A method and system for precise control of magnetic steel forming based on magnetic field orientation

Abstract

The application relates to the technical field of artificial intelligence regulation, and relates to a magnetic steel forming precision control method and system based on magnetic field orientation, which comprises the following steps: performing a pressing operation on magnetic powder particle raw materials based on external magnetic field configuration, and in the process: obtaining a pressing intermediate, obtaining a real-time pressure parameter, performing an edge-to-center magnetic field strength adjustment operation on a default magnetic field strength according to the real-time pressure parameter to obtain a target magnetic field matrix; judging whether the real-time pressure parameter is less than a pre-constructed process pressure in the magnetic field orientation forming process parameter; when the real-time pressure parameter is greater than or equal to the process pressure, the pressing operation based on the external magnetic field configuration is configured, the process pressure and the target magnetic field matrix are maintained until a preset maintenance time, and a pressing finished product is obtained; and performing a processing forming operation on the pressing finished product according to a preset sintering and post-processing strategy to obtain a target magnetic steel product. The application can improve the precision of the magnetic field orientation setting process.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence control technology, and in particular to a method and system for precise control of magnetic steel forming based on magnetic field orientation. Background Technology

[0002] The current process of forming magnets generally includes material selection, smelting, magnetic field orientation and shaping, heat treatment, and coating. Among them, magnetic field orientation and shaping refers to the process of applying a uniform external magnetic field to make the magnetic powder inside the magnet align in an orderly manner along the direction of the magnetic field, while simultaneously pressing the magnetic powder.

[0003] However, the requirements for magnet products are becoming increasingly stringent, and the product sizes are getting larger, leading to larger and larger errors in the traditional magnetic field orientation and shaping process. This results in problems such as "low magnet edge orientation" and "large orientation error," making it impossible for traditional magnet production processes to meet the hardware requirements of high-precision equipment. Summary of the Invention

[0004] This invention provides a precise control method for magnetic steel forming based on magnetic field orientation, the main purpose of which is to improve the accuracy of the magnetic field orientation forming process.

[0005] To achieve the above objectives, the present invention provides a method for precise control of magnetic steel forming based on magnetic field orientation, comprising:

[0006] Obtain magnetic powder particle raw materials based on pre-constructed order information;

[0007] According to preset magnetic field orientation forming process parameters, the magnetic powder particle raw material is subjected to a pressing operation based on an external magnetic field configuration, and during the pressing operation based on the external magnetic field configuration:

[0008] Obtain the pressing intermediate, obtain the real-time pressure parameters, and use the pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation molding process parameters based on the real-time pressure parameters, thereby obtaining the target magnetic field matrix.

[0009] Determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters;

[0010] When the real-time pressure parameter is less than the process pressure, the default magnetic field strength is updated using the target magnetic field matrix, and the process returns to the steps described above for obtaining the pressing intermediate.

[0011] When the real-time pressure parameter is greater than or equal to the process pressure, the pressing operation based on the external magnetic field configuration is configured to maintain the process pressure and the target magnetic field matrix until the preset holding time, thereby obtaining the pressed finished product.

[0012] According to the preset sintering and post-processing strategy, the pressed finished product is processed and shaped to obtain the target magnetic steel product.

[0013] Optionally, obtaining the magnetic powder particle raw material based on the pre-constructed order information includes:

[0014] Text keyword recognition is performed on the pre-built order information to obtain the magnet type, magnet size and density requirements;

[0015] If the magnet type is a preset first type of magnet, then the size and density requirements of the magnet are multiplied to obtain the raw material requirement.

[0016] According to the required amount of raw materials, obtain magnetic powder mixed raw materials, and perform melting and cooling operations on the magnetic powder mixed raw materials to obtain alloy strip sheets;

[0017] Using a pre-constructed powder-making device, the alloy strip is crushed into powder of a preset particle size to obtain magnetic powder particle raw material;

[0018] If the magnet type is a preset second type of magnet, then a set of sub-component particles is obtained, and the set of sub-component particles is uniformly mixed to obtain magnetic powder particle raw material.

[0019] Optionally, the step of using a pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation forming process parameters based on the real-time pressure parameters, according to the edge-to-center magnetic field strength adjustment operation, to obtain the target magnetic field matrix, includes:

[0020] Using a pre-built edge-magnetic field configuration network in a pre-trained magnetic field configuration model, the real-time size of the compression intermediate is predicted based on the real-time pressure parameters.

[0021] Based on the real-time dimensions of the intermediate body, a magnetic field configuration operation based on edge orientation degree is performed on the default magnetic field strength to obtain the magnetic field strength gradient distribution, wherein the edge-magnetic field configuration network is represented as:

[0022] ,

[0023] In the formula, This represents the magnetic field strength gradient distribution. This represents a regression function relationship for predicting the distribution of the internal magnetic field strength of a magnet based on the external magnetic field and the magnet's dimensions. This indicates the real-time size of the intermediate. This indicates the default magnetic field strength. This represents the relationship coefficient between the real-time dimensions of the intermediate and the real-time pressure parameter. This indicates the real-time pressure parameter. Indicates bias;

[0024] Using the pre-constructed pressure-magnetic field configuration network in the magnetic field configuration model, the real-time internal pressure distribution of the compression intermediate is predicted based on the real-time pressure parameters and the real-time size of the intermediate.

[0025] Based on the real-time internal pressure distribution, a compensating magnetic field prediction operation based on the pressure attenuation of the magnetic field strength gradient distribution is performed on the compression intermediate to obtain the compensating magnetic field strength distribution, wherein the pressure-magnetic field configuration network is represented as:

[0026] ,

[0027] In the formula, This indicates compensation for the distribution of magnetic field strength. This represents the weighting coefficient between the compensating magnetic field strength distribution and the real-time internal pressure distribution. This represents the real-time internal pressure distribution. This represents a regression function relationship that predicts the internal pressure distribution of a magnet based on its surface pressure and size.

[0028] Using the edge-magnetic field configuration network, the compensation magnetic field intensity distribution is calculated in reverse to obtain the external compensation magnetic field intensity distribution. The external compensation magnetic field intensity distribution and the default magnetic field intensity are then summed to obtain the target magnetic field matrix.

[0029] Optionally, before utilizing the pre-built edge-magnetic field configuration network in the pre-trained magnetic field configuration model, the method further includes:

[0030] Obtain a pressure and size relationship sample based on the magnetic powder particle raw material, and use a pre-built regression network model to perform machine learning on the pressure and size relationship sample to obtain a compression size prediction network;

[0031] Obtain a set of magnetic field orientation experimental parameters based on a combination of random external magnetic field strength and random billet size, and obtain the magnetic steel orientation degree corresponding to each magnetic field orientation experimental parameter in the set of magnetic field orientation experimental parameters to obtain an orientation degree set;

[0032] A training sample set is obtained by constructing key-value pairs from the set of magnetic field orientation experimental parameters and the set of orientation degrees.

[0033] The regression network model is trained using the training sample set to obtain an orientation recognition network based on magnetic field and size. The orientation recognition network based on magnetic field and size is minimized to obtain a magnetic field recognition network with the minimum orientation.

[0034] The suppression size prediction network is connected to the magnetic field recognition network to obtain the edge-magnetic field configuration network.

