Permanent magnet rotor magnetizing uniformity control method and device

Abstract

The invention discloses a method and a device for controlling magnetizing uniformity of a permanent magnet rotor, relates to the technical field of permanent magnet rotors, and aims to evaluate the risk of uniformity during magnetizing of the permanent magnet rotor in advance and provide effective early warning for magnetizing operation by analyzing temperature change data, shape data and surface defect data and constructing a magnetic field response capability coefficient. Based on monitoring of the internal magnetic domain data and the magnetic data of the permanent magnet rotor in the step S3, the magnetic field stability of the rotor can be dynamically analyzed in combination with the influence of an external magnetic field generated by a magnetizing source on magnetic domain alignment. The comprehensive magnetic performance stability index is generated by comprehensively considering the magnetic field response capability coefficient and the magnetizing state prediction model, and a scientific basis is provided for control and optimization in the subsequent magnetizing process. Intelligent control and optimization are carried out, grade scoring is carried out by setting a risk threshold value R and combining the comprehensive magnetic performance stability index, the magnetizing operation effect can be accurately evaluated, and corresponding control measures are taken under different grades.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet rotors, and particularly to a method and device for controlling the magnetization uniformity of a permanent magnet rotor. Background Art

[0002] Permanent magnet materials are widely used in many fields such as motors, sensors, transformers, etc. Especially in high-performance drive systems, permanent magnet materials have received more and more attention due to their high magnetic energy density and excellent magnetic stability. With the continuous development of industrial technology, the application of permanent magnet materials has gradually expanded from traditional equipment such as motors and generators to equipment with higher precision and efficiency. Among them, as a core component, the permanent magnet rotor has been widely used in fields such as wind power generation and brushless DC motors. In order to ensure the efficient operation and long-term stability of the permanent magnet rotor during operation, the magnetization uniformity has become a key factor in ensuring its magnetic performance. The magnetization uniformity of the permanent magnet rotor directly affects its magnetic field distribution, power density and reliability. Therefore, how to accurately control the magnetization uniformity has become an important issue in the design and manufacture of permanent magnet rotors.

[0003] In the analysis of the magnetization uniformity of permanent magnet rotors, the influence of factors such as shape regularity, material defects, and heat treatment process on the magnetization effect cannot be ignored. The shape regularity of the permanent magnet rotor is directly related to the magnetic field distribution during the magnetization process. The temperature change on the rotor surface and the control of the cooling process determine the behavior of the magnetic material during the magnetization process, while surface defects may lead to uneven local magnetic flux distribution. Therefore, in the analysis of magnetization uniformity, scientifically and reasonably monitoring these data has become the basis for controlling the magnetization quality. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a method and device for controlling the magnetization uniformity of a permanent magnet rotor, which solves the problems in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: A method for controlling the magnetization uniformity of a permanent magnet rotor includes the following steps. S1. Set fixed points at different positions of the permanent magnet rotor in advance to collect the shape regularity of the permanent magnet rotor between the fixed points, obtain relevant shape data, and during the manufacturing stage of the permanent magnet rotor, monitor the relevant temperature change data of each area on the surface of the permanent magnet rotor during the heat treatment and cooling processes, and before performing the magnetization operation on the permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor. S2. Based on the relevant temperature change data, relevant shape data, and relevant surface defect data, construct a magnetic field response ability coefficient Cxxs, and based on the magnetic field response ability coefficient Cxxs, preliminarily predict the uniformity of the permanent magnet rotor during the magnetization operation. S3. On the basis of step S2, monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and analyze the alignment of the magnetic domains in the permanent magnet rotor and the magnetic field stability of the permanent magnet rotor under the action of an external magnetic field according to the relevant statistical data and relevant magnetic data. Then, combine the magnetic field response ability coefficient Cxxs and the trained magnetization state prediction model to construct a comprehensive magnetic performance stability index Zwzs. S4. Preset a risk threshold R in advance, and combine the value of the comprehensive magnetic performance stability index Zwzs to grade the magnetization state of the permanent magnet rotor. Based on the grade score, take corresponding control measures for the permanent magnet rotor.

[0006] Preferably, the specific steps of S1 include: S11. On the edge of the cross-section of the permanent magnet rotor, set fixed points on average, and collect the shape regularity of the permanent magnet rotor among the fixed points to obtain relevant shape data. The relevant shape data includes the radius values between the fixed points, and extract the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor and the minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor. ; S12. During the manufacturing stage of the permanent magnet rotor, divide the surface of the permanent magnet rotor into several groups of regions on average to monitor the relevant temperature change data of each region on the surface of the permanent magnet rotor during the heat treatment and cooling processes. Among them, the relevant temperature change data includes the temperature difference of each region on the surface of the permanent magnet rotor and the cooling rate difference. ; S13. After the manufactured permanent magnet rotor and before the magnetization operation of the permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor. Among them, the relevant surface defect data includes the surface roughness of the permanent magnet rotor .

