Heat treatment method and device of magnetic material, storage medium and computer equipment

Abstract

The invention discloses a heat treatment method and device for a magnetic material, a storage medium and computer equipment, relates to the technical field of vacuum sintering, and mainly aims at improving the special treatment effect on the magnetic material and improving the heat treatment efficiency. Comprising the following steps: in response to a heat treatment signal of a target magnetic material, obtaining material attribute information and heat treatment demand information of the target magnetic material and heat treatment furnace attribute information of a target heat treatment furnace; determining a material feature vector of the material attribute information, a demand feature vector of the heat treatment demand information and a heat treatment furnace feature vector of the heat treatment furnace attribute information; performing fusion processing on the material feature vector, the demand feature vector and the heat treatment furnace feature vector to obtain a heat treatment fusion feature vector; the heat treatment fusion feature vector is input into a preset treatment mode prediction model for heat treatment mode prediction, and a target heat treatment mode is obtained; and controlling the target heat treatment furnace to perform heat treatment on the target magnetic material based on the target heat treatment mode.

Description

Technical Field

[0001] The present invention relates to the field of vacuum sintering technology, and in particular to a method, a device, a storage medium and a computer device for heat treatment of magnetic materials. Background Art

[0002] Heat treatment can change the magnetic structure of magnetic materials, such as magnetic domain arrangement and magnetic moment direction, thereby increasing the magnetization intensity and saturation magnetic induction intensity of magnetic materials, thereby optimizing the magnetic properties of the materials and making them perform better in specific applications.

[0003] At present, in the process of heat treatment of magnetic materials, the heat treatment process is usually set manually. However, this method of manually setting the heat treatment process is greatly affected by human subjective factors, which will lead to poor heat treatment effect and low heat treatment efficiency. Summary of the invention

[0004] The present invention provides a method, device, storage medium and computer equipment for heat treatment of magnetic materials, which are mainly capable of improving the special treatment effect on magnetic materials and improving the heat treatment efficiency.

[0005] According to a first aspect of the present invention, there is provided a method for heat treatment of a magnetic material, comprising:

[0006] In response to a heat treatment signal of a target magnetic material, obtaining material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material;

[0007] Respectively determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information;

[0008] The material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fusion feature vector;

[0009] Inputting the heat treatment fusion feature vector into a preset treatment method prediction model to predict the heat treatment method, and obtaining a target heat treatment method corresponding to the target magnetic material;

[0010] Based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain a heat-treated magnetic material.

[0011] According to a second aspect of the present invention, there is provided a heat treatment device for magnetic materials, comprising:

[0012] An acquisition unit, configured to acquire material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material in response to a heat treatment signal of the target magnetic material;

[0013] A determination unit, used to respectively determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information;

[0014] A fusion unit, used for fusing the material feature vector, the demand feature vector, and the heat treatment furnace feature vector to obtain a heat treatment fusion feature vector;

[0015] A prediction unit, used for inputting the heat treatment fusion feature vector into a preset treatment method prediction model to predict the heat treatment method, and obtaining a target heat treatment method corresponding to the target magnetic material;

[0016] The heat treatment unit is used to control the target heat treatment furnace to perform heat treatment on the target magnetic material based on the target heat treatment method to obtain the magnetic material after heat treatment.

[0017] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned method for heat treatment of magnetic materials when executed by a processor.

[0018] According to a fourth aspect of the present invention, there is provided a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for heat treatment of magnetic materials when executing the program.

[0019] According to a method, device, storage medium and computer equipment for heat treatment of magnetic materials provided by the present invention, compared with the current method of manually setting the heat treatment process in the process of heat treatment of magnetic materials, the present invention obtains material property information, heat treatment requirement information and heat treatment furnace property information of a target heat treatment furnace for heat treatment of the target magnetic material by responding to the heat treatment signal of the target magnetic material; and respectively determines the material feature vector corresponding to the material property information, the demand feature vector corresponding to the heat treatment requirement information and the heat treatment furnace feature vector corresponding to the heat treatment furnace property information; at the same time, the material feature vector, the demand feature vector and the heat treatment furnace feature vector are fused to obtain a heat treatment fused feature vector; then the heat treatment fused feature vector is input into a preset processing method prediction model to predict the heat treatment method and obtain the target heat treatment method corresponding to the target magnetic material; finally, based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment. Therefore, a preset processing method prediction model is used to comprehensively analyze the material property information of the target magnetic material, the heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace to determine the heat treatment method. Finally, the heat treatment method is used to heat treat the magnetic material, which can avoid the problems of low determination accuracy and low determination efficiency caused by manual determination of the heat treatment process. Therefore, the present invention uses a model to determine that the determination efficiency and accuracy of the heat treatment method can be improved, thereby improving the heat treatment effect on the magnetic material. At the same time, the present invention can mine more hidden features and deeper features by fusing various feature vectors, and then predict the heat treatment method based on the hidden features and deeper features, which can further improve the determination accuracy of the heat treatment method, thereby further improving the heat treatment effect on the magnetic material, so that the magnetic material after heat treatment can better meet the performance requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0021] Figure 1 A flow chart of a heat treatment method for a magnetic material provided by an embodiment of the present invention is shown;

[0022] Figure 2 A flow chart of another method for heat treatment of magnetic materials provided by an embodiment of the present invention is shown;

[0023] Figure 3 A schematic structural diagram of a heat treatment device for magnetic materials provided by an embodiment of the present invention is shown;

[0024] Figure 4 A schematic structural diagram of another magnetic material heat treatment device provided by an embodiment of the present invention is shown;

[0025] Figure 5 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0026] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0027] At present, in the process of heat treatment of magnetic materials, the heat treatment process is manually set, which is greatly affected by human subjective factors, resulting in poor heat treatment effect and low heat treatment efficiency.

[0028] In order to solve the above problems, the present invention provides a method for heat treatment of magnetic materials. Figure 1 As shown, the method includes:

[0029] 101. In response to a heat treatment signal of a target magnetic material, obtain material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material.

[0030] Among them, material attribute information includes the type, shape, size and other information of the target magnetic material; heat treatment requirement information refers to the expected requirements to be achieved after heat treatment of the target magnetic material, such as expected magnetic properties, mechanical properties, chemical properties, etc.; heat treatment furnace attribute information includes: the furnace volume of the target heat treatment furnace, the type of heat treatment furnace, etc.