[0035] Optionally, before utilizing the pre-built pressure-magnetic field configuration network in the magnetic field configuration model, the method further includes:

[0036] Using the pre-constructed finite element method, the pre-constructed target steel billet is discretized into a set of element steel billets;

[0037] Based on preset extrusion parameters, the extrusion process of the unit steel billet assembly is simulated to obtain the three-dimensional pressure distribution of the target steel billet;

[0038] The three-dimensional pressure distribution is weighted according to preset weighting coefficients to obtain the magnetic field utilization rate distribution;

[0039] Obtain the billet size parameters of the target steel billet and the surface pressure parameters in the extrusion parameters, and construct a pressure-magnetic field distribution sample based on the billet size parameters, surface pressure parameters and magnetic field utilization rate distribution;

[0040] Based on the pressure-magnetic field distribution samples, the regression network model is trained to obtain the pressure-magnetic field utilization recognition network;

[0041] The output layer of the pressure-magnetic field utilization identification network is obtained, and the output layer is configured using a pre-constructed activation function to obtain the pressure-magnetic field configuration network.

[0042] Optionally, the method further includes maintaining the process pressure and the target magnetic field matrix until a preset holding time:

[0043] Determine the shape of the compressed intermediate;

[0044] When the shape of the pressing intermediate is a preset rectangle or column, the pressing intermediate is classified by size according to the preset size rules;

[0045] If the shape of the pressed intermediate is a preset small-sized rectangle or column, the holding time is configured as a preset first holding time;

[0046] If the shape of the pressed intermediate is a preset large-sized rectangle or column, the holding time is configured as a preset second holding time;

[0047] When the shape of the pressing intermediate is a preset irregular shape, the holding time is configured as a preset third holding time.

[0048] Optionally, the step of processing the pressed finished product into a shape according to a preset sintering and post-processing strategy to obtain the target magnetic steel product includes:

[0049] According to the preset sintering and post-processing strategy, the pressed finished product is sintered to obtain a sintered product, and the sintered product is quenched and tempered to obtain a heat-treated magnet.

[0050] According to the product specifications pre-constructed in the sintering and post-processing strategy, the heat-treated magnet is machined to obtain a shaped product, and the shaped product is then subjected to surface electroplating to obtain a corrosion-resistant product.

[0051] The corrosion-resistant product is magnetized using a pre-built magnetization device to obtain the target magnetic steel product.

[0052] Optionally, after processing the pressed product into a shape according to a preset sintering and post-processing strategy to obtain the target magnetic steel product, the method further includes:

[0053] Obtain the target magnetic field matrix corresponding to each suppression intermediate in the suppression operation based on the external magnetic field configuration, and obtain the magnetic field matrix change sequence;

[0054] Using a pre-constructed finite element simulation model, the magnetic field matrix change sequence is simulated to obtain the simulated control signal;

[0055] The pre-constructed electromagnetic coil array is configured using the analog control signal.

[0056] Optionally, after obtaining the target magnet product, the method further includes:

[0057] The target magnet product is subjected to orientation degree detection to obtain the true orientation degree;

[0058] Determine whether the actual orientation degree is less than a preset qualified threshold;

[0059] When the true orientation degree is less than the qualified threshold, the target magnet product is determined to be unqualified.

[0060] To achieve the above objectives, the present invention also provides a precision control system for magnetic steel forming based on magnetic field orientation, comprising:

[0061] The raw material acquisition module is used to acquire magnetic powder particle raw materials based on pre-built order information;

[0062] The magnetic field orientation and shaping module is used to perform a pressing operation on the magnetic powder particle raw material based on the external magnetic field configuration according to the preset magnetic field orientation molding process parameters. During the pressing operation based on the external magnetic field configuration, the module obtains the pressing intermediate, acquires the real-time pressure parameters, and uses a pre-trained magnetic field configuration model to adjust the default magnetic field strength pre-constructed in the magnetic field orientation molding process parameters based on the real-time pressure parameters to obtain the target magnetic field matrix.

[0063] The parameter adjustment module is used to determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters, and when the real-time pressure parameter is less than the process pressure, update the default magnetic field strength using the target magnetic field matrix and return to the above steps of obtaining the pressing intermediate, and when the real-time pressure parameter is greater than or equal to the process pressure, configure the pressing operation based on the external magnetic field configuration, maintain the process pressure and the target magnetic field matrix until the preset holding time, and obtain the pressed finished product;

[0064] The post-processing module is used to process and shape the pressed finished product according to a preset sintering and post-processing strategy to obtain the target magnetic steel product.

[0065] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0066] Memory, storing at least one instruction;

[0067] The processor executes the instructions stored in the memory to implement the above-described method for precise control of magnetic steel forming based on magnetic field orientation.

[0068] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the above-described method for precise control of magnetic steel forming based on magnetic field orientation.

[0069] To address the problems described in the background section, this invention first performs a pressing operation on magnetic powder particles based on an external magnetic field configuration. During the pressing process, real-time pressure parameters of the pressing intermediate are continuously acquired. Then, based on a pre-trained magnetic field configuration model, a suitable magnetic field matrix is ​​pre-processed. This allows the magnetic field to change in response to pressure variations during pressing, ensuring uniform orientation of the magnetic powder particles within the pressing intermediate, thereby reducing the degree of orientation and improving the precision of the magnet product. The magnetic field configuration model in this application predicts the size and internal pressure distribution of the pressing intermediate based on real-time pressure. Then, using a regression neural network, it learns the mapping relationship between magnet size and the distribution of magnetic field orientation capability within the magnet, and the mapping relationship between the internal pressure distribution and the magnetic field orientation efficiency, thus outputting a target magnetic field matrix. This target magnetic field matrix eliminates the interference of size distribution (center to surface) and pressure distribution on magnetic field orientation. Therefore, this invention improves the accuracy of the magnetic field orientation process. Attached Figure Description

[0070] Figure 1 A flowchart illustrating a precise control method for forming magnetic steel based on magnetic field orientation provided in an embodiment of the present invention;

[0071] Figure 2 A functional block diagram of a magnetic steel forming precision control system based on magnetic field orientation provided in an embodiment of the present invention;

[0072] Figure 3 This is a schematic diagram of the structure of an electronic device that implements the precise control method for forming magnetic steel based on magnetic field orientation, according to an embodiment of the present invention.

[0073] Explanation of reference numerals in the attached figures:

[0074] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0075] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0076] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0077] This application provides a method for precise control of magnet forming based on magnetic field orientation. The executing entity of this method includes, but is not limited to, at least one electronic device configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, where the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0078] Reference Figure 1 The diagram shown is a flowchart illustrating a method for precise control of magnet forming based on magnetic field orientation, according to an embodiment of the present invention. In this embodiment, the method for precise control of magnet forming based on magnetic field orientation includes:

[0079] S1. Obtain magnetic powder particle raw materials based on the pre-constructed order information.

[0080] The order information refers to the collection of text information of customer orders received by the magnet manufacturer, such as information on magnet shape, type, and size.

[0081] The magnetic powder raw material refers to particles made by crushing alloy or pure metal materials to a certain degree, such as Nd2Fe with a particle size of 1-5μm in sintered NdFeB magnets. 14 B.

[0082] In detail, in this embodiment of the invention, obtaining magnetic powder particle raw materials based on pre-constructed order information includes:

[0083] Text keyword recognition is performed on the pre-built order information to obtain the magnet type, magnet size and density requirements;

[0084] If the magnet type is a preset first type of magnet, then the size and density requirements of the magnet are multiplied to obtain the raw material requirement.

