Regulation and control method and system for quenching treatment of metal product

Abstract

The invention belongs to the technical field of factory control. The invention provides a regulation and control method and system for quenching treatment of a metal product. The method comprises the following steps: collecting training data, and training a constructed first depth prediction model and a constructed second depth prediction model by using the training data; determining a required target hardness after quenching treatment of the metal part, and performing deep analysis on the target hardness by using a first depth prediction model and a second depth prediction model to obtain a target heat preservation time length, a cooling medium type and a cooling medium temperature; and heat preservation operation is conducted on the metal part which is heated to form austenite according to the target heat preservation time, a target cooling medium is selected according to the type of the cooling medium and the temperature of the cooling medium, and the target cooling medium is used for conducting rapid cooling treatment on the austenite metal part. According to the method, the optimal heat preservation parameters and rapid cooling parameters can be determined according to the target hardness of the quenched metal part by a customer, so that the individual requirements of the customer can be met.

Description

Technical Field

[0001] The present invention relates to the field of factory control technology, and in particular to a control method and system for quenching metal products. Background Art

[0002] Quenching is a key step in the metal heat treatment process, mainly used to improve the hardness and strength of the metal. Quenching includes three main process steps: heating, heat preservation, and rapid cooling. The specific introduction is as follows: Heating: The metal workpiece is heated to a specific temperature, which is called the austenitizing temperature, which is usually higher than the critical temperature of the material. At this temperature, the microstructure inside the metal will change and form austenite.

[0003] Holding: After reaching the austenitizing temperature, the workpiece needs to be held for a period of time to ensure that the entire workpiece is heated evenly and the microstructure is fully transformed to austenite.

[0004] Rapid cooling: After holding, the workpiece needs to be cooled quickly to form martensite or other hardening phases. The cooling rate must be fast enough to avoid the formation of undesirable microstructures such as pearlite or bainite. The cooling medium can be water, oil, air or other special quenching liquids.

[0005] It can be seen that heat preservation and rapid cooling are the key processes of quenching, which directly affect the quality of metal parts after quenching, such as whether there are cracks on the surface of the metal parts, and whether there are undesirable microstructures such as pearlite or bainite inside the metal parts.

[0006] The existing technology currently has a technical problem, that is, it is not possible to specifically adjust the heat preservation parameters and rapid cooling parameters in the quenching process according to the hardness requirements of the metal parts after quenching treatment required by the customer, resulting in difficulty in meeting the customer's personalized needs for quenching treatment of metal parts. The present invention aims to solve this technical problem. Summary of the invention

[0007] In order to solve the above-mentioned technical problems, the present invention provides a control method, system, electronic equipment, computer storage medium and computer program product for quenching metal products.

[0008] The present invention discloses a control method for quenching a metal product, the method comprising the following steps: collecting training data, and using the training data to train a first deep prediction model and a second deep prediction model; determining a target hardness of a metal part after quenching, and using the first deep prediction model and the second deep prediction model to perform a deep analysis on the target hardness to obtain a target insulation time, a cooling medium type, and a cooling medium temperature; performing insulation operation on a metal part heated to form austenite according to the target insulation time, and selecting a target cooling medium according to the cooling medium type and the cooling medium temperature, and using the target cooling medium to perform rapid cooling treatment on the austenite metal part.

[0009] Optionally, collecting training data and using the training data to train the constructed first deep prediction model and the second deep prediction model include: collecting a group of historical quenching data, and separating from the group of historical quenching data first historical quenching data corresponding to the insulation operation and second historical quenching data corresponding to the rapid cooling operation; constructing a first training data set and a first test data set, a second training data set and a second test data set according to the first historical quenching data and the second historical quenching data; wherein the first training data set and the first test data set, the second training data set and the second test data set all include label one and label two, the label one is the hardness of the metal part after quenching, and the label two is the quality parameter of the metal part after quenching; using the first training data set and the first test data set to train and test the first deep prediction model, and using the second training data set and the second test data set to train and test the second deep prediction model, until the first deep prediction model and the second deep prediction model are trained to meet the standards.