[0031] For the embodiment of the present invention, the present invention realizes the heat treatment process of magnetic materials through a fully automatic multiple continuous sintering control system for a heat treatment furnace. The fully automatic multiple continuous sintering control system for a heat treatment furnace is a visual human-computer interaction system for operating and managing a heat treatment furnace. Through the control system, the user can flexibly set various parameters on the touch screen and the temperature control table, and intuitively obtain equipment operation information and data, thereby realizing easy and effective equipment operation and management. Most of the heat treatment processes of magnetic materials require 3 or even more high-temperature sintering, cooling processes, etc. to achieve ideal characteristics. The development of this control system provides customers with efficient solutions for processing different products. This system uses the time signal of the temperature control table (the time signal of the cooling process can also be stored in the PLC (Programmable Logic Controller) according to the number of steps), and can realize multiple heating and cooling processes in a temperature control curve. The number of heating and cooling times can be set arbitrarily, which is sufficient to meet the customer's needs for customizing different production processes for different products. Combined with the parameter setting of the HMI (HumanMachine Interface) and the logic program of the PLC, the fully automatic multiple continuous sintering function is easier to achieve. The control system includes: Touch screen: The Mitsubishi embedded touch screen on the control cabinet is the main control touch screen of the equipment. It has the functions of operating the operation of each mechanism of the equipment, displaying the operation status, editing the process curve, setting the process parameters and recording the operation data. The following introduces the specific content and operation methods of each screen of several touch screens in the control system: 1. Main screen: There is a preset QR code of the company at the preset position of the main screen. Scan the QR code to follow the company's official account to understand the company's corporate dynamics in real time. The main interface also contains the switch buttons of each window. 2. Operating system: Contains the main operation area of ​​the equipment. The upper right corner of each mechanism indicates whether the operating conditions of the system are met. "×" means not met, and "√" means met. For example, when the operating conditions of the system are met, long press the mechanism button for 2 seconds to start running, and long press the mechanism button for 2 seconds again to stop running. When the operating conditions of the mechanism are not met, long press the mechanism button for 2 seconds, and the mechanism operating condition window pops up, which can clearly see which conditions are not met. The lower part mainly displays real-time data, including heating curve data, furnace vacuum data, and temperature data. 3. Operation condition window: For example, click the operation condition button on the operation screen to open the operation condition window. Click each mechanism button again to open the corresponding operation condition window. A gray button indicates that the operation condition of the mechanism is not met, and a green button indicates that it is met.4. Temperature control system: For example, the top of the interface can be a real-time temperature curve, and the right side is a heating hold function button. Pressing the button will pop up the hold time setting window. If the time is set to 0, it will be permanently held until the hold function is manually canceled. When the time is set to a specific value, the hold function will be automatically canceled after the hold time reaches the set value. The right side is the heating curve skip function button. To prevent misoperation, you need to press the button for 2 seconds to execute the skip function. The bottom of the screen can display real-time data, including heating curve data, furnace vacuum data, and temperature data. 5. Password window: A password is required to enter the parameter setting. 6. Parameter setting window: Parameter settings can be entered manually or automatically generated. 7. Historical data window: This window records important data of the equipment. For example, it can be set to record once every 10 seconds and save it to the database regularly, or it can be saved manually. 8. Historical curve window: You can accurately find the curve of a certain period of time or the data at a certain moment. 9. Running time window: Record the running time or running times of each mechanism, and each mechanism can be regularly inspected and maintained based on this record. 10. Action record window: This window records the action status of each mechanism and the limit position, which is convenient for customers to check whether the operation is performed in accordance with the process requirements, and is convenient for maintenance workers to repair when a fault occurs. 11. Alarm screen: For example, the alarm bar is red to indicate that this alarm has occurred, and the electrode alarm button can view detailed alarm countermeasures. 12. Alarm record window: This window records all alarms that have occurred. In addition, the control system also includes a temperature control meter. The temperature control meter mainly uses its time signal function as a cooling signal. Multiple cooling signals can be set in a temperature control curve, that is, multiple "sintering-cooling" processes can be realized. The fully automatic multiple continuous sintering control system of the heat treatment furnace in the embodiment of the present invention is composed of control and execution elements such as human-machine interface configuration, PLC, temperature control meter and sensor, which provides users with a safe, efficient, simple and practical equipment control and management method.

[0032] Specifically, when the target magnetic material needs to be heat treated, a heat treatment signal for the target magnetic material is triggered in the control system, and the control system is linked to a database. Based on the identification information of the target magnetic material carried in the control signal, the material property information of the target magnetic material is obtained in the database. At the same time, the staff can select or input the heat treatment requirement information in the corresponding interface of the control system. The control system will select a target heat treatment furnace that meets the heat treatment conditions from multiple heat treatment furnaces, and obtain the heat treatment furnace property information of the target heat treatment furnace in the database. Then, the material property information, the heat treatment requirement information, and the heat treatment furnace property information are comprehensively analyzed to determine the best heat treatment method for the target magnetic material. Through the best heat treatment method, a temperature control curve can be generated. The heat treatment furnace fully automatic multiple continuous sintering control system performs heat treatment on the magnetic material according to the process parameters in the temperature control curve. If the magnetic material needs to be sintered and cooled multiple times, it can be achieved through the temperature control curve, avoiding the time wasted by determining multiple heating process curves for multiple heating and cooling. Therefore, the embodiment of the present invention can improve the heat treatment efficiency of the magnetic material. The embodiment of the present invention is mainly applied to the scenario of heat treatment of magnetic materials. The execution subject of the embodiment of the present invention is a device or equipment capable of heat treatment of magnetic materials, which can be specifically set on the client or server side.

[0033] 102. Determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information.

[0034] 103. The material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fusion feature vector.

[0035] For the embodiments of the present invention, word embedding and other methods can be used to convert material attribute information into material feature vectors, heat treatment requirement information into requirement feature vectors, and heat treatment furnace attribute information into heat treatment furnace feature vectors. The material feature vectors, requirement feature vectors, and heat treatment furnace feature vectors are then fused, and finally the fused feature vectors are used to predict the heat treatment method. By fusing the material feature vectors, requirement feature vectors, and heat treatment furnace feature vectors, it is possible to combine multiple features and extract more implicit features at the same time, thereby improving the heat treatment accuracy of magnetic materials. At the same time, by fusing multiple features into one feature, the input of the model can be reduced, that is, the input dimension of the model is reduced, thereby optimizing computing resources, improving the prediction efficiency of the model, and further improving the heat treatment efficiency of magnetic materials.