[0085] According to the required amount of raw materials, obtain magnetic powder mixed raw materials, and perform melting and cooling operations on the magnetic powder mixed raw materials to obtain alloy strip sheets;

[0086] Using a pre-constructed powder-making device, the alloy strip is crushed into powder of a preset particle size to obtain magnetic powder particle raw material;

[0087] If the magnet type is a preset second type of magnet, then a set of sub-component particles is obtained, and the set of sub-component particles is uniformly mixed to obtain magnetic powder particle raw material.

[0088] The text keyword recognition refers to the process of extracting information such as magnet type and size from the order information.

[0089] The types of magnets include sintered NdFeB, bonded NdFeB, permanent magnet ferrite, and samarium cobalt magnets.

[0090] The size of the magnet is measured in mm, for example, a rectangular magnet measuring 250mm × 250mm × 50mm.

[0091] The density requirement refers to a parameter in the final product, such as a customer requiring a magnet density of 7.5 g / cm³. 3 .

[0092] The first type of magnet refers to a type of magnet that is made by first mixing and melting various raw materials and then crushing them into particles, such as sintered neodymium iron boron.

[0093] The raw material requirement refers to the product of the required size and density of the magnet.

[0094] The magnetic powder mixture refers to the mixture of various raw materials that have been proportioned.

[0095] The melting and cooling operation refers to the process of heating the magnetic powder mixture into an alloy and then cooling the alloy.

[0096] The alloy strip refers to an alloy sheet prepared by rapid solidification technology.

[0097] The powder-making equipment refers to a device that combines hydrogen crushing and air jet milling technology, used to crush alloy strips into powder.

[0098] The preset particle size is configured to be 1 to 5 micrometers.

[0099] The second type of magnet refers to a type of magnet in which each raw material is individually ground into powder and then mixed in proportion to form magnetic powder particles, such as permanent magnet ferrite.

[0100] The sub-component particle set refers to the powder of each raw material corresponding to the second type of magnet.

[0101] The uniform mixing refers to the process of uniformly distributing the particles of each sub-component, which requires the participation of a stirrer.

[0102] Specifically, in this embodiment of the invention, the corresponding product information, such as magnet type, magnet size, and density requirements, is first obtained based on the order information. Then, the raw material quality is weighed according to the magnet size and density requirements. However, depending on different production processes, the raw materials need to be processed separately. If the magnet type is a first type of magnet, multiple raw materials are mixed and processed through melting, solidification, crushing, etc., to obtain magnetic powder particles. If the magnet type is a second type of magnet, each sub-component needs to be powdered separately and then mixed evenly to obtain magnetic powder particles.

[0103] S2. According to the preset magnetic field orientation forming process parameters, the magnetic powder particle raw material is pressed based on the external magnetic field configuration, and during the pressing operation based on the external magnetic field configuration.

[0104] The magnetic field orientation and shaping process parameters refer to the executable parameters configured by the enterprise in the magnetic steel production line, such as a uniform magnetic field of 1.5T (10,000 gauss) and a shaping pressure of 300MPa.

[0105] The pressing operation based on the external magnetic field configuration refers to the process of extruding magnetic powder particles using a 300MPa pressure mold while maintaining a stable magnetic field of 1.5T.

[0106] Specifically, in this embodiment of the invention, the magnetic powder particles are first oriented and shaped using default magnetic field and pressure parameters. Then, the changes in various data are monitored as the pressure mold is gradually lowered and the pressure gradually increases.

[0107] S3. Obtain the pressing intermediate, obtain the real-time pressure parameters, and use the pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation molding process parameters based on the real-time pressure parameters, thereby obtaining the target magnetic field matrix.

[0108] The compressed intermediate refers to an intermediate in which the magnetic powder particles are compressed, but not completely compressed.

[0109] The real-time pressure parameter refers to the pressure data displayed on the pressure mold, and also represents the pressure parameter of the surface of the pressed intermediate.

[0110] The magnetic field configuration model refers to a model built based on a regression neural network. The model learns the mapping relationship between the size of the compression intermediate and the distribution of the internal magnetic field orientation capability, as well as the mapping relationship between size, surface pressure and internal pressure distribution, and the mapping relationship between internal pressure distribution and internal magnetic field orientation capability distribution.

[0111] The default magnetic field strength refers to a uniform magnetic field of 1.5T.

[0112] The target magnetic field matrix refers to the default magnetic field strength after configuration by the magnetic field configuration model.

[0113] In detail, in this embodiment of the invention, the step of using a pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation forming process parameters based on the edge to the center, according to the real-time pressure parameters, to obtain the target magnetic field matrix, includes:

[0114] Using a pre-built edge-magnetic field configuration network in a pre-trained magnetic field configuration model, the real-time size of the compression intermediate is predicted based on the real-time pressure parameters.

[0115] Based on the real-time dimensions of the intermediate body, a magnetic field configuration operation based on edge orientation degree is performed on the default magnetic field strength to obtain the magnetic field strength gradient distribution, wherein the edge-magnetic field configuration network is represented as:

[0116] ,

[0117] In the formula, This represents the magnetic field strength gradient distribution. This represents a regression function relationship for predicting the distribution of the internal magnetic field strength of a magnet based on the external magnetic field and the magnet's dimensions. This indicates the real-time size of the intermediate. This indicates the default magnetic field strength. This represents the relationship coefficient between the real-time dimensions of the intermediate and the real-time pressure parameter. This indicates the real-time pressure parameter. Indicates bias;

[0118] Using the pre-constructed pressure-magnetic field configuration network in the magnetic field configuration model, the real-time internal pressure distribution of the compression intermediate is predicted based on the real-time pressure parameters and the real-time size of the intermediate.

[0119] Based on the real-time internal pressure distribution, a compensating magnetic field prediction operation based on the pressure attenuation of the magnetic field strength gradient distribution is performed on the compression intermediate to obtain the compensating magnetic field strength distribution, wherein the pressure-magnetic field configuration network is represented as:

[0120] ,

[0121] In the formula, This indicates compensation for the distribution of magnetic field strength. This represents the weighting coefficient between the compensating magnetic field strength distribution and the real-time internal pressure distribution. This represents the real-time internal pressure distribution. This represents a regression function relationship that predicts the internal pressure distribution of a magnet based on its surface pressure and size.

[0122] Using the edge-magnetic field configuration network, the compensation magnetic field intensity distribution is calculated in reverse to obtain the external compensation magnetic field intensity distribution. The external compensation magnetic field intensity distribution and the default magnetic field intensity are then summed to obtain the target magnetic field matrix.

[0123] The edge-magnetic field configuration network refers to a model built based on a regression neural network, which is used to learn the mapping relationship between external pressure and the volume change of the pressing intermediate, as well as the mapping relationship between the volume change of the pressing intermediate and the magnetic field orientation capability distribution of each magnetic powder in the pressing intermediate.

[0124] In this application, the magnetic field orientation capability refers to the ability of an external magnetic field to control the rotation of the magnetic domains of individual magnetic powders in the pressing intermediate. For example, when the external magnetic field is constant, the faster the magnetic domains of the magnetic powder rotate, the higher the magnetic field orientation capability. When the magnetic domains of the magnetic powder rotate at a uniform speed, the weaker the external magnetic field, the higher the magnetic field orientation capability.