[0010] Optionally, after the first depth prediction model and the second depth prediction model are trained to meet the standards, the method also includes: using the first test data set to test the first depth prediction model that has been trained to meet the standards to obtain a set of predicted insulation times; screening out second training data that belong to the same historical quenching data as each first test data from the second training data set, using each screened second training data as the data body of the third training data, using the corresponding predicted insulation time as the label three of the third training data, and using the original label of each second training data, that is, the hardness and quality parameters of the metal part after quenching, as the label one and label two of the third training data, respectively, and each third training data constitutes a third training data set; using the third training data set to perform secondary training on the second depth prediction model that has been trained to meet the standards, and using the second test data set to test the second depth prediction model after the secondary training, until the second depth prediction model is trained to meet the standards, that is, the final first depth prediction model and second depth prediction model are obtained.

[0011] Optionally, the first deep prediction model that has been trained to meet the standards is tested using the first test data set to obtain a set of predicted insulation times, including: determining the number of types of metal plates in the first test data set, determining a screening number based on the number of types, randomly screening some first test data from the first test data set according to the screening number, and the screened first test data constitute a third test data set; the first deep prediction model that has been trained to meet the standards is tested using the third test data set to obtain a set of predicted insulation times.

[0012] Optionally, the screening quantity is calculated according to the number of types and the following formula, including: ; In the formula, is the screening quantity; is the total amount of first test data in the first test data set; is the number of roughly classified types of metal plates in the first test data set, is the number of subdivision types of metal plates in the first test data set, is the total number of subdivision types of known metal plates; a is a constant value.

[0013] Optionally, the first depth prediction model is constructed based on CNN, and the second depth prediction model is constructed based on Transformer.

[0014] The present invention also discloses a control system for quenching metal products, the system comprising a processing device and a storage device, the computer code stored in the storage device is called and executed by the processing device to implement the following steps: The present invention also discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, the processor executing the computer program to implement any of the methods described above.

[0015] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0016] The present invention also discloses a computer program product, which includes computer codes. When the computer codes are executed by a processor of an electronic device, any of the above methods is implemented.

[0017] The beneficial effect of the present invention is at least that: the scheme of the present invention can determine the optimal insulation parameters and rapid cooling parameters according to the customer's personalized demand for the target hardness of the metal parts after quenching, and quench the metal parts separately according to the optimal insulation parameters and rapid cooling parameters, thereby meeting the personalized needs of the customer and helping to improve the performance of the quenching company. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flow chart of a control method for quenching a metal product disclosed in an embodiment of the present invention.

[0020] Figure 2 It is a flow chart of a method for generating a predicted heat preservation time disclosed in an embodiment of the present invention.

[0021] Figure 3 It is a structural schematic diagram of a control system for quenching a metal product disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following is a description of the implementation of the present application by specific specific embodiments. People familiar with the technology can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0023] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0024] like Figure 1 As shown, an embodiment of the present invention discloses a control method for quenching a metal product, the method comprising the following steps: collecting training data, and using the training data to train a first deep prediction model and a second deep prediction model; determining a target hardness of the metal part after quenching, and using the first deep prediction model and the second deep prediction model to perform a deep analysis on the target hardness to obtain a target insulation time, a cooling medium type, and a cooling medium temperature; performing insulation operation on the metal part heated to form austenite according to the target insulation time, and selecting a target cooling medium according to the cooling medium type and the cooling medium temperature, and using the target cooling medium to quickly cool the austenite metal part.