[0036] 104. Input the heat treatment fusion feature vector into a preset processing method prediction model to predict the heat treatment method, and obtain a target heat treatment method corresponding to the target magnetic material.

[0037] 105. Based on the target heat treatment method, control the target heat treatment furnace to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment.

[0038] Among them, the target heat treatment method includes: heating method, insulation method and cooling method. The heating method includes heating form (such as electric heating, etc.), heating temperature, heating time and other information. The insulation method includes: insulation time, etc. The cooling method includes: cooling form (water quenching, air cooling, etc.), cooling time, cooling speed, etc.

[0039] For the embodiment of the present invention, in order to improve the prediction accuracy of the preset processing method prediction model, it is necessary to pre-train and construct the preset processing method prediction model. Based on this, the method includes: constructing multiple preset initial models and obtaining sample data, wherein the sample data includes sample material attribute information of the sample magnetic material, sample heat treatment requirement information, sample heat treatment furnace attribute information of the sample heat treatment furnace for heat treating the sample magnetic material, and obtaining the actual heat treatment method of the sample magnetic material; based on the number of preset initial models, the sample data is divided into multiple groups of training data and multiple groups of test data; using each of the training data to train the corresponding preset initial model to obtain the trained preset initial model, wherein the training process is: determining the sample material attribute information in the training data The sample material feature vector corresponding to the sample heat treatment requirement information, the sample requirement feature vector corresponding to the sample heat treatment furnace attribute information, and the sample furnace feature vector corresponding to the sample heat treatment furnace attribute information are fused to obtain the sample fused feature vector, and the sample fused feature vector is used as the input data of the preset initial model, and the actual heat treatment method is used as the output data of the preset initial model; the prediction accuracy of the corresponding trained preset initial model is tested by using each of the test data to obtain the prediction accuracy of each of the trained preset initial models; the maximum prediction accuracy is determined among the prediction accuracies, and the trained preset initial model corresponding to the maximum prediction accuracy is determined as the preset processing method prediction model. Furthermore, after the material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused, the fused heat treatment fusion feature vector is input into a preset processing method prediction model to predict the heat treatment method, and finally the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material using the target heat treatment method output by the model. The optimal heat treatment method is predicted by the model, which can avoid wasted time and erroneous determination of the heat treatment method manually. Therefore, the embodiment of the present invention can improve the heat treatment efficiency and heat treatment accuracy of the magnetic material, thereby improving the heat treatment effect on the magnetic material, so that the magnetic material after heat treatment can meet user needs.

[0040] According to a method for heat treatment of magnetic materials provided by the present invention, compared with the current method of manually setting the heat treatment process in the process of heat treatment of magnetic materials, the present invention obtains material property information, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treatment of the target magnetic material by responding to a heat treatment signal of the target magnetic material; and respectively determines a material feature vector corresponding to the material property information, a demand feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace property information; at the same time, the material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fused feature vector; then the heat treatment fused feature vector is input into a preset processing method prediction model to predict the heat treatment method, and the target heat treatment method corresponding to the target magnetic material is obtained; finally, based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment. Therefore, a preset processing method prediction model is used to comprehensively analyze the material property information of the target magnetic material, the heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace to determine the heat treatment method. Finally, the heat treatment method is used to heat treat the magnetic material, which can avoid the problems of low determination accuracy and low determination efficiency caused by manual determination of the heat treatment process. Therefore, the present invention uses a model to determine that the determination efficiency and accuracy of the heat treatment method can be improved, thereby improving the heat treatment effect on the magnetic material. At the same time, the present invention can mine more hidden features and deeper features by fusing various feature vectors, and then predict the heat treatment method based on the hidden features and deeper features, which can further improve the determination accuracy of the heat treatment method, thereby further improving the heat treatment effect on the magnetic material, so that the magnetic material after heat treatment can better meet the performance requirements.

[0041] Further, in order to better illustrate the above process of classifying data, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another method for heat treatment of magnetic materials, such as Figure 2 As shown, the method includes:

[0042] 201. In response to a heat treatment signal of a target magnetic material, obtain material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material.

[0043] For an embodiment of the present invention, after receiving a heat treatment signal of a target magnetic material, in order to perform heat treatment on the target magnetic material, it is first necessary to determine a target heat treatment furnace that meets the heat treatment conditions. Based on this, the method includes: determining storage location information corresponding to the target magnetic material, and determining different idle heat treatment furnaces in the heat treatment site where the target magnetic material is located, and obtaining furnace attribute information of each idle heat treatment furnace; determining an attribute feature vector corresponding to the furnace attribute information, and clustering each idle heat treatment furnace based on the attribute feature vector to obtain idle heat treatment furnaces under different clustering categories; based on the material attribute information and the heat treatment requirement information, determining a target clustering category that meets the heat treatment requirement in the different clustering categories; determining the furnace location information of each idle heat treatment furnace under the target clustering category, and determining the target heat treatment furnace among each idle heat treatment furnace under the target clustering category based on the storage location information of the target magnetic material and the furnace location information of each idle heat treatment furnace.

[0044] Among them, an idle heat treatment furnace refers to a heat treatment furnace that can operate normally after receiving a heat treatment signal, in which no heat treatment work is performed in the special treatment site where the target magnetic material is located, and there is no material that needs to be processed in the furnace body; the furnace attribute information of the idle heat treatment furnace includes: the furnace type, controllable temperature range, heating method, furnace structure, furnace chamber volume, etc. of the idle heat treatment furnace; different clustering categories can be divided according to furnace type, controllable temperature range, heating method, furnace structure, furnace chamber volume, etc., and the embodiment of the present invention does not specifically limit the different clustering categories.

[0045] For the embodiment of the present invention, in order to accurately and quickly select the target heat treatment furnace that meets the heat treatment requirements, it is first necessary to cluster the idle heat treatment furnaces. Based on this, the method includes: initializing the centroid vectors corresponding to different clusters; calculating the cosine similarities between each of the attribute feature vectors and the centroid vectors corresponding to the different clusters, and based on each of the cosine similarities, dividing each of the idle heat treatment furnaces into the different clusters; based on the attribute feature vectors corresponding to the idle heat treatment furnaces in the different clusters, determining the updated centroid vectors corresponding to the different clusters; based on the updated centroid vectors, re-dividing each of the idle heat treatment furnaces into the different clusters until the updated centroid vector does not change, and determining the idle heat treatment furnaces finally divided into the different clusters as idle heat treatment furnaces under the different clustering categories.