[0125] The real-time size of the intermediate refers to the size of the pressed intermediate.

[0126] The magnetic field configuration operation based on edge orientation adjustment refers to the process of adjusting the magnetic field strength at various locations along the cross-section of the pressed intermediate body, from the edge to the center of the cross-section, perpendicular to the direction of the external magnetic field. The magnetic field strength gradient distribution refers to the default magnetic field strength modified by the magnetic field configuration operation based on edge orientation adjustment.

[0127] The pressure-magnetic field configuration network refers to a regression network model used to learn the mapping relationship between size and surface pressure, and between internal pressure distribution, as well as the mapping relationship between pressure distribution and magnetic field orientation capability distribution.

[0128] The real-time internal pressure distribution refers to the pressure distribution information of the pressing intermediate from the surface to the center.

[0129] The pressure attenuation-based magnetic field prediction operation for the pressed intermediate refers to the following process: Based on the theory that higher pressure leads to denser magnetic powder arrangement, higher magnetic permeability, and ultimately stronger magnetic field orientation capability, and the theory that the surface pressure of the pressed intermediate is the highest while the internal pressure gradually decreases during pressing, it can be inferred that the magnetic field orientation capability of the magnetic powder in different parts of the pressed intermediate is different. Therefore, the lack of internal magnetic field orientation capability in various regions can be compensated by increasing the external magnetic field, thereby obtaining the compensation magnetic field intensity distribution. For example, in this embodiment of the invention, the pressure distribution of the pressed intermediate (such as a magnetic powder compact) usually conforms to the gradient attenuation law. Assuming that the pressure decreases exponentially from the surface to the interior (in actual processes, it is only close to exponential attenuation due to pressure loss caused by internal particle friction), it can be expressed as:

[0130] ,

[0131] In the formula, express Deep pressure, Indicates surface pressure (i.e.) ), Indicates surface to The distance between the depths, This represents the pressure attenuation coefficient (an empirical constant determined by material properties and pressing speed, which needs to be calibrated experimentally).

[0132] Furthermore, it should be known that the magnetic powder packing density increases monotonically with increasing pressure (pressure compresses the gaps between particles, causing the density to rise), and the relative permeability is positively correlated with the magnetic powder density (the higher the density, the stronger the magnetic coupling between particles, and the stronger the ability to conduct magnetic fields). The magnetic field orientation capability is essentially the degree to which the magnetic powder particles align along the "easy magnetization axis" of the external magnetic field direction; the higher the permeability, the easier it is for the particles to respond to the external magnetic field and align in a specific direction. Therefore, the magnetic field orientation capability is directly proportional to the permeability.

[0133] Furthermore, it can be inferred that the distribution of magnetic field orientation capability inside the magnet changes with the pressure distribution. Therefore, during the pressing process, if the external magnetic field is uniform but the internal pressure distribution of the magnet is uneven, the distribution of magnetic field orientation capability inside the magnet will be uneven. Thus, the distribution of the external magnetic field can be adjusted and changed by regressing the network mapping relationship, thereby making the distribution of magnetic field orientation capability inside the magnet uniform.

[0134] The compensation magnetic field intensity distribution refers to the prediction result of the real-time internal pressure distribution by the compensation magnetic field prediction operation based on pressure attenuation.

[0135] The network reverse calculation process refers to the process of configuring the edge-magnetic field network. As input parameters, The process of calculating as an output result. The external compensating magnetic field strength distribution refers to incorporating the compensating magnetic field strength distribution into... When parameters are included, the output is The output value of the parameter.

[0136] The matrix summation refers to the process of summing the values ​​of the external compensating magnetic field strength distribution and the default magnetic field strength at the same location.

[0137] Specifically, in this embodiment of the invention, a pre-trained edge-magnetic field configuration network is used to predict the real-time size of the intermediate pressing body based on real-time pressure parameters. Then, based on the real-time size of the intermediate pressing body, the conversion efficiency between the external magnetic field and the orientation capability of the internal magnetic field of the intermediate pressing body is predicted. Based on the conversion efficiency, the default external magnetic field strength is adjusted once to obtain the magnetic field strength gradient distribution, so that the magnetic field orientation capability in the intermediate pressing body is uniform and is not affected by the magnetic powder at the center or edge position.

[0138] Specifically, this invention uses a pressure-magnetic field configuration network to predict the real-time internal pressure distribution of the pressing intermediate based on real-time pressure parameters and the real-time dimensions of the intermediate. Then, it calculates the external magnetic field that needs to be compensated based on the real-time internal pressure distribution, obtaining a compensation magnetic field strength distribution. This ensures that the magnetic field orientation capability within the pressing intermediate is uniform and unaffected by pressure at the magnetic powder locations. Furthermore, through the inverse calculation of the edge-magnetic field configuration network, the compensation magnetic field strength distribution is output as an external compensation magnetic field strength distribution.

[0139] Finally, by summing the external compensating magnetic field strength distribution and the default magnetic field strength, a target magnetic field matrix that takes into account both positional and pressure disturbances is obtained.

[0140] In detail, in this embodiment of the invention, before utilizing the pre-built edge-magnetic field configuration network in the pre-trained magnetic field configuration model, the method further includes:

[0141] Obtain a pressure and size relationship sample based on the magnetic powder particle raw material, and use a pre-built regression network model to perform machine learning on the pressure and size relationship sample to obtain a compression size prediction network;

[0142] Obtain a set of magnetic field orientation experimental parameters based on a combination of random external magnetic field strength and random billet size, and obtain the magnetic steel orientation degree corresponding to each magnetic field orientation experimental parameter in the set of magnetic field orientation experimental parameters to obtain an orientation degree set;

[0143] A training sample set is obtained by constructing key-value pairs from the set of magnetic field orientation experimental parameters and the set of orientation degrees.

[0144] The regression network model is trained using the training sample set to obtain an orientation recognition network based on magnetic field and size. The orientation recognition network based on magnetic field and size is minimized to obtain a magnetic field recognition network with the minimum orientation.

[0145] The suppression size prediction network is connected to the magnetic field recognition network to obtain the edge-magnetic field configuration network.

[0146] The pressure and size relationship sample based on the magnetic powder particle raw material refers to the data recording the relationship between the surface pressure and size of the pressing intermediate under the condition that the mass fraction of the magnetic powder particle raw material remains unchanged.

[0147] The regression network model refers to a regression function with automatic parameter updating capabilities, used to learn the mapping relationships between multiple variables.

[0148] The machine learning and subsequent training processes described herein are all processes of updating the parameters in the network model based on the cross-entropy loss algorithm and the gradient descent method, which will not be elaborated upon in this application.

[0149] The compression size prediction network refers to a regression network model that has learned the relationship between the surface pressure and size of the compression intermediate.

[0150] The set of magnetic field orientation experimental parameters refers to the data records of magnetic field orientation experiments conducted by randomly configuring the external magnetic field strength and billet size.

[0151] The orientation degree of the magnets refers to the angle between the C-axis of each magnet and the external magnetic field. The C-axis refers to the direction of easy magnetization in the magnet structure.

[0152] The orientation set refers to the orientation detection results corresponding to each magnetic field orientation experiment.