[0025] The above scheme of the present invention supports the selection of the best insulation parameters and rapid cooling parameters according to the target hardness of the quenched metal parts required by the customer. Specifically, the pre-constructed first depth prediction model and the second depth prediction model are trained using training data. Then, the target hardness of the metal part after quenching treatment is received according to the customer's needs, and the first depth prediction model is used to perform a deep analysis on the target hardness to obtain the best insulation parameters, that is, the target insulation time, and the second depth prediction model is used to perform a deep analysis on the target hardness to obtain the best rapid cooling parameters, that is, the cooling medium type and the cooling medium temperature. Finally, the insulation operation of the metal part heated to form austenite is controlled according to the target insulation time, and the target cooling medium is selected according to the cooling medium type and the cooling medium temperature, and the austenite metal part is rapidly cooled using the target cooling medium, so that a metal part that meets the target hardness and meets the quality standards can be obtained. Among them, meeting the quality standards means that there are no cracks on the surface of the metal part and there are no undesirable microstructures such as pearlite or bainite inside the metal part.

[0026] Therefore, the solution of the present invention can determine the optimal insulation parameters and rapid cooling parameters according to the customer's personalized demand for the target hardness of the metal parts after quenching, and quench the metal parts separately according to the optimal insulation parameters and rapid cooling parameters, thereby meeting the customer's personalized needs and helping to improve the performance of the quenching company.

[0027] Optionally, collecting training data and using the training data to train the constructed first deep prediction model and the second deep prediction model include: collecting a group of historical quenching data, and separating from the group of historical quenching data first historical quenching data corresponding to the insulation operation and second historical quenching data corresponding to the rapid cooling operation; constructing a first training data set and a first test data set, a second training data set and a second test data set according to the first historical quenching data and the second historical quenching data; wherein the first training data set and the first test data set, the second training data set and the second test data set all include label one and label two, the label one is the hardness of the metal part after quenching, and the label two is the quality parameter of the metal part after quenching; using the first training data set and the first test data set to train and test the first deep prediction model, and using the second training data set and the second test data set to train and test the second deep prediction model, until the first deep prediction model and the second deep prediction model are trained to meet the standards.

[0028] In this embodiment, the present invention collects a set of historical quenching data recorded during the actual quenching operation, each piece of historical quenching data includes quenching data in the heat preservation operation (i.e., heat preservation time) and quenching data in the rapid cooling operation (i.e., cooling medium type, cooling medium temperature) and the corresponding hardness and quality parameters of the metal part after quenching (whether there are cracks on the surface of the metal part, whether there are undesirable microstructures such as pearlite or bainite inside the metal part), and the historical quenching data are separated one by one according to the heat preservation operation and the rapid cooling operation, that is, the first historical quenching data and the second historical quenching data mentioned above are obtained. The first historical quenching data and the second historical quenching data are used as the training data body, and the hardness and quality parameters of the metal part after quenching are used as label one and label two respectively, that is, two sets of training data sets are formed, and the two sets of training data sets are divided into the first training data set and the first test data set, and the second training data set and the second test data set respectively.

[0029] Then, the first training data set and the first test data set constructed above are used to train and test the first depth prediction model, and the second training data set and the second test data set are used to train and test the second depth prediction model, until both the first depth prediction model and the second depth prediction model are trained to meet the standards, that is, the convergence conditions are met.

[0030] Optionally, after the first depth prediction model and the second depth prediction model are trained to meet the standards, the method also includes: using the first test data set to test the first depth prediction model that has been trained to meet the standards to obtain a set of predicted insulation times; screening out second training data that belong to the same historical quenching data as each first test data from the second training data set, using each screened second training data as the data body of the third training data, using the corresponding predicted insulation time as the label three of the third training data, and using the original label of each second training data, that is, the hardness and quality parameters of the metal part after quenching, as the label one and label two of the third training data, respectively, and each third training data constitutes a third training data set; using the third training data set to perform secondary training on the second depth prediction model that has been trained to meet the standards, and using the second test data set to test the second depth prediction model after the secondary training, until the second depth prediction model is trained to meet the standards, that is, the final first depth prediction model and second depth prediction model are obtained.