[0046] Specifically, the centroid vectors corresponding to the initial centroids corresponding to K clusters are selected, and the cosine similarities between each attribute feature vector and the K centroid vectors are calculated for the attribute feature vectors corresponding to each idle heat treatment furnace. Furthermore, after calculating the cosine similarities between each idle heat treatment furnace and the centroid vectors of each cluster, each attribute feature vector is assigned to the cluster corresponding to the centroid vector with the largest cosine similarity. Then, for each cluster, the centroid of each cluster and its corresponding centroid vector are recalculated, and each idle heat treatment furnace is re-divided into different clusters. In this way, each idle heat treatment furnace is continuously divided until the position of the centroid does not change, and finally the idle heat treatment furnaces divided into different clusters are determined as idle heat treatment furnaces under different clustering categories.

[0047] For example, different idle heat treatment furnaces are clustered according to the controllable temperature range and the furnace chamber volume to obtain idle heat treatment furnaces under different clustering categories, and then the target clustering category matching the volume and heating temperature requirements of the magnetic material is selected according to the volume of the target magnetic material and the heating temperature requirements in the heat treatment demand information, and then the idle heat treatment furnace closest to the target magnetic material is selected from each idle heat treatment furnace under the target clustering category, and the idle heat treatment furnace is determined as the target heat treatment furnace that will perform heat treatment on the target magnetic material. In the embodiment of the present invention, if the heat treatment furnaces are not clustered and a suitable heat treatment furnace is selected according to the clustering results, but a heat treatment furnace is randomly selected, if a heat treatment furnace with a smaller volume is selected to perform heat treatment on the material, the material needs to be heat treated multiple times, thereby increasing the heat treatment time. Therefore, by clustering the idle heat treatment furnaces, a heat treatment furnace that matches and meets the heat treatment requirements can be found, thereby improving the heat treatment efficiency of the material and also improving the heat treatment effect of the magnetic material.

[0048] 202. Determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information.

[0049] 203. Determine the principal component eigenvector between every two eigenvectors among the material eigenvector, the demand eigenvector, and the heat treatment furnace eigenvector, including: determining the first principal component eigenvector between the material eigenvector and the demand eigenvector, the second principal component eigenvector between the material eigenvector and the heat treatment furnace eigenvector, and the third principal component eigenvector between the demand eigenvector and the heat treatment furnace eigenvector.

[0050] 204. Transversely concatenate the first principal component eigenvector, the second principal component eigenvector, and the third principal component eigenvector respectively to obtain a concatenated eigenvector.

[0051] 205. Determine the transformation coefficient corresponding to the concatenated feature vector, and based on the transformation coefficient, perform linear transformation on the concatenated feature vector to obtain a heat-treated fusion feature vector.

[0052] For the embodiment of the present invention, the word embedding method such as Word2Vec can be used to respectively determine the material feature vector corresponding to the material attribute information, the demand feature vector corresponding to the heat treatment requirement information, and the heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information. Then, in order to extract deeper features in the material attribute information, the heat treatment requirement information, and the heat treatment furnace attribute information, it is necessary to fuse the material feature vector, the demand feature vector, and the heat treatment furnace feature vector. The specific fusion processing method is: determine the principal component feature vector between every two feature vectors in the material feature vector, the demand feature vector, and the heat treatment furnace feature vector, wherein the principal component feature vector between every two feature vectors is specifically determined. The method for determining the principal component eigenvector between two eigenvectors includes: constructing a feature matrix based on every two eigenvectors; determining the element mean of each element in the feature matrix, and subtracting each element in the feature matrix from the element mean to obtain a centralized feature matrix; determining the covariance matrix corresponding to the centralized feature matrix; performing eigenvalue decomposition on the covariance matrix to obtain matrix eigenvalues ​​and matrix eigenvectors; based on the size of the matrix eigenvalues, selecting a preset number of matrix eigenvectors from the matrix eigenvectors, and determining the preset number of matrix eigenvectors as the principal component eigenvectors between every two eigenvectors.

[0053] Among them, the preset number is a value set according to actual demand (experience); every two feature vectors include a material feature vector and a demand feature vector, a material feature vector and a heat treatment furnace feature vector, and a demand feature vector and a heat treatment furnace feature vector. Specifically, if every two feature vectors are: a material feature vector and a demand feature vector, the material feature vector is [11 2 4 2], and the demand feature vector is [1 3 3 4 4], the feature matrix composed of the material feature vector and the demand feature vector is as follows:

[0054]

[0055] Among them, T represents the feature matrix. Then, the element mean corresponding to the elements in the first row of the feature matrix is ​​determined, and each element in the first row of elements is subtracted from the mean to obtain the subtraction result corresponding to each element in the first row. At the same time, the element mean corresponding to each element in the second row is determined, and each element in the second row is subtracted from the corresponding element mean to obtain the subtraction result corresponding to each element in the second row. Finally, the subtraction result corresponding to each element in the first row and the subtraction result corresponding to each element in the second row are used to form the centralized feature matrix as follows:

[0056]

[0057] Among them, T z Represents the centralized feature matrix, and then calculates the covariance between each row of the centralized feature matrix. For example, the covariance Cov(X 1 , X 1 )=[(-1) 2 +(-1) 2 +(0) 2 +(2) 2 +(0) 2 ] / (5-1)=1.5, the covariance Cov(X 1 , X 2 )=[(-1)×(-2)+(-1)×0+0×0+2×1+0×1] / (5-1)=1, thereby calculating the covariance between all row elements, and then forming a covariance matrix from all covariances, and performing eigenvalue decomposition on the covariance matrix. The specific decomposition method is: first determine the eigenvector group of the covariance matrix, and calculate the corresponding eigenvalue according to the eigenvector group. For example, if the obtained eigenvector group is k groups, then the corresponding eigenvalues ​​are k, and each eigenvalue has its corresponding eigenvector. Then, the k eigenvalues ​​are sorted from large to small to obtain the sorted eigenvalues, and then the eigenvalues ​​are sorted. The eigenvalues ​​ranked in the top n (preset number) are selected from the sorted eigenvalues, and the eigenvectors corresponding to the top n eigenvalues ​​are determined as the principal component eigenvectors, that is, the first principal component eigenvector between the material eigenvector and the demand eigenvector. Similarly, the second principal component eigenvector between the material eigenvector and the heat treatment furnace eigenvector, and the third principal component eigenvector between the demand eigenvector and the heat treatment furnace eigenvector can be determined in the above manner. Then, the first principal component eigenvector, the second principal component eigenvector, and the third principal component eigenvector are horizontally spliced ​​to obtain a spliced ​​eigenvector. For example, if the first principal component eigenvector is f(a 1 , b 1 ), the second principal component eigenvector is f(a 2 , b 2 ), the third principal component eigenvector is f(a 3 , b 3 ), the transformation coefficient is preset as w, the above three principal component feature vectors are horizontally spliced, and the heat treatment fusion feature vector is obtained by transformation based on the transformation coefficient as f(w(a 1 , b 1 , a 2 , b 2 , a 3 , b 3)), wherein the transformation coefficient can be set according to actual needs. The material property information, heat treatment requirement information, and heat treatment furnace property information in the embodiment of the present invention belong to data from different data sources. By fusing the data from different data sources into more significant and deeper features, and using the model to analyze the more significant and deeper features, the prediction performance of the model can be improved, thereby improving the heat treatment effect of the magnetic material.