[0153] The key-value pair construction refers to the data storage process of using the set of magnetic field orientation experimental parameters and the set of orientation degrees as "keys" and "values," respectively. Key-value pairs are a data storage method. The training sample set refers to the set of key-value pairs constructed from the magnetic field orientation experimental parameters and orientation degrees.

[0154] The orientation recognition network based on magnetic field and size refers to the result of training a regression network model through a training sample set.

[0155] The magnetic field recognition network refers to a network that minimizes the orientation degree recognition network based on magnetic field and size. For example, in the orientation degree recognition network based on magnetic field and size, when the size is fixed and the orientation degree is at its minimum value, the corresponding magnetic field will be output.

[0156] The process of connecting the compression size prediction network to the magnetic field recognition network refers to the process of inputting the output of the compression size prediction network into the magnetic field recognition network.

[0157] Specifically, in this embodiment of the invention, a regression network model is first trained using pressure and size relationship samples to obtain a suppression size prediction network. Then, an orientation degree set is extracted using a magnetic field orientation experimental parameter set, and a training sample set is constructed to train the regression network model, resulting in an orientation degree recognition network based on magnetic field and size. This orientation degree recognition network is then minimized to obtain a magnetic field recognition network with the lowest possible orientation degree. Finally, the suppression size prediction network is connected to the orientation degree recognition network based on magnetic field and size to obtain an edge-magnetic field configuration network.

[0158] Furthermore, it should be understood that the uniformity and strength of the magnetic field inside the magnet are directly affected by its shape and size. Therefore, in another embodiment of the present invention, a parameter describing the weakening of the internal magnetic field of the magnet due to its shape can be set, namely the demagnetization factor, the value of which depends on the aspect ratio of the magnet (the ratio of the length in the magnetization direction to the dimension in the perpendicular direction).

[0159] This invention takes an infinitely long magnet (assuming a demagnetization factor of 0) as an example. The internal magnetic field strength is equal to the remanence of the magnet (the remanent magnetic induction intensity of the magnet after the external magnetic field is removed, which is an inherent property of the material, such as the remanence of N50 neodymium iron boron which is 1.43T). However, considering demagnetization, the internal magnetic field can be assumed to be:

[0160] ,

[0161] In the formula, Indicates the internal magnetic field. Indicates remanence. This represents the demagnetization factor.

[0162] Depending on the shape, if the magnet is a cuboid, it can be predicted that when both the length and width are much larger than the thickness in the magnetization direction, the internal magnetic field should be close to 0. Therefore, the pre-configured demagnetization factor of this invention is: In the formula, the magnetization direction is the thickness D, and the other two directions are the length L and width W. Similarly, if the shape is cylindrical, the magnetization direction is the height D (equivalent to the thickness D mentioned above), and the radius is R. The demagnetization factor that can be configured in this invention is... This allows for a demagnetization factor of approximately 0 when R is much larger than D.

[0163] In another embodiment of the present invention, the shape features of each sample can be extracted and demagnetization factors can be configured during the model training process, thereby reducing the training difficulty of network parameters in the regression model and improving the recognition accuracy of the edge-magnetic field configuration network.

[0164] In detail, in this embodiment of the invention, before utilizing the pre-constructed pressure-magnetic field configuration network in the magnetic field configuration model, the method further includes:

[0165] Using the pre-constructed finite element method, the pre-constructed target steel billet is discretized into a set of element steel billets;

[0166] Based on preset extrusion parameters, the extrusion process of the unit steel billet assembly is simulated to obtain the three-dimensional pressure distribution of the target steel billet;

[0167] The three-dimensional pressure distribution is weighted according to preset weighting coefficients to obtain the magnetic field utilization rate distribution;

[0168] Obtain the billet size parameters of the target steel billet and the surface pressure parameters in the extrusion parameters, and construct a pressure-magnetic field distribution sample based on the billet size parameters, surface pressure parameters and magnetic field utilization rate distribution;

[0169] Based on the pressure-magnetic field distribution samples, the regression network model is trained to obtain the pressure-magnetic field utilization recognition network;

[0170] The output layer of the pressure-magnetic field utilization identification network is obtained, and the output layer is configured using a pre-constructed activation function to obtain the pressure-magnetic field configuration network.

[0171] The finite element method refers to the method of discretizing a continuous steel billet into a finite number of smaller steel billets.

[0172] The target billet refers to a digital model of a pressing intermediate.

[0173] The term "unit billet set" refers to the entirety of multiple unit billets.

[0174] The extrusion parameter can be a pressure value that is randomly generated within the range of 100MPa to 300MPa.

[0175] The simulation of the extrusion process of the unit billet assembly refers to the virtual extrusion process achieved through digital simulation of multiphysics fields.

[0176] The three-dimensional pressure distribution refers to the set of pressure vectors corresponding to the unit steel billet at each location.

[0177] The weighting coefficient refers to the coefficient that transforms the order of magnitude of each data point in the three-dimensional pressure distribution into a percentage.

[0178] The magnetic field utilization rate distribution refers to the specific numerical value of the magnetic field orientation capability mentioned above. It can be understood as: "Three-dimensional pressure distribution, proportional to the magnetic permeability distribution, and proportional to the magnetic field utilization rate distribution."

[0179] The billet size parameters refer to the size information of the target steel billet.

[0180] The surface pressure parameter refers to the extrusion parameter corresponding to the target steel billet during this extrusion process.

[0181] The pressure-magnetic field distribution sample refers to a key-value pair constructed using the embryo size parameter and surface pressure parameter as the key and the magnetic field utilization rate distribution as the value.

[0182] The pressure-magnetic field utilization identification network refers to a regression network model trained using pressure-magnetic field distribution samples.

[0183] The output layer refers to the network layer that changes the output result.

[0184] The activation function refers to the function that modifies the output result. In this invention, it is configured as an inverse proportional function, which means that the higher the utilization rate of the identified magnetic field, the lower the compensation strength of the output magnetic field.

[0185] Specifically, in this embodiment of the invention, a target steel billet is discretized using the finite element method to obtain a finite set of unit billets. Then, the set of unit billets is used to simulate a compression scenario to obtain the pressure distribution of each unit billet, thus obtaining the three-dimensional pressure distribution of the target steel billet. Furthermore, using weighting coefficients, the three-dimensional pressure distribution is transformed into a magnetic field utilization rate distribution. Then, based on the billet size parameters, surface pressure parameters, and magnetic field utilization rate distribution, a pressure-magnetic field distribution sample is constructed. Finally, a regression network model is trained using the pressure-magnetic field distribution sample to obtain a pressure-magnetic field utilization rate recognition network.

[0186] S4. Determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters.

[0187] In this embodiment of the invention, the pressure mold gradually presses down on the pressing intermediate, causing the real-time pressure parameters to continuously increase. When the pressure reaches the process pressure, the pressure can be maintained and no further increase is required.

[0188] When the real-time pressure parameter is less than the process pressure, S5, update the default magnetic field strength using the target magnetic field matrix, and return to the above steps of obtaining the pressing intermediate.

[0189] When the real-time pressure parameter is less than the process pressure, it indicates that the pressure mold is still extruding. Therefore, the default magnetic field strength can be replaced according to the target magnetic field matrix obtained above, thereby eliminating the magnetic field orientation interference caused by the changes in size and pressure distribution of the pressing intermediate during the pressing process.