[0031] In this embodiment, the first training data set and the first test data set, the second training data set and the second test data set constructed above correspond to the heat preservation operation and the rapid cooling operation, respectively. For the second deep prediction model, the austenite transformation of the metal parts in the previous process, i.e., the heat preservation operation (i.e., whether it is completely converted to austenite) is not considered during training. Therefore, the aforementioned one-time training of the second deep prediction model is essentially insufficient and a second training is required.

[0032] In response to the above technical problems, the present invention further uses a first test data set (including multiple first test data) to test the first deep prediction model that has been trained to meet the standards, and then collects a set of predicted insulation time predicted by the first deep prediction model; and selects the second training data that belongs to the same historical quenching data as each first test data from the second training data set (including multiple second training data), and uses each second training data obtained by screening as the data body of the third training data, and uses the corresponding predicted insulation time as the label three of the third training data, and uses the original label of each second training data, that is, the hardness and quality parameters of the metal parts after quenching, as the label one and label two of the third training data, respectively, to obtain a new third training data set. It can be seen that the new third training data set contains the label of the predicted insulation time output by the first deep prediction model. Using it to perform a second training on the second deep prediction model that has been trained to meet the standards once, the second deep prediction model can gradually master the optimal insulation parameters predicted by the first deep prediction model, and analyze and obtain the optimal rapid cooling parameters. At the same time, the second depth prediction model that has undergone the second training is trained and tested using the original second test data set. If the test result meets the standard, it is used as the final second depth prediction model, and the first depth prediction model that meets the standard after the first training is the final first depth prediction model.

[0033] Therefore, the solution of the present invention uses training data corresponding to the heat preservation operation and training data corresponding to the rapid cooling operation to independently train the first depth prediction model and the second depth prediction model, and then uses the test output data of the first depth prediction model to fuse new training data to perform another fusion training on the second depth prediction model, so that the second depth prediction model can be coordinated with the first depth prediction model, so that the heat preservation parameters and rapid cooling parameters predicted by the first depth prediction model and the second depth prediction model respectively obtained are matched, ensuring that the actual hardness of the metal plate obtained after quenching treatment is closest to the target hardness required by the customer, and the quality parameters of the metal plate are still up to standard.

[0034] Alternatively, if Figure 2 As shown, the first deep prediction model that has been trained to meet the standards is tested using the first test data set to obtain a set of predicted insulation times, including: determining the number of types of metal plates in the first test data set, determining a screening number based on the number of types, randomly screening some first test data from the first test data set according to the screening number, and the screened first test data constitute a third test data set; the first deep prediction model that has been trained to meet the standards is tested using the third test data set to obtain a set of predicted insulation times.

[0035] In this embodiment, metal plates are divided into multiple types, such as carbon steel, low alloy steel (such as 45#, T7, T8, etc.), alloy structural steel (such as 40Cr, 40MnB, 35CrMo, etc.), high alloy tool steel (such as 9SiCr, CrWMn, Crl2MoV, W6, W8, etc.), and different types of metal plates have different requirements for insulation parameters and rapid cooling parameters. Correspondingly, the historical quenching data collected above will also include historical quenching data of different types of metal plates, and then the first test data set also corresponds to the historical quenching data of multiple different types of metal plates.

[0036] Based on the above situation, the present invention first determines the number of types of metal plates involved in the first test data set, determines the number of screenings based on the number of types, randomly screens part of the first test data from the first test data set according to the screening number, and the screened first test data constitute the third test data set. Then, the third test data set is used to test the first deep prediction model that has been trained to the standard, and a set of predicted insulation time is obtained.

[0037] Among them, the screening number and the number of types of metal plates are negatively correlated, that is, the more types of metal plates there are, the greater the probability that the first deep prediction model has been fully trained in one training process, and the higher the confidence of the predicted insulation time output by the first deep prediction model. At this time, a smaller number of first test data are screened from the first test data set to form the third test data set; the fewer types of metal plates there are, the smaller the probability that the first deep prediction model has been fully trained in one training process, and the lower the confidence of the predicted insulation time output by the first deep prediction model. At this time, a larger number of first test data are screened from the first test data set to form the third test data set to ensure the effect of the secondary training of the second deep prediction model.

[0038] Optionally, the screening quantity is calculated according to the number of types and the following formula, including: ; In the formula, is the screening quantity; is the total amount of first test data in the first test data set; is the number of roughly classified types of metal plates in the first test data set, is the number of subdivision types of metal plates in the first test data set, is the total number of subdivision types of known metal plates; a is a constant value.

[0039] In this embodiment, the present invention roughly divides and subdivides the types of metal plates. The roughly divided types refer to dividing metal plates into carbon steel, low alloy steel, alloy structural steel, high alloy tool steel, etc., while the subdivided types refer to dividing metal plates into carbon steel, low alloy steel (45#), low alloy steel (T7), low alloy steel (T8), alloy structural steel (40Cr), alloy structural steel (40MnB), alloy structural steel (35CrMo), high alloy tool steel (9SiCr), high alloy tool steel (CrWMn), high alloy tool steel (Crl2MoV), high alloy tool steel (W6), high alloy tool steel (W8), etc. Then, the number of roughly divided types of metal plates in the first test data set can be calculated. , the number of subdivision types of metal plates in the first test dataset ; and, the total number of known metal sheet subdivision types Substituting the above parameters into the above calculation formula, the screening quantity can be calculated, and a part of the first test data is screened from the first test data set according to the screening quantity to form a third test data set.

[0040] It should be noted that the constant value a in the above formula is used to ensure A value less than 1.

[0041] Optionally, the first depth prediction model is constructed based on CNN, and the second depth prediction model is constructed based on Transformer.

[0042] In this embodiment, both the first deep prediction model and the second deep prediction model can be constructed based on CNN. However, since the first deep prediction model does not need to consider the rapid cooling parameters of the subsequent quenching process when predicting the optimal insulation parameters, the prediction difficulty of the first deep prediction model is relatively low, so it can be constructed using a conventional CNN (convolutional neural network). In contrast, the second deep prediction model also needs to consider the insulation parameters predicted by the first deep prediction model of the previous quenching process when predicting the optimal rapid cooling parameters, and its prediction difficulty is greater, so it is preferred to set the second deep prediction model to be constructed based on Transformer.

[0043] like Figure 3As shown, an embodiment of the present invention also discloses a control system for quenching metal products, the system comprising a processing device and a storage device, the computer code stored in the storage device is called and executed by the processing device to implement the following steps: collecting training data, and using the training data to train the constructed first deep prediction model and the second deep prediction model; determining the required target hardness of the metal part after quenching treatment, using the first deep prediction model and the second deep prediction model to perform a deep analysis on the target hardness to obtain the target insulation time, the cooling medium type and the cooling medium temperature; performing insulation operation on the metal part heated to form austenite according to the target insulation time, and selecting the target cooling medium according to the cooling medium type and the cooling medium temperature, and using the target cooling medium to perform rapid cooling treatment on the austenitic metal part.

[0044] An embodiment of the present invention further discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method described in the above embodiment.

[0045] An embodiment of the present invention further discloses a computer storage medium, wherein the computer storage medium stores a computer program, and the computer program is executed by a processor to implement the method described in the above embodiment.

[0046] An embodiment of the present invention further discloses a computer program product, which includes computer code. When the computer code is executed by a processor of an electronic device, the method described in the above embodiment is implemented.

[0047] The computer-readable storage medium described above may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media may include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0048] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0049] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0050] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling quenching of a metal product, characterized in that: The method includes the following steps: collecting training data, and using the training data to train a first deep prediction model and a second deep prediction model; determining a target hardness of a metal part after quenching, and using the first deep prediction model and the second deep prediction model to perform a deep analysis on the target hardness to obtain a target insulation time, a cooling medium type, and a cooling medium temperature; performing insulation operations on a metal part heated to form austenite according to the target insulation time, and selecting a target cooling medium according to the cooling medium type and the cooling medium temperature, and using the target cooling medium to perform rapid cooling on the austenite metal part.

2. A control method for quenching a metal product according to claim 1, characterized in that: The collecting of training data and using the training data to train the constructed first deep prediction model and the second deep prediction model include: collecting a group of historical quenching data, and separating from the group of historical quenching data the first historical quenching data corresponding to the heat preservation operation and the second historical quenching data corresponding to the rapid cooling operation; constructing a first training data set and a first test data set, a second training data set and a second test data set according to the first historical quenching data and the second historical quenching data; wherein the first training data set and the first test data set, the second training data set and the second test data set all include label one and label two, the label one is the hardness of the metal part after quenching, and the label two is the quality parameter of the metal part after quenching; using the first training data set and the first test data set to train and test the first deep prediction model, and using the second training data set and the second test data set to train and test the second deep prediction model, until the first deep prediction model and the second deep prediction model are trained to meet the standards.

3. A control method for quenching a metal product according to claim 2, characterized in that: After the first depth prediction model and the second depth prediction model are trained to meet the standards, the method also includes: using the first test data set to test the first depth prediction model that has been trained to meet the standards to obtain a set of predicted insulation times; screening out second training data that belong to the same historical quenching data as each first test data from the second training data set, using each screened second training data as the data body of the third training data, using the corresponding predicted insulation time as the label three of the third training data, and using the original label of each second training data, that is, the hardness and quality parameters of the metal part after quenching, as the label one and label two of the third training data, respectively, and each third training data constitutes a third training data set; using the third training data set to perform secondary training on the second depth prediction model that has been trained to meet the standards, and using the second test data set to test the second depth prediction model after the secondary training until the second depth prediction model is trained to meet the standards, that is, the final first depth prediction model and second depth prediction model are obtained.

4. A method for controlling quenching of a metal product according to claim 3, characterized in that: The first deep prediction model that has been trained to meet the standards is tested using the first test data set to obtain a set of predicted insulation times, including: determining the number of types of metal plates in the first test data set, determining a screening number based on the number of types, randomly screening some first test data from the first test data set according to the screening number, and the screened first test data constitute a third test data set; the first deep prediction model that has been trained to meet the standards is tested using the third test data set to obtain a set of predicted insulation times.

5. A method for controlling quenching of a metal product according to claim 4, characterized in that: The number of screenings is calculated based on the number of types and the following formula, including: ; In the formula, is the screening quantity; is the total amount of first test data in the first test data set; is the number of roughly classified types of metal plates in the first test data set, is the number of subdivision types of metal plates in the first test data set, is the total number of subdivision types of known metal plates; a is a constant value.

6. A method for controlling quenching of a metal product according to claim 5, characterized in that: The first depth prediction model is built based on CNN, and the second depth prediction model is built based on Transformer.

7. A control system for quenching a metal product, the system comprising a processing device and a storage device, characterized in that: The computer code stored in the storage device is called and executed by the processing device to implement the following steps: collecting training data, and using the training data to train the constructed first deep prediction model and the second deep prediction model; determining the target hardness of the metal part after quenching treatment, and using the first deep prediction model and the second deep prediction model to perform a deep analysis on the target hardness to obtain the target insulation time, cooling medium type and cooling medium temperature; performing insulation operation on the metal part heated to form austenite according to the target insulation time, and selecting the target cooling medium according to the cooling medium type and the cooling medium temperature, and using the target cooling medium to perform rapid cooling treatment on the austenitic metal part.

8. An electronic device comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 6.

9. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that: The computer program product includes computer codes, and when the computer codes are executed by a processor of an electronic device, the method according to any one of claims 1 to 6 is implemented.

Images