[0058] 206. Input the heat treatment fusion feature vector into a preset processing method prediction model to predict the heat treatment method, and obtain a target heat treatment method corresponding to the target magnetic material.

[0059] For the embodiment of the present invention, after the material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain the heat treatment fused feature vector, the heat treatment fused feature vector is input into the preset processing method prediction model to predict the heat treatment method. The preset processing method prediction model can output the target heat treatment method corresponding to the target magnetic material, and finally the target heat treatment method is used to heat treat the target magnetic material. Thus, the heat treatment method is predicted by the pre-trained preset processing method prediction model, which can avoid wasted time and prediction errors in manually determining the heat treatment method. Therefore, the embodiment of the present invention can improve the prediction efficiency and prediction accuracy of the heat treatment method, and further improve the heat treatment efficiency and heat treatment effect of the magnetic material.

[0060] 207. Based on the target heat treatment method, control the target heat treatment furnace to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment.

[0061] Among them, the target heat treatment method includes: heating treatment method, insulation treatment method and cooling treatment method. The heating treatment method includes: heating method (resistance heating, etc.), heating time, heating temperature, etc. The insulation treatment method includes: insulation method, insulation time, etc. The cooling treatment method includes: cooling method (water cooling, air cooling, etc.), cooling speed, etc.

[0062] For the embodiment of the present invention, the preset processing method prediction model can predict the heating treatment method, insulation treatment method, and cooling treatment method that meet the requirements of the heat treatment process. Then, the target magnetic material needs to be heat treated in the above-mentioned manner. Based on this, step 207 specifically includes: based on the heating method, heating time, and heating temperature, controlling the target heat treatment furnace to heat the target magnetic material to obtain the magnetic material after the heating treatment; based on the insulation method and insulation time, controlling the target heat treatment furnace to insulation treat the magnetic material after the heating treatment to obtain the magnetic material after insulation treatment; based on the cooling method and cooling speed, controlling the target heat treatment furnace to cool the magnetic material after the insulation treatment to obtain the magnetic material after the heat treatment.

[0063] Specifically, first, according to the heating method, heating time, and heating temperature, the target heat treatment furnace is controlled to heat the target magnetic material, and then according to the insulation method and insulation time, the target heat treatment furnace is controlled to perform insulation treatment on the magnetic material after the heating treatment, and finally according to the cooling method and cooling speed, the target heat treatment furnace is controlled to cool the magnetic material after the insulation treatment to obtain the heat-treated magnetic material. Thus, the model is used to predict the heating method, insulation method, and cooling method of the magnetic material during the heat treatment process, which can realize the full automation of the heat treatment process and improve the heat treatment efficiency of the magnetic material.

[0064] Furthermore, during the heat treatment process of the target magnetic material, in order to ensure the safety and stability of the heat treatment process and to ensure the heat treatment effect, it is also necessary to monitor the heat treatment process in real time. Based on this, the method includes: during the heat treatment process of the target magnetic material, obtaining the heating property information of the target heat treatment furnace, wherein the heating property information includes: at least one of the real-time temperature, real-time vacuum degree, and real-time pressure in the target heat treatment furnace; based on the heating property information, judging whether the target heat treatment furnace meets the preset conditions for the alarm; if the target heat treatment furnace meets the preset conditions, generating an alarm message based on the heating property information, and calling a preset communication tool interface to send the alarm message to the monitoring terminal.

[0065] Specifically, in the process of heat treating the target magnetic material, the real-time temperature, real-time vacuum degree, and real-time pressure of the target heat treatment furnace during the heat treatment process are obtained in real time, wherein the real-time temperature can be the temperature of each position in the target heat treatment furnace. According to the temperature of each position, it is judged whether the target heat treatment furnace meets the condition of uniform temperature distribution. If the temperature distribution is uneven, an alarm message is generated. At the same time, it is judged whether the real-time vacuum degree is within a preset vacuum degree range (wherein, the preset vacuum degree range is set according to actual needs). If it is not within the preset vacuum degree range, an alarm message is generated. At the same time, it is also necessary to judge whether the real-time pressure is within a preset pressure range (wherein, the preset pressure range is set according to actual needs). If not, an alarm message is generated. Finally, the communication tool interface such as email and text message is called to send the alarm message to the monitoring personnel of the monitoring terminal, so that the monitoring personnel of the monitoring terminal can correct the heat treatment process based on the alarm information, thereby achieving stable operation of the heat treatment process and ensuring the heat treatment effect of the magnetic material.

[0066] Furthermore, during the heat treatment process, in order to further improve the safety of the heat treatment, the following settings can also be made:

[0067] A. Overpressure setting in the furnace: When the pressure in the furnace exceeds the set value, the system will alarm for overpressure.

[0068] B. Over-temperature setting in the furnace: When the temperature in the furnace exceeds the set value, the system will give an over-temperature alarm.

[0069] C. Roots pump starting pressure: Below the set pressure, the Roots pump is allowed to start.

[0070] D. Heating operating pressure: When heating is turned on and running, the pressure in the furnace must be below the set value, otherwise the heating will stop.

[0071] E. Heating to start vacuum: Heating is allowed to be turned on below the set value. Once heating is turned on, it will not be limited by the vacuum degree. Even if the vacuum degree rises above the set value, heating will not stop.

[0072] F. Heating tolerance vacuum: After heating is turned on, when the vacuum degree rises above the set value, heating will stop.

[0073] G. Vent door opening temperature: If the temperature is lower than this set value, it means the cooling temperature is in place, allowing the vent valve to open, the lock ring to loosen, and the gate valve to open.