[0007] Preferably, the specific steps of S2 include: S21. According to the relevant shape data obtained in step S11, analyze the shape regularity degree between each fixed point in the permanent magnet rotor to obtain the flatness factor Bpyz between each fixed point. Specifically, obtain the flatness factor Bpyz between the corresponding fixed points through the following formula: ; In the formula, represents the minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor, represents the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor; S22. Analyze the temperature change differences in each area on the surface of the permanent magnet rotor based on the relevant temperature change data obtained in step S12, and after dimensionless processing, obtain the heat treatment factor Rcyz for each area. Specifically, the heat treatment factor Rcyz for the corresponding area is obtained through the following formula: ; In the formula, and are both weight values, represents the temperature difference, represents the difference in cooling rate. Among them, , and The specific values are set by the user according to the situation.

[0008] Preferably, the specific steps of S2 further include: S23. Based on the flatness factor Bpyz between each fixed point and the heat treatment factor Rcyz for each area, and combined with the relevant surface defect data of the permanent magnet rotor, preliminarily analyze the risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor, and after dimensionless processing, construct the magnetic field response ability coefficient Cxxs. The magnetic field response ability coefficient Cxxs is obtained through the following formula: ; In the formula, represents the flatness factor between the th fixed points, represents the average flatness factor, represents the heat treatment factor for the th area, represents the average heat treatment factor represents the surface roughness, represents the number of fixed points, represents the number of areas, , , and are all weight values. Among them, , and The specific values are set by the user according to the situation.

[0009] Preferably, the specific steps of S2 further include: S24. By comparing the magnetic field response ability coefficient Cxxs with a preset threshold value, preliminarily predict the uniformity of the permanent magnet rotor during the magnetizing operation. The specific content is as follows: When the magnetic field response ability coefficient Cxxs does not exceed the preset threshold, it is preliminarily analyzed that there is a risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is unevenness in the permanent magnet rotor during the magnetization operation, triggering the risk mechanism; When the magnetic field response ability coefficient Cxxs exceeds the preset threshold, it is preliminarily analyzed that there is no risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is no unevenness in the permanent magnet rotor during the magnetization operation, and the risk mechanism is not triggered temporarily.

[0010] Preferably, the specific steps of S3 include: S31. According to the risk mechanism triggered in step S24, monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor. Among them, the relevant statistical data includes the number of magnetic domains per unit volume and the volume of the permanent magnet rotor ; the relevant magnetic data includes the magnetic moment of a single magnetic domain and the magnetic induction intensity ; S32. According to the relevant statistical data and relevant magnetic data, analyze the alignment of magnetic domains in the permanent magnet rotor under the action of an external magnetic field and the energy storage ability of the permanent magnet rotor in the external magnetic field, so as to calculate and obtain the magnetization intensity Chq and the magnetic energy density Cnmd: ; In the formula, represents the number of magnetic domains per unit volume; represents the magnetic moment of a single magnetic domain; represents the volume of the permanent magnet rotor.

[0011] Preferably, the specific steps of S3 also include: S33. The magnetic energy density Cnmd is obtained through the following formula: ; In the formula, represents the magnetic induction intensity, and Cxxs represents the magnetic field response ability coefficient.

[0012] Preferably, the specific steps of S3 also include: S34. Based on the magnetization intensity Chq and the magnetic energy density Cnmd obtained in steps S32 and S33, and combined with deep learning technology, construct a magnetization state prediction model and dimensionless processing, and fit and output the comprehensive magnetic performance stability index Zwzs. The comprehensive magnetic performance stability index Zwzs is obtained through the following formula: ; In the formula, and All are weight values, represents the logarithmic function with a constant as the base, represents the Euler number; represents a correction constant, where , and The specific values are set by the user according to the situation.

[0013] Preferably, the specific steps of S4 include: S41. By comparing the comprehensive magnetic performance stability index Zwzs with the risk threshold R, the magnetization state of the permanent magnet rotor is rated. The specific content is as follows: If the comprehensive magnetic performance stability index Zwzs exceeds the risk threshold R, a first-level rating is generated. At this time, the current magnetization process setting is maintained, and the magnetic field response and magnetic energy density are continuously monitored to ensure the stability of the magnetization process, and periodic monitoring is carried out to track the change of the comprehensive magnetic performance stability index Zwzs; If the comprehensive magnetic performance stability index Zwzs does not exceed the risk threshold R, a second-level rating is generated. At this time, according to the shape rule of the permanent magnet rotor, the magnetization time is increased and the magnetization current is adjusted, and the warning system is started to analyze various indicators, identify potential failure modes, and adaptively adjust the selection and layout of the permanent magnet material according to the analysis results of the magnetic energy density Cnmd and the magnetization intensity Chq to optimize the magnetic permeability and magnetization uniformity.

[0014] A device for controlling the magnetization uniformity of a permanent magnet rotor includes a data acquisition module, a preliminary analysis module, a stability comprehensive analysis module, and a level feedback module; The data acquisition module is used to pre-set fixed points at different positions of the permanent magnet rotor to collect the shape rule of the permanent magnet rotor between the fixed points, obtain relevