being trained based on historical heat treatment process parameters of the workpiece and a knowledge graph of the workpiece, and being used to represent influence parameters affecting the total length of the workpiece; inputting the influence parameters into a pre-trained graph attention model to obtain a predicted total length of the workpiece after the heat treatment process; the graph attention model is trained based on influence parameters of a plurality of workpieces, and is used to predict the total length of the workpiece after the heat treatment process; and optimizing the influence parameters based on the predicted total length and a pre-stored expected total length of the workpiece after the heat treatment process.

[0006] Embodiments of the present invention also provide an optimization device for heat treatment process parameters, comprising: an acquisition module for acquiring a factor graph model of a workpiece, the factor graph model being trained based on the historical heat treatment process parameters of the workpiece and the knowledge graph of the workpiece, used to characterize the influence parameters affecting the total length of the workpiece; an input module for inputting the influence parameters into a pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process; the graph attention model being trained based on the influence parameters of multiple workpieces to predict the total length of the workpiece after the heat treatment process; and an optimization module for optimizing the influence parameters based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process.

[0007] Embodiments of the present invention also provide an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described method for optimizing heat treatment process parameters.

[0008] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for optimizing heat treatment process parameters.

[0009] Compared to existing technologies, the embodiments of this invention train a factor graph model of the workpiece based on the workpiece's historical heat treatment process parameters and the workpiece's knowledge graph. Since the factor graph model represents the influencing parameters affecting the final total length of the workpiece during the heat treatment process, by obtaining the factor graph model of the workpiece, the influencing parameters affecting the total length of the workpiece are obtained. These influencing parameters are then input into a graph attention model. Since the graph attention model is pre-trained based on the influencing parameters of multiple workpieces and is used to predict the total length of the workpiece after the heat treatment process, the predicted total length of the workpiece after the heat treatment process can be obtained after inputting the influencing parameters into the pre-trained graph attention model. Based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process, the influencing parameters are optimized. Finally, the workpiece is heat-treated according to the optimized parameters to ensure that the final total length of the workpiece is qualified.

[0010] Furthermore, the influencing parameters are input into a pre-trained graph attention model to obtain optimized parameters. This process includes: inputting the influencing parameters into the pre-trained graph attention model to obtain feature vectors of the influencing parameters; obtaining the first predicted total length of the workpiece after heat treatment based on the feature vectors; and optimizing the influencing parameters based on the first predicted total length and the pre-stored total length of the workpiece after heat treatment to obtain optimized parameters. The influencing parameters are optimized by comparing the first predicted total length of the workpiece after heat treatment with the actual total length of the workpiece after heat treatment.

[0011] In addition, the factor graph model is also used to characterize the influence weights between different influencing parameters. The influencing parameters are input into a pre-trained graph attention model to obtain the predicted total length of the workpiece after heat treatment. This includes inputting the influencing parameters and their influence weights into the graph attention model to obtain the predicted total length of the workpiece after heat treatment. The predicted total length of the workpiece after heat treatment can be obtained when the factor graph model is also used to characterize the influence weights between different influencing parameters.

[0012] In addition, the factor graph model is also used to characterize the influence weights between different influencing parameters; the influencing parameters and the interactions between different influencing parameters are input into the graph attention model to obtain the predicted total length of the workpiece after heat treatment. The predicted total length of the workpiece after heat treatment can be obtained when the factor graph model is also used to characterize the interactions between different influencing parameters.

[0013] In addition, the factor graph model is a directed acyclic graph model, specifically including: influencing parameters and variables related to the influencing parameters; wherein, the influencing parameters and variables are connected by directed edges and are used as input to the graph attention model.

[0014] In addition, based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process, the influencing parameters are optimized, including: determining the relationship between the influencing parameters and the total length of the workpiece after the heat treatment process based on the predicted total length and the expected total length; optimizing the influencing parameters based on the relationship between the influencing parameters and the total length of the workpiece after the heat treatment process, so as to obtain the final optimized influencing parameters.

[0015] In addition, after optimizing the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment, the process further includes: adjusting the parameters of the workpiece before quenching according to the optimized influencing parameters to obtain target parameters, so as to perform heat treatment on the workpiece using the target parameters, so that the total length of the workpiece is qualified after tempering. Attached Figure Description

[0016] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0017] Figure 1 This is a detailed flowchart of workpiece manufacturing according to an embodiment of the present invention;

[0018] Figure 2 This is a flowchart of a method for optimizing heat treatment process parameters according to an embodiment of the present invention. Figure 1 ;

[0019] Figure 3 This is a structural diagram of a factor graph model provided according to an embodiment of the present invention;

[0020] Figure 4 This is a structural diagram of a graph attention model provided according to an embodiment of the present invention;

[0021] Figure 5 This is a flow chart of a method for optimizing heat treatment process parameters according to another embodiment of the present invention. Figure 2 ;

[0022] Figure 6 This is a flow chart of a method for optimizing heat treatment process parameters according to another embodiment of the present invention. Figure 3 ;

[0023] Figure 7 This is a schematic diagram of an apparatus for optimizing heat treatment process parameters according to another embodiment of the present invention;

[0024] Figure 8 This is a structural diagram of an electronic device provided according to another embodiment of the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are presented in the various embodiments of the present invention to facilitate a better understanding of this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments. The division of the various embodiments below is for ease of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined with and referenced by each other without contradiction.

[0026] One embodiment of the present invention relates to a method for optimizing heat treatment process parameters, applied in the heat treatment process of a workpiece. It is understood that the heat treatment process is only one step in the workpiece manufacturing process. In one example, the specific manufacturing process of the workpiece is as follows: Figure 1 As shown, it includes: rough machining, surface heat treatment, heat treatment process, and finish machining, wherein the heat treatment process includes: annealing, normalizing, quenching, and tempering.

[0027] To avoid the problem of the final total length of the workpiece being unqualified during the heat treatment process, this embodiment of the invention first trains a factor graph model of the workpiece based on the historical heat treatment process parameters of the workpiece and the knowledge graph of the workpiece to characterize the parameters affecting the total length of the workpiece. Based on the factor graph models of multiple workpieces, a graph attention model for predicting the total length of the workpiece after the heat treatment process is pre-trained. The parameters affecting the total length of the workpiece are input into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process. Based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process, the parameters can be optimized. Then, the workpiece is heat-treated according to the optimized parameters to ensure that the final total length of the workpiece is qualified.

[0028] The following provides a detailed explanation of the implementation details of the optimization method for the heat treatment process parameters in this embodiment. The following details are provided for ease of understanding and are not essential for implementing this solution. The specific flowchart of the optimization method for the heat treatment process parameters in this embodiment is shown below. Figure 2 As shown, it includes:

[0029] Step 201: Obtain the factor graph model of the workpiece.

[0030] Specifically, based on the workpiece's historical heat treatment process parameters and its knowledge graph, a machine learning algorithm is used to train a factor graph model of the workpiece. The historical heat treatment process parameters mainly include parameters of the workpiece before quenching, as well as parameters that, based on experience, will affect the overall length of the workpiece during quenching, such as roundness 1 and roundness 2. The workpiece's knowledge graph represents the process flow steps. The factor graph model is used to characterize the parameters that influence the overall length of the workpiece during the quenching process of the heat treatment process.

[0031] In one example, the factor graph model is a type of directed acyclic graph model, such as... Figure 3 As shown, this includes variable nodes and factor nodes. The factor nodes are the parameters that affect the total length of the workpiece in this embodiment. Variable nodes and factor nodes are connected by directed edges. A variable node connected to a factor node is a variable of that factor, and the variable can be roundness.

[0032] Step 202: Input the influencing parameters into the pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0033] Specifically, after determining the influencing parameters of the total length of the workpiece through the factor graph model of the workpiece, the influencing parameters are input into the pre-trained graph attention model to obtain the feature vector of the influencing parameters. The feature vector is used to better describe the characteristics of the influencing parameters. Based on the feature vector, a regression algorithm is used to obtain the predicted total length of the workpiece after the heat treatment process.

[0034] The structure diagram of the graph attention model is as follows: Figure 4 As shown, the input node is the factor, which is the parameter that affects the total length of the workpiece, and the output node is the predicted total length of the workpiece after the heat treatment process.

[0035] In its implementation, the graph attention model is trained using machine learning algorithms based on the influence parameters of multiple workpieces and is used to predict the total length of the workpiece after heat treatment.

[0036] Step 203: Optimize the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process.

[0037] Specifically, based on the obtained predicted total length, the expected total length of the workpiece after heat treatment is pre-stored, along with the relationship between influencing parameters and the total length of the workpiece after heat treatment. The influencing parameters are then optimized. This relationship can be a function, allowing for automated configuration of the influencing parameters to obtain optimized parameters. For example, if a larger influencing parameter value results in a larger total length of the workpiece after heat treatment, then the influencing parameter is increased to determine its value when the total length of the workpiece approaches the expected total length, thus completing the optimization process. Conversely, if a larger influencing parameter value results in a smaller total length of the workpiece after heat treatment, then the influencing parameter is decreased to determine its value when the total length of the workpiece approaches the expected total length, thus completing the optimization process.

[0038] In one example, some parameters of the workpiece may change during the quenching process of the heat treatment process, resulting in the overall length of the product being unqualified after tempering. Therefore, after obtaining the optimized influence parameters, the parameters of the workpiece before quenching are adjusted according to the optimized influence parameters to obtain the target parameters. The heat treatment process is then carried out on the workpiece using the target parameters so that the overall length of the workpiece is qualified after tempering.

[0039] In this embodiment, a factor graph model of the workpiece is trained based on the workpiece's historical heat treatment process parameters and the workpiece's knowledge graph. Since the factor graph model represents the influence parameters affecting the final total length of the workpiece during the heat treatment process, the influence parameters affecting the total length of the workpiece are obtained by acquiring the factor graph model of the workpiece. These influence parameters are then input into the graph attention model. Since the graph attention model is pre-trained based on the influence parameters of multiple workpieces and is used to predict the total length of the workpiece after the heat treatment process, the predicted total length of the workpiece after the heat treatment process can be obtained after inputting the influence parameters into the pre-trained graph attention model. Based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process, the influence parameters are optimized. Finally, the workpiece is heat-treated according to the optimized parameters to ensure that the final total length of the workpiece is qualified.

[0040] In another embodiment, the factor graph model is used not only to characterize the influence parameters of the total workpiece length, but also to characterize the influence weights of those parameters. See also Figure 3 For example, regarding variable 1, factor 1, i.e., the influence weight of parameter 1 is 0.2, the influence weight of factor 2 is 0.3, and the influence weight of factor 3 is 0.5.

[0041] In one example, the method for optimizing heat treatment process parameters according to an embodiment of this application can be implemented through the following steps, the specific process of which is as follows: Figure 5 As shown:

[0042] Step 501: Obtain the factor graph model of the workpiece.

[0043] Step 502: Input the influencing parameters and their influence weights into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0044] Step 503: Optimize the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process.

[0045] In this embodiment, a factor graph model of the workpiece is trained based on the workpiece's historical heat treatment process parameters and the workpiece's knowledge graph. Since the factor graph model represents the influencing parameters and their weights in the heat treatment process of the workpiece, by obtaining the factor graph model, the influencing parameters and their weights are obtained. These parameters and their weights are then input into a pre-trained graph attention model. Since the graph attention model is pre-trained based on the influencing parameters of multiple workpieces and is used to predict the total length of the workpiece after heat treatment, the predicted total length of the workpiece after heat treatment can be obtained after inputting the influencing parameters into the pre-trained graph attention model. Based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment, the influencing parameters are optimized. Finally, the workpiece is heat-treated according to the optimized parameters to ensure that the final total length of the workpiece is qualified.

[0046] In another embodiment, the factor graph model, in addition to characterizing the parameters affecting the total length of the workpiece, is also used to characterize the interactions between different parameters, see [link to relevant documentation]. Figure 3 For example, Factor 1 and Factor 2 influence each other.

[0047] In one example, the optimized parameters can be obtained through the following steps, the specific process of which is as follows: Figure 6 As shown:

[0048] Step 601: Obtain the factor graph model of the workpiece.

[0049] Step 602: Input the influencing parameters and the mutual influence between different influencing parameters into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0050] Among them, the mutual influence between different influencing parameters, i.e. Figure 3 The interaction between factors, for example, if the value of factor 1 increases, the value of factor 2 will decrease, resulting in a change in the total length of the workpiece after the heat treatment process. Therefore, it is necessary to input the influencing parameters and the interaction between different influencing parameters into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process under this condition.

[0051] In one example, the factor graph model simultaneously represents the influencing parameters affecting the total length of the workpiece, the influence weights of the influencing parameters, and the mutual influences between different influencing parameters. Therefore, the influencing parameters, the influence weights of the influencing parameters, and the mutual influences between different influencing parameters can all be input into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0052] Step 603: Optimize the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process.

[0053] In this embodiment, a factor graph model of the workpiece is trained based on the workpiece's historical heat treatment process parameters and the workpiece's knowledge graph. Since the factor graph model represents the influencing parameters affecting the final total length of the workpiece during the heat treatment process, the influence weights of the influencing parameters, and the mutual influences between different influencing parameters, by obtaining the factor graph model of the workpiece, the influencing parameters affecting the total length of the workpiece, the influence weights of the influencing parameters, and the mutual influences between different influencing parameters are obtained. The influencing parameters, the influence weights of the influencing parameters, and the mutual influences between different influencing parameters are then input into a pre-trained graph attention model. Since the graph attention model is pre-trained based on the influencing parameters of multiple workpieces and is used to predict the total length of the workpiece after the heat treatment process, the predicted total length of the workpiece after the heat treatment process can be obtained after inputting the influencing parameters into the pre-trained graph attention model. Based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process, the influencing parameters are optimized. Finally, the workpiece is heat-treated according to the optimized parameters so that the final total length of the workpiece is qualified.

[0054] It should be noted that the examples described above in this embodiment are merely illustrative for ease of understanding and do not constitute a limitation on the technical solution of the present invention.

[0055] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this patent.

[0056] Another embodiment of the present invention relates to a device for optimizing heat treatment process parameters. The details of this device are described below for ease of understanding and are not essential for implementing this example. Figure 7 This is a schematic diagram of the heat treatment process parameter optimization device described in this embodiment, including: acquisition module 701, input module 702 and optimization module 703.

[0057] Specifically, the acquisition module 701 is used to acquire the factor graph model of the workpiece. The factor graph model is trained based on the historical heat treatment process parameters of the workpiece and the knowledge graph of the workpiece, and is used to characterize the influence parameters that affect the total length of the workpiece.

[0058] The input module 702 is used to input the influence parameters into the pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process; wherein, the graph attention model is trained based on the influence parameters of multiple workpieces and is used to predict the total length of the workpiece after the heat treatment process.

[0059] In one example, while the factor graph model is also used to characterize the influence weights of the influence parameters, the input module 702 is also used to input the influence parameters and their influence weights into a pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0060] In one example, while the factor graph model is also used to characterize the influence weights between different influence parameters, the input module 702 is also used to input the influence parameters and the mutual influences between different influence parameters into the pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

[0061] The optimization module 703 is used to optimize the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after the heat treatment process.

[0062] In one example, the optimization module 703 is further configured to determine the relationship between the influencing parameters and the total length of the workpiece after the heat treatment process based on the predicted total length and the expected total length; and to optimize the influencing parameters based on the relationship between the influencing parameters and the total length of the workpiece after the heat treatment process.

[0063] It is not difficult to see that this embodiment is a device embodiment corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the first embodiment.

[0064] It is worth mentioning that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by this invention; however, this does not mean that other units are absent from this embodiment.

[0065] Another embodiment of the present invention relates to an electronic device, such as... Figure 8As shown, it includes: at least one processor 801; and a memory 802 communicatively connected to the at least one processor 801; wherein the memory 802 stores instructions executable by the at least one processor 801, the instructions being executed by the at least one processor 801 to enable the at least one processor 801 to execute the heat treatment process parameter optimization method in the above embodiments.

[0066] The memory and processor are connected via a bus, which can include any number of interconnecting buses and bridges, connecting various circuits of one or more processors and memories. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and will not be described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over the wireless medium via an antenna, which further receives data and transmits it to the processor.

[0067] The processor manages the bus and general processing, and also provides various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory is used to store data used by the processor during operation.

[0068] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the method embodiments described above.

[0069] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0070] Those skilled in the art will understand that the above embodiments are specific examples of implementing the present invention, and in practical applications, various changes in form and detail may be made without departing from the spirit and scope of the present invention.

Claims

1. A method for optimizing heat treatment process parameters, characterized in that, include: A factor graph model of the workpiece is obtained. The factor graph model is trained based on the historical heat treatment process parameters of the workpiece and the knowledge graph of the workpiece, and is used to characterize the influence parameters affecting the total length of the workpiece. The influencing parameters are input into a pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process. The graph attention model is trained based on the influence parameters of multiple workpieces and is used to predict the total length of the workpiece after heat treatment. Based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment, the influencing parameters are optimized. The factor graph model is a directed acyclic graph model, specifically including: the influencing parameters and the variables related to the influencing parameters; wherein the influencing parameters and the variables are connected by directed edges.

2. The method for optimizing heat treatment process parameters according to claim 1, characterized in that, The factor graph model is also used to characterize the influence weights of the influence parameters; The step of inputting the influencing parameters into a pre-trained graph attention model to obtain the predicted total length of the workpiece after heat treatment includes: The influence parameters and their influence weights are input into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

3. The method for optimizing heat treatment process parameters according to claim 1, characterized in that, The factor graph model is also used to characterize the mutual influence between different influencing parameters; The step of inputting the influencing parameters into a pre-trained graph attention model to obtain the predicted total length of the workpiece after heat treatment includes: The influence parameters and the interactions between the different influence parameters are input into the graph attention model to obtain the predicted total length of the workpiece after the heat treatment process.

4. The method for optimizing heat treatment process parameters according to any one of claims 1 to 3, characterized in that, The graph attention model is trained based on a regression algorithm.

5. The method for optimizing heat treatment process parameters according to claim 1, characterized in that, The optimization of the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment includes: Based on the predicted total length, the expected total length, and the relationship between the influencing parameters and the total length of the workpiece after heat treatment, the influencing parameters are optimized.

6. The method for optimizing heat treatment process parameters according to claim 5, characterized in that, After optimizing the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment, the method further includes: The parameters of the workpiece before quenching are adjusted according to the optimized influence parameters to obtain the target parameters, and the workpiece is then subjected to heat treatment process using the target parameters.

7. An apparatus for optimizing heat treatment process parameters, characterized in that, include: The acquisition module is used to acquire the factor graph model of the workpiece. The factor graph model is trained based on the historical heat treatment process parameters of the workpiece and the knowledge graph of the workpiece, and is used to characterize the influence parameters affecting the total length of the workpiece. The factor graph model is a directed acyclic graph model, specifically including: the influence parameters and variables related to the influence parameters; wherein the influence parameters and the variables are connected by directed edges. An input module is used to input the influence parameters into a pre-trained graph attention model to obtain the predicted total length of the workpiece after the heat treatment process; the graph attention model is trained based on the influence parameters of multiple workpieces and is used to predict the total length of the workpiece after the heat treatment process. An optimization module is used to optimize the influencing parameters based on the predicted total length and the pre-stored expected total length of the workpiece after heat treatment.

8. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the optimization method for heat treatment process parameters as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for optimizing the heat treatment process parameters according to any one of claims 1 to 6.

Images