[0074] H. Furnace charging pressure: The pressure of charging before the material is discharged after heating and cooling are completed. This pressure is set to be equal to the atmospheric pressure.

[0075] I. Door opening pressure: When the lock ring is loosened and the air is released to this pressure, the furnace door can be opened. This pressure is set to be equal to the atmospheric pressure.

[0076] J. Inflation timeout: The time starts from when the inflation valve is opened. When the time to inflate to the upper limit of the air-cooled inflation pressure is greater than the set value, an alarm will be triggered.

[0077] K. Partial pressure inflation timeout: The timing starts when the partial pressure inflation valve is opened. When the time to inflate to the upper pressure limit is greater than the set value, an alarm will be triggered.

[0078] L. Air-cooled inflation pressure upper limit: set the inflation pressure upper limit during air-cooling; Air-cooled inflation pressure lower limit: set the inflation pressure lower limit during air-cooling. First, inflate to the middle value between the upper limit and the lower limit (for example, if the middle value exceeds 100, this value becomes 85), then you can turn on the fan, and continue to inflate to the upper limit after the fan is turned on. When the pressure drops to the lower limit, inflate again to the upper limit.

[0079] M. Small pressure, large pressure partial pressure setting: partial pressure inflates to the upper pressure limit, then opens the exhaust valve to exhaust to the lower pressure limit, and repeats the "inflation-exhaust" action.

[0080] N. The following three parameters are the parameters for automatic operation: Cooling and charging temperature: charging is allowed below this set value; Fan start temperature: charging is completed, air cooling is allowed below this set value; Air cooling end temperature: below this set value, it indicates that air cooling has ended.

[0081] O. Curve measurement temperature difference: When the temperature difference between the three zones exceeds the set value, an alarm will be triggered and the heating curve will be maintained. When it is lower than the set value, the heating curve will be cancelled and the temperature will continue to rise. This function is invalid when the set value is 0: Starting temperature: This function is valid when the starting temperature is exceeded.

[0082] P. Curve measurement temperature difference: When the difference between the actual temperature value and the S measured temperature value exceeds the set value, an alarm will be triggered and the heating curve will be maintained. When it is lower than the set value, the heating curve will be canceled and the temperature will continue to rise. This function is invalid when the set value is 0: Starting temperature: This function is valid when the starting temperature is exceeded.

[0083] Q. Heating curve setting: Setting of heating curve number. It can be input manually or generated automatically.

[0084] According to another method for heat treatment of magnetic materials provided by the present invention, compared with the current method of manually setting the heat treatment process in the process of heat treatment of magnetic materials, the present invention obtains the material property information, heat treatment requirement information, and heat treatment furnace property information of the target heat treatment furnace for heat treatment of the target magnetic material by responding to the heat treatment signal of the target magnetic material; and respectively determines the material feature vector corresponding to the material property information, the demand feature vector corresponding to the heat treatment requirement information, and the heat treatment furnace feature vector corresponding to the heat treatment furnace property information; at the same time, the material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fused feature vector; then the heat treatment fused feature vector is input into a preset processing method prediction model to predict the heat treatment method, and the target heat treatment method corresponding to the target magnetic material is obtained; finally, based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment. Therefore, a preset processing method prediction model is used to comprehensively analyze the material property information of the target magnetic material, the heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace to determine the heat treatment method. Finally, the heat treatment method is used to heat treat the magnetic material, which can avoid the problems of low determination accuracy and low determination efficiency caused by manual determination of the heat treatment process. Therefore, the present invention uses a model to determine that the determination efficiency and accuracy of the heat treatment method can be improved, thereby improving the heat treatment effect on the magnetic material. At the same time, the present invention can mine more hidden features and deeper features by fusing various feature vectors, and then predict the heat treatment method based on the hidden features and deeper features, which can further improve the determination accuracy of the heat treatment method, thereby further improving the heat treatment effect on the magnetic material, so that the magnetic material after heat treatment can better meet the performance requirements.

[0085] Further, as Figure 1 In a specific implementation, an embodiment of the present invention provides a heat treatment device for magnetic materials, such as Figure 3 As shown, the device includes: an acquisition unit 31, a determination unit 32, a fusion unit 33, a prediction unit 34, and a heat treatment unit 35.

[0086] The acquisition unit 31 can be used to respond to the heat treatment signal of the target magnetic material to acquire material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of the target heat treatment furnace for heat treating the target magnetic material.

[0087] The determination unit 32 may be used to respectively determine a material feature vector corresponding to the material property information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace property information.

[0088] The fusion unit 33 may be used to fuse the material feature vector, the demand feature vector, and the heat treatment furnace feature vector to obtain a heat treatment fusion feature vector.

[0089] The prediction unit 34 may be used to input the heat treatment fusion feature vector into a preset treatment method prediction model to perform heat treatment method prediction, and obtain a target heat treatment method corresponding to the target magnetic material.

[0090] The heat treatment unit 35 may be used to control the target heat treatment furnace to perform heat treatment on the target magnetic material based on the target heat treatment method to obtain the heat-treated magnetic material.

[0091] In specific application scenarios, in order to heat treat the target magnetic material, such as Figure 4 As shown, the heat treatment unit 35 includes a heating module 351 , a heat preservation module 352 , and a cooling module 353 .

[0092] The heating module 351 can be used to control the target heat treatment furnace to heat the target magnetic material based on the heating method, heating time, and heating temperature to obtain the magnetic material after the heating treatment.

[0093] The insulation module 352 can be used to control the target heat treatment furnace to perform insulation treatment on the magnetic material after the heating treatment based on the insulation method and insulation time, so as to obtain the magnetic material after the insulation treatment.

[0094] The cooling module 353 can be used to control the target heat treatment furnace to cool the magnetic material after the insulation treatment based on the cooling method and cooling speed to obtain the magnetic material after the heat treatment.

[0095] In a specific application scenario, in order to monitor the heat treatment process in real time, the device further includes: a judgment unit 36 ​​and an alarm unit 37 .

[0096] The acquisition unit 31 can also be used to acquire heating property information of the target heat treatment furnace during the heat treatment of the target magnetic material, wherein the heating property information includes at least one of real-time temperature, real-time vacuum degree, and real-time pressure in the target heat treatment furnace.

[0097] The judgment unit 36 ​​may be configured to judge whether the target heat treatment furnace meets preset alarm conditions based on the heating property information.

[0098] The alarm unit 37 can be used to generate alarm information based on the heating property information if the target heat treatment furnace meets the preset conditions, and call a preset communication tool interface to send the alarm information to the monitoring terminal.