[0190] When the real-time pressure parameter is greater than or equal to the process pressure, S6, configure the pressing operation based on the external magnetic field configuration, maintain the process pressure and the target magnetic field matrix until the preset holding time, and obtain the pressed finished product.

[0191] When the real-time pressure parameter is greater than or equal to the process pressure, it indicates that the process parameters have been met. Based on production experience, the process pressure and the target magnetic field matrix are maintained.

[0192] The holding time is 5 seconds to 1 minute.

[0193] The pressed finished product refers to a pressed intermediate that maintains the process pressure and the target magnetic field matrix for a preset holding time.

[0194] In detail, in this embodiment of the invention, the method of maintaining the process pressure and the target magnetic field matrix until the preset holding time further includes:

[0195] Determine the shape of the compressed intermediate;

[0196] When the shape of the pressing intermediate is a preset rectangle or column, the pressing intermediate is classified by size according to the preset size rules;

[0197] If the shape of the pressed intermediate is a preset small-sized rectangle or column, the holding time is configured to be a preset first holding time;

[0198] If the shape of the pressed intermediate is a preset large-sized rectangle or column, the holding time is configured as a preset second holding time;

[0199] When the shape of the pressing intermediate is a preset irregular shape, the holding time is configured as a preset third holding time.

[0200] The shapes include rectangles, columns, steps, grooves, etc.

[0201] The size rules are configured as follows: magnets with a diameter less than 20mm and a thickness less than 10mm are considered small-sized, while magnets with a diameter greater than 50mm and a thickness greater than 30mm are considered large-sized.

[0202] The size classification refers to the process of dividing the pressed intermediates according to the size rules.

[0203] The first holding time is configured to be 5 to 10 seconds. The second holding time is configured to be 15 to 30 seconds. The third holding time is 30 seconds to 1 minute, as a longer holding time is required to adjust the orientation due to uneven magnetic powder accumulation in complex structures.

[0204] The irregular shape can be stepped, grooved, or similar.

[0205] Specifically, in this embodiment of the invention, the holding time of the compression intermediate can be configured according to the above rules to adapt to different types of compression intermediates.

[0206] S7. According to the preset sintering and post-processing strategy, the pressed finished product is processed and shaped to obtain the target magnetic steel product.

[0207] The sintering and post-processing strategy refers to the production rules for magnets after the magnetic field orientation forming process.

[0208] The processing and forming operation refers to the process of transforming the pressed finished product into a marketable commodity according to the sintering and post-processing strategy.

[0209] The target magnet product is the magnet generated in the final result of this application.

[0210] In detail, in this embodiment of the invention, the step of processing and shaping the pressed finished product according to a preset sintering and post-processing strategy to obtain the target magnetic steel product includes:

[0211] According to the preset sintering and post-processing strategy, the pressed finished product is sintered to obtain a sintered product, and the sintered product is quenched and tempered to obtain a heat-treated magnet.

[0212] According to the product specifications pre-constructed in the sintering and post-processing strategy, the heat-treated magnet is machined to obtain a shaped product, and the shaped product is then subjected to surface electroplating to obtain a corrosion-resistant product.

[0213] The corrosion-resistant product is magnetized using a pre-built magnetization device to obtain the target magnetic steel product.

[0214] The sintering process refers to raising the temperature to 1050℃ to densify the magnetic powder. The sintered product refers to a pressed finished product that has undergone the sintering process.

[0215] The quenching and tempering operation refers to the process of cooling the sintered product to room temperature (20°C) and then heating it to 500°C. The heat-treated magnet refers to the sintered product after quenching and tempering.

[0216] The product specifications refer to the shape and size into which the product needs to be cut.

[0217] The formed product refers to heat-treated magnets that have undergone machine cutting.

[0218] The surface electroplating process involves covering the surface of the molded product with electroplating technology to prevent oxidation.

[0219] The magnetization equipment refers to devices such as pulse magnetizers that can provide a magnetic field. The magnetization process refers to the working process of the magnetization equipment.

[0220] Specifically, in this embodiment of the invention, the pressed finished product is sintered according to a preset sintering and post-processing strategy to obtain a sintered product. Then, quenching and tempering operations are performed to obtain a heat-treated magnet. Next, according to product specifications, the heat-treated magnet is machined to obtain a shaped product, and surface electroplating is completed to obtain a corrosion-resistant product. Finally, the corrosion-resistant product is magnetized using a magnetization device to obtain the target magnet product.

[0221] In detail, in this embodiment of the invention, after processing the pressed finished product into a shape according to a preset sintering and post-processing strategy to obtain the target magnetic steel product, the method further includes:

[0222] Obtain the target magnetic field matrix corresponding to each suppression intermediate in the suppression operation based on the external magnetic field configuration, and obtain the magnetic field matrix change sequence;

[0223] Using a pre-constructed finite element simulation model, the magnetic field matrix change sequence is simulated to obtain the simulated control signal;

[0224] The pre-constructed electromagnetic coil array is configured using the analog control signal.

[0225] The magnetic field matrix change sequence refers to the target magnetic field matrix corresponding to each suppression intermediate in the suppression operation based on the external magnetic field configuration.

[0226] The finite element simulation model refers to a model used for simulating multiphysics.

[0227] The simulation refers to the process of simulating a sequence of changes in the magnetic field matrix to obtain the corresponding circuit signals. The simulated control signal refers to an electrical signal capable of realizing the sequence of changes in the magnetic field matrix.

[0228] The electromagnetic coil array refers to a cluster of multiple magnetic field generating devices.

[0229] Specifically, in this embodiment of the invention, after the process of manufacturing the magnet, the sequence of changes in the magnetic field matrix can be recorded. Then, through a finite element simulation model, a simulated control signal capable of realizing the sequence of changes in the magnetic field matrix can be simulated. Thus, using the simulated control signal, the electromagnetic coil array can be directly configured, and the magnetic field configuration can be automated, without the need for calculations as described in the scheme of this invention.

[0230] In detail, in this embodiment of the invention, after obtaining the target magnet product, the method further includes:

[0231] The target magnet product is subjected to orientation degree detection to obtain the true orientation degree;

[0232] Determine whether the actual orientation degree is less than a preset qualified threshold;

[0233] When the true orientation degree is less than the qualified threshold, the target magnet product is determined to be unqualified.

[0234] The orientation degree detection refers to the process of identifying the orientation degree of the target magnetic steel product. The true orientation degree refers to the orientation degree detection result.

[0235] The qualified threshold is configured as 95%.

[0236] Specifically, in this embodiment of the invention, if the actual orientation degree is less than 95%, the target magnet product can be determined to be unqualified, which also indicates that the above magnetic field strength configuration process is incorrect and the parameters of the magnetic field configuration model need to be optimized.

[0237] To address the problems described in the background section, this invention first performs a pressing operation on magnetic powder particles based on an external magnetic field configuration. During the pressing process, real-time pressure parameters of the pressing intermediate are continuously acquired. Then, based on a pre-trained magnetic field configuration model, a suitable magnetic field matrix is ​​pre-processed. This allows the magnetic field to change in response to pressure variations during pressing, ensuring uniform orientation of the magnetic powder particles within the pressing intermediate, thereby reducing the degree of orientation and improving the precision of the magnet product. The magnetic field configuration model in this application predicts the size and internal pressure distribution of the pressing intermediate based on real-time pressure. Then, using a regression neural network, it learns the mapping relationship between magnet size and the distribution of magnetic field orientation capability within the magnet, and the mapping relationship between the internal pressure distribution and the magnetic field orientation efficiency, thus outputting a target magnetic field matrix. This target magnetic field matrix eliminates the interference of size distribution (center to surface) and pressure distribution on magnetic field orientation. Therefore, this invention improves the accuracy of the magnetic field orientation process.