[0099] In a specific application scenario, in order to perform fusion processing on the feature vectors, the fusion unit 33 includes a determination module 331 , a splicing module 332 , and a transformation module 333 .

[0100] The determination module 331 can be used to determine the principal component feature vector between every two feature vectors among the material feature vector, the demand feature vector, and the heat treatment furnace feature vector, including: determining the first principal component feature vector between the material feature vector and the demand feature vector, the second principal component feature vector between the material feature vector and the heat treatment furnace feature vector, and the third principal component feature vector between the demand feature vector and the heat treatment furnace feature vector.

[0101] The splicing module 332 can be used to horizontally splice the first principal component eigenvector, the second principal component eigenvector, and the third principal component eigenvector to obtain a spliced ​​eigenvector.

[0102] The transformation module 333 may be used to determine the transformation coefficient corresponding to the concatenated feature vector, and based on the transformation coefficient, perform a linear transformation on the concatenated feature vector to obtain a heat-processed fused feature vector.

[0103] In a specific application scenario, in order to determine the principal component eigenvector, the determination module 331 can be specifically used to construct a feature matrix based on every two eigenvectors; determine the element mean of each element in the feature matrix, and subtract each element in the feature matrix from the element mean to obtain a centralized feature matrix; determine the covariance matrix corresponding to the centralized feature matrix; perform eigenvalue decomposition on the covariance matrix to obtain matrix eigenvalues ​​and matrix eigenvectors; based on the size of the matrix eigenvalues, select a preset number of matrix eigenvectors from the matrix eigenvectors, and determine the preset number of matrix eigenvectors as the principal component eigenvectors between every two eigenvectors.

[0104] In a specific application scenario, in order to determine a target heat treatment furnace that meets the heat treatment requirements, the device further includes a clustering unit 38 .

[0105] The determination unit 32 may also be used to determine storage location information corresponding to the target magnetic material, determine different idle heat treatment furnaces in the heat treatment site where the target magnetic material is located, and obtain furnace attribute information of each of the idle heat treatment furnaces.

[0106] The clustering unit 38 may be used to determine the attribute feature vector corresponding to the furnace attribute information, and based on the attribute feature vector, perform clustering processing on each of the idle heat treatment furnaces to obtain idle heat treatment furnaces in different clustering categories.

[0107] The determining unit 32 may be specifically configured to determine a target clustering category that meets the heat treatment requirement from among the different clustering categories based on the material property information and the heat treatment requirement information.

[0108] The determination unit 32 can be specifically used to determine the furnace location information of each idle heat treatment furnace under the target clustering category, and determine the target heat treatment furnace among the idle heat treatment furnaces under the target clustering category based on the storage location information of the target magnetic material and the furnace location information of each idle heat treatment furnace.

[0109] In a specific application scenario, in order to cluster each idle heat treatment furnace, the clustering unit 38 can be specifically used to initialize the centroid vectors corresponding to different clusters; calculate the cosine similarity between each attribute feature vector and the centroid vector corresponding to the different clusters, and divide each idle heat treatment furnace into the different clusters based on each cosine similarity; based on the attribute feature vectors corresponding to the idle heat treatment furnaces in the different clusters, determine the updated centroid vectors corresponding to the different clusters; based on the updated centroid vectors, re-divide each idle heat treatment furnace into the different clusters until the updated centroid vector does not change, and the idle heat treatment furnaces finally divided into the different clusters are determined as idle heat treatment furnaces under the different clustering categories.

[0110] It should be noted that for other corresponding descriptions of the functional modules involved in the heat treatment device for magnetic materials provided in the embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0111] Based on the above Figure 1The method shown, accordingly, the embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the following steps are implemented: in response to the heat treatment signal of the target magnetic material, the material property information, heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace for heat treating the target magnetic material are obtained; the material feature vector corresponding to the material property information, the requirement feature vector corresponding to the heat treatment requirement information, and the heat treatment furnace feature vector corresponding to the heat treatment furnace property information are respectively determined; the material feature vector, the requirement feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fused feature vector; the heat treatment fused feature vector is input into a preset processing method prediction model to predict the heat treatment method, and the target heat treatment method corresponding to the target magnetic material is obtained; based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment.

[0112] Based on the above Figure 1 The method shown and Figure 3 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 5 As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor, wherein the memory 42 and the processor 41 are both arranged on a bus 43, and the processor 41 implements the following steps when executing the program: in response to a heat treatment signal of a target magnetic material, obtaining material property information, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material; respectively determining a material feature vector corresponding to the material property information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace property information; fusing the material feature vector, the requirement feature vector, and the heat treatment furnace feature vector to obtain a heat treatment fusion feature vector; inputting the heat treatment fusion feature vector into a preset processing method prediction model to perform heat treatment method prediction to obtain a target heat treatment method corresponding to the target magnetic material; and based on the target heat treatment method, controlling the target heat treatment furnace to perform heat treatment on the target magnetic material to obtain a heat-treated magnetic material.

[0113] Through the technical solution of the present invention, the present invention obtains the material property information, heat treatment requirement information, and heat treatment furnace property information of the target magnetic material in response to the heat treatment signal of the target magnetic material; and respectively determines the material feature vector corresponding to the material property information, the demand feature vector corresponding to the heat treatment requirement information, and the heat treatment furnace feature vector corresponding to the heat treatment furnace property information; at the same time, the material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fused feature vector; then the heat treatment fused feature vector is input into a preset processing method prediction model to predict the heat treatment method, and the target heat treatment method corresponding to the target magnetic material is obtained; finally, based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain the magnetic material after heat treatment. Therefore, a preset processing method prediction model is used to comprehensively analyze the material property information of the target magnetic material, the heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace to determine the heat treatment method. Finally, the heat treatment method is used to heat treat the magnetic material, which can avoid the problems of low determination accuracy and low determination efficiency caused by manual determination of the heat treatment process. Therefore, the present invention uses a model to determine that the determination efficiency and accuracy of the heat treatment method can be improved, thereby improving the heat treatment effect on the magnetic material. At the same time, the present invention can mine more hidden features and deeper features by fusing various feature vectors, and then predict the heat treatment method based on the hidden features and deeper features, which can further improve the determination accuracy of the heat treatment method, thereby further improving the heat treatment effect on the magnetic material, so that the magnetic material after heat treatment can better meet the performance requirements.

[0114] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for heat treatment of magnetic materials, characterized in that: include: In response to a heat treatment signal of a target magnetic material, obtaining material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material; Respectively determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information; The material feature vector, the demand feature vector, and the heat treatment furnace feature vector are fused to obtain a heat treatment fusion feature vector; Inputting the heat treatment fusion feature vector into a preset treatment method prediction model to predict the heat treatment method, and obtaining a target heat treatment method corresponding to the target magnetic material; Based on the target heat treatment method, the target heat treatment furnace is controlled to perform heat treatment on the target magnetic material to obtain a heat-treated magnetic material.

2. The method according to claim 1, characterized in that The target heat treatment method includes: a heating treatment method, a heat preservation treatment method and a cooling treatment method. The heating treatment method includes: a heating method, a heating time and a heating temperature. The heat preservation treatment method includes: a heat preservation method and a heat preservation time. The cooling treatment method includes: a cooling method and a cooling speed. Based on the target heat treatment method, controlling the target heat treatment furnace to heat treat the target magnetic material to obtain the heat-treated magnetic material includes: Based on the heating method, heating time, and heating temperature, controlling the target heat treatment furnace to heat the target magnetic material to obtain the magnetic material after the heating treatment; Based on the insulation method and insulation time, controlling the target heat treatment furnace to perform insulation treatment on the magnetic material after the heating treatment to obtain the magnetic material after the insulation treatment; Based on the cooling method and cooling speed, the target heat treatment furnace is controlled to cool the magnetic material after the insulation treatment to obtain the magnetic material after the heat treatment.

3. The method according to claim 1, characterized in that The method further comprises: During the heat treatment of the target magnetic material, heating property information of the target heat treatment furnace is obtained, wherein the heating property information includes at least one of real-time temperature, real-time vacuum degree, and real-time pressure in the target heat treatment furnace; Based on the heating property information, determining whether the target heat treatment furnace meets the preset conditions for the alarm; If the target heat treatment furnace meets the preset conditions, an alarm message is generated based on the heating property information, and a preset communication tool interface is called to send the alarm message to a monitoring terminal.

4. The method according to claim 1, characterized in that: The fusing process of the material feature vector, the demand feature vector, and the heat treatment furnace feature vector to obtain the heat treatment fusion feature vector includes: Determine the principal component eigenvector between every two eigenvectors among the material eigenvector, the demand eigenvector, and the heat treatment furnace eigenvector, including: determining a first principal component eigenvector between the material eigenvector and the demand eigenvector, a second principal component eigenvector between the material eigenvector and the heat treatment furnace eigenvector, and a third principal component eigenvector between the demand eigenvector and the heat treatment furnace eigenvector; Respectively horizontally splicing the first principal component eigenvector, the second principal component eigenvector, and the third principal component eigenvector to obtain a spliced ​​eigenvector; The transformation coefficient corresponding to the concatenated feature vector is determined, and based on the transformation coefficient, the concatenated feature vector is linearly transformed to obtain a heat-processed fused feature vector.

5. The method according to claim 4, characterized in that The determining of the principal component feature vector between every two feature vectors in the material feature vector, the demand feature vector, and the heat treatment furnace feature vector comprises: Based on every two eigenvectors, construct a feature matrix; Determine the element mean of each element in the feature matrix, and subtract each element in the feature matrix from the element mean to obtain a centralized feature matrix; Determine the covariance matrix corresponding to the centralized feature matrix; Performing eigenvalue decomposition on the covariance matrix to obtain matrix eigenvalues ​​and matrix eigenvectors; Based on the magnitude of the matrix eigenvalues, a preset number of matrix eigenvectors are selected from the matrix eigenvectors, and the preset number of matrix eigenvectors are determined as principal component eigenvectors between every two eigenvectors.

6. The method according to claim 1, characterized in that Before obtaining the material property information of the target magnetic material, the heat treatment requirement information, and the heat treatment furnace property information of the target heat treatment furnace for heat treating the target magnetic material, the method further includes: Determine the storage location information corresponding to the target magnetic material, determine different idle heat treatment furnaces in the heat treatment site where the target magnetic material is located, and obtain furnace attribute information of each of the idle heat treatment furnaces; Determine the attribute feature vector corresponding to the furnace attribute information, and based on the attribute feature vector, perform clustering processing on each of the idle heat treatment furnaces to obtain idle heat treatment furnaces under different clustering categories; Based on the material property information and the heat treatment requirement information, determining a target cluster category that meets the heat treatment requirement from among the different cluster categories; The furnace location information of each idle heat treatment furnace under the target clustering category is determined, and based on the storage location information of the target magnetic material and the furnace location information of each idle heat treatment furnace, the target heat treatment furnace is determined among the idle heat treatment furnaces under the target clustering category.

7. The method according to claim 6, characterized in that The method of clustering the idle heat treatment furnaces based on the attribute feature vector to obtain idle heat treatment furnaces in different clustering categories includes: Initialize the centroid vectors corresponding to different clusters; Calculating the cosine similarity between each of the attribute feature vectors and the centroid vectors corresponding to the different clusters, and dividing each of the idle heat treatment furnaces into the different clusters based on the cosine similarities; Based on the attribute feature vectors corresponding to the idle heat treatment furnaces in the different clusters, determining the updated centroid vectors corresponding to the different clusters; Based on the updated centroid vector, each of the idle heat treatment furnaces is re-divided into the different clusters until the updated centroid vector does not change, and the idle heat treatment furnaces finally divided into the different clusters are determined as idle heat treatment furnaces under the different clustering categories.

8. A heat treatment device for magnetic materials, characterized in that: include: An acquisition unit, configured to acquire material property information of the target magnetic material, heat treatment requirement information, and heat treatment furnace property information of a target heat treatment furnace for heat treating the target magnetic material in response to a heat treatment signal of the target magnetic material; A determination unit, used to respectively determine a material feature vector corresponding to the material attribute information, a requirement feature vector corresponding to the heat treatment requirement information, and a heat treatment furnace feature vector corresponding to the heat treatment furnace attribute information; A fusion unit, used for fusing the material feature vector, the demand feature vector, and the heat treatment furnace feature vector to obtain a heat treatment fusion feature vector; A prediction unit, used for inputting the heat treatment fusion feature vector into a preset treatment method prediction model to predict the heat treatment method, and obtaining a target heat treatment method corresponding to the target magnetic material; The heat treatment unit is used to control the target heat treatment furnace to perform heat treatment on the target magnetic material based on the target heat treatment method to obtain the magnetic material after heat treatment.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images