[0238] like Figure 2 The diagram shown is a functional block diagram of a magnetic steel forming precision control system based on magnetic field orientation provided in an embodiment of the present invention.

[0239] The magnetic field orientation-based precision control system 100 for forming magnetic steel described in this invention can be installed in an electronic device. Depending on the functions implemented, the magnetic field orientation-based precision control system 100 may include a raw material acquisition module 101, a magnetic field orientation and shaping module 102, a parameter adjustment module 103, and a post-processing module 104. The module described in this invention can also be called a unit, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0240] The raw material acquisition module 101 is used to acquire magnetic powder particle raw materials according to pre-constructed order information;

[0241] The magnetic field orientation shaping module 102 is used to perform a pressing operation on the magnetic powder particle raw material based on the external magnetic field configuration according to the preset magnetic field orientation forming process parameters. During the pressing operation based on the external magnetic field configuration, the module obtains a pressing intermediate, obtains real-time pressure parameters, and uses a pre-trained magnetic field configuration model to adjust the default magnetic field strength pre-constructed in the magnetic field orientation forming process parameters based on the real-time pressure parameters to obtain the target magnetic field matrix.

[0242] The parameter adjustment module 103 is used to determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters, and when the real-time pressure parameter is less than the process pressure, update the default magnetic field strength using the target magnetic field matrix and return to the above steps of obtaining the pressing intermediate, and when the real-time pressure parameter is greater than or equal to the process pressure, configure the pressing operation based on the external magnetic field configuration, maintain the process pressure and the target magnetic field matrix until the preset holding time, and obtain the pressed finished product;

[0243] The post-processing module 104 is used to process and shape the pressed finished product according to a preset sintering and post-processing strategy to obtain the target magnetic steel product.

[0244] In detail, the modules in the magnetic field orientation-based precision control system 100 for forming magnetic steel described in this embodiment of the invention employ the same methods as described above. Figure 1 The method used is the same as the precise control method for magnetic steel forming based on magnetic field orientation described in the article, and can produce the same technical effect, so it will not be repeated here.

[0245] like Figure 3 The diagram shown is a structural schematic of an electronic device that implements a precise control method for forming magnetic steel based on magnetic field orientation, according to an embodiment of the present invention.

[0246] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a program for a precise control method for forming magnetic steel based on magnetic field orientation.

[0247] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a method for precise control of magnetic steel forming based on magnetic field orientation, but also to temporarily store data that has been output or will be output.

[0248] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a precise control method program for magnet forming based on magnetic field orientation), and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0249] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0250] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0251] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0252] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0253] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0254] The program for precise control of magnetic steel forming based on magnetic field orientation, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0255] Obtain magnetic powder particle raw materials based on pre-constructed order information;

[0256] According to preset magnetic field orientation forming process parameters, the magnetic powder particle raw material is subjected to a pressing operation based on an external magnetic field configuration, and during the pressing operation based on the external magnetic field configuration:

[0257] Obtain the pressing intermediate, obtain the real-time pressure parameters, and use the pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation molding process parameters based on the real-time pressure parameters, thereby obtaining the target magnetic field matrix.

[0258] Determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters;

[0259] When the real-time pressure parameter is less than the process pressure, the default magnetic field strength is updated using the target magnetic field matrix, and the process returns to the steps described above for obtaining the pressing intermediate.

[0260] When the real-time pressure parameter is greater than or equal to the process pressure, the pressing operation based on the external magnetic field configuration is configured to maintain the process pressure and the target magnetic field matrix until the preset holding time, thereby obtaining the pressed finished product.

[0261] According to the preset sintering and post-processing strategy, the pressed finished product is processed and shaped to obtain the target magnetic steel product.

[0262] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0263] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0264] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0265] Obtain magnetic powder particle raw materials based on pre-constructed order information;

[0266] According to preset magnetic field orientation forming process parameters, the magnetic powder particle raw material is subjected to a pressing operation based on an external magnetic field configuration, and during the pressing operation based on the external magnetic field configuration:

[0267] Obtain the pressing intermediate, obtain the real-time pressure parameters, and use the pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation molding process parameters based on the real-time pressure parameters, thereby obtaining the target magnetic field matrix.

[0268] Determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters;

[0269] When the real-time pressure parameter is less than the process pressure, the default magnetic field strength is updated using the target magnetic field matrix, and the process returns to the steps described above for obtaining the pressing intermediate.

[0270] When the real-time pressure parameter is greater than or equal to the process pressure, the pressing operation based on the external magnetic field configuration is configured to maintain the process pressure and the target magnetic field matrix until the preset holding time, thereby obtaining the pressed finished product.

[0271] According to the preset sintering and post-processing strategy, the pressed finished product is processed and shaped to obtain the target magnetic steel product.

[0272] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0273] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0274] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0275] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0276] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for precise control of magnetic steel forming based on magnetic field orientation, characterized in that, The method includes: Obtain magnetic powder particle raw materials based on pre-constructed order information; According to preset magnetic field orientation forming process parameters, the magnetic powder particle raw material is subjected to a pressing operation based on an external magnetic field configuration, and during the pressing operation based on the external magnetic field configuration: Obtain the pressing intermediate, obtain the real-time pressure parameters, and use the pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation molding process parameters based on the real-time pressure parameters, thereby obtaining the target magnetic field matrix. Determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters; When the real-time pressure parameter is less than the process pressure, the default magnetic field strength is updated using the target magnetic field matrix, and the process returns to the steps described above for obtaining the pressing intermediate. When the real-time pressure parameter is greater than or equal to the process pressure, the pressing operation based on the external magnetic field configuration is configured to maintain the process pressure and the target magnetic field matrix until the preset holding time, thereby obtaining the pressed finished product. According to the preset sintering and post-processing strategy, the pressed finished product is processed and shaped to obtain the target magnetic steel product.

2. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 1, characterized in that, The step of obtaining magnetic powder particle raw materials based on pre-constructed order information includes: Text keyword recognition is performed on the pre-built order information to obtain the magnet type, magnet size and density requirements; If the magnet type is a preset first type of magnet, then the size and density requirements of the magnet are multiplied to obtain the raw material requirement. According to the required amount of raw materials, obtain magnetic powder mixed raw materials, and perform melting and cooling operations on the magnetic powder mixed raw materials to obtain alloy strip sheets; Using a pre-constructed powder-making device, the alloy strip is crushed into powder of a preset particle size to obtain magnetic powder particle raw material; If the magnet type is a preset second type of magnet, then a set of sub-component particles is obtained, and the set of sub-component particles is uniformly mixed to obtain magnetic powder particle raw material.

3. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 2, characterized in that, The method utilizes a pre-trained magnetic field configuration model to adjust the default magnetic field strength in the magnetic field orientation forming process parameters based on the real-time pressure parameters, performing an edge-to-center magnetic field strength adjustment operation to obtain the target magnetic field matrix, including: Using a pre-built edge-magnetic field configuration network in a pre-trained magnetic field configuration model, the real-time size of the compression intermediate is predicted based on the real-time pressure parameters. Based on the real-time dimensions of the intermediate body, a magnetic field configuration operation based on edge orientation degree is performed on the default magnetic field strength to obtain the magnetic field strength gradient distribution, wherein the edge-magnetic field configuration network is represented as: ， In the formula, This represents the magnetic field strength gradient distribution. This represents a regression function relationship for predicting the distribution of the internal magnetic field strength of a magnet based on the external magnetic field and the magnet's dimensions. This indicates the real-time size of the intermediate. This indicates the default magnetic field strength. This represents the relationship coefficient between the real-time dimensions of the intermediate and the real-time pressure parameter. This indicates the real-time pressure parameter. Indicates bias; Using the pre-constructed pressure-magnetic field configuration network in the magnetic field configuration model, the real-time internal pressure distribution of the compression intermediate is predicted based on the real-time pressure parameters and the real-time size of the intermediate. Based on the real-time internal pressure distribution, a compensating magnetic field prediction operation based on the pressure attenuation of the magnetic field strength gradient distribution is performed on the compression intermediate to obtain the compensating magnetic field strength distribution, wherein the pressure-magnetic field configuration network is represented as: ， In the formula, This indicates compensation for the distribution of magnetic field strength. This represents the weighting coefficient between the compensating magnetic field strength distribution and the real-time internal pressure distribution. This represents the real-time internal pressure distribution. This represents a regression function relationship that predicts the internal pressure distribution of a magnet based on its surface pressure and size. Using the edge-magnetic field configuration network, the compensation magnetic field intensity distribution is calculated in reverse to obtain the external compensation magnetic field intensity distribution. The external compensation magnetic field intensity distribution and the default magnetic field intensity are then summed to obtain the target magnetic field matrix.

4. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 3, characterized in that, Prior to utilizing the pre-built edge-magnetic field configuration network in the pre-trained magnetic field configuration model, the method further includes: Obtain a pressure and size relationship sample based on the magnetic powder particle raw material, and use a pre-built regression network model to perform machine learning on the pressure and size relationship sample to obtain a compression size prediction network; Obtain a set of magnetic field orientation experimental parameters based on a combination of random external magnetic field strength and random billet size, and obtain the magnetic steel orientation degree corresponding to each magnetic field orientation experimental parameter in the set of magnetic field orientation experimental parameters to obtain an orientation degree set; A training sample set is obtained by constructing key-value pairs from the set of magnetic field orientation experimental parameters and the set of orientation degrees. The regression network model is trained using the training sample set to obtain an orientation recognition network based on magnetic field and size. The orientation recognition network based on magnetic field and size is minimized to obtain a magnetic field recognition network with the minimum orientation. The suppression size prediction network is connected to the magnetic field recognition network to obtain the edge-magnetic field configuration network.

5. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 4, characterized in that, Prior to utilizing the pre-constructed pressure-magnetic field configuration network in the magnetic field configuration model, the method further includes: Using the pre-constructed finite element method, the pre-constructed target steel billet is discretized into a set of element steel billets; Based on preset extrusion parameters, the extrusion process of the unit steel billet assembly is simulated to obtain the three-dimensional pressure distribution of the target steel billet; The three-dimensional pressure distribution is weighted according to preset weighting coefficients to obtain the magnetic field utilization rate distribution; Obtain the billet size parameters of the target steel billet and the surface pressure parameters in the extrusion parameters, and construct a pressure-magnetic field distribution sample based on the billet size parameters, surface pressure parameters and magnetic field utilization rate distribution; Based on the pressure-magnetic field distribution samples, the regression network model is trained to obtain the pressure-magnetic field utilization recognition network; The output layer of the pressure-magnetic field utilization identification network is obtained, and the output layer is configured using a pre-constructed activation function to obtain the pressure-magnetic field configuration network.

6. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 5, characterized in that, The method further includes maintaining the process pressure and the target magnetic field matrix until a preset holding time, wherein: Determine the shape of the compressed intermediate; When the shape of the pressing intermediate is a preset rectangle or column, the pressing intermediate is classified by size according to the preset size rules; If the shape of the pressed intermediate is a preset small-sized rectangle or column, the holding time is configured to be a preset first holding time; If the shape of the pressed intermediate is a preset large-sized rectangle or column, the holding time is configured as a preset second holding time; When the shape of the pressing intermediate is a preset irregular shape, the holding time is configured as a preset third holding time.

7. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 6, characterized in that, The step of processing and shaping the pressed product according to a preset sintering and post-processing strategy to obtain the target magnetic steel product includes: According to the preset sintering and post-processing strategy, the pressed finished product is sintered to obtain a sintered product, and the sintered product is quenched and tempered to obtain a heat-treated magnet. According to the product specifications pre-constructed in the sintering and post-processing strategy, the heat-treated magnet is machined to obtain a shaped product, and the shaped product is then subjected to surface electroplating to obtain a corrosion-resistant product. The corrosion-resistant product is magnetized using a pre-built magnetization device to obtain the target magnetic steel product.

8. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 7, characterized in that, After processing the pressed finished product into the target magnetic steel product according to the preset sintering and post-processing strategy, the method further includes: Obtain the target magnetic field matrix corresponding to each suppression intermediate in the suppression operation based on the external magnetic field configuration, and obtain the magnetic field matrix change sequence; Using a pre-constructed finite element simulation model, the magnetic field matrix change sequence is simulated to obtain the simulated control signal; The pre-constructed electromagnetic coil array is configured using the analog control signal.

9. The method for precise control of magnetic steel forming based on magnetic field orientation as described in claim 8, characterized in that, After obtaining the target magnet product, the method further includes: The target magnet product is subjected to orientation degree detection to obtain the true orientation degree; Determine whether the actual orientation degree is less than a preset qualified threshold; When the true orientation degree is less than the qualified threshold, the target magnet product is determined to be unqualified.

10. A precision control system for forming magnetic steel based on magnetic field orientation, characterized in that, The system includes: The raw material acquisition module is used to acquire magnetic powder particle raw materials based on pre-built order information; The magnetic field orientation and shaping module is used to perform a pressing operation on the magnetic powder particle raw material based on the external magnetic field configuration according to the preset magnetic field orientation molding process parameters. During the pressing operation based on the external magnetic field configuration, the module obtains the pressing intermediate, acquires the real-time pressure parameters, and uses a pre-trained magnetic field configuration model to adjust the default magnetic field strength pre-constructed in the magnetic field orientation molding process parameters based on the real-time pressure parameters to obtain the target magnetic field matrix. The parameter adjustment module is used to determine whether the real-time pressure parameter is less than the pre-constructed process pressure in the magnetic field orientation molding process parameters, and when the real-time pressure parameter is less than the process pressure, update the default magnetic field strength using the target magnetic field matrix and return to the above steps of obtaining the pressing intermediate, and when the real-time pressure parameter is greater than or equal to the process pressure, configure the pressing operation based on the external magnetic field configuration, maintain the process pressure and the target magnetic field matrix until the preset holding time, and obtain the pressed finished product; The post-processing module is used to process and shape the pressed finished product according to a preset sintering and post-processing strategy to obtain the target magnetic steel product.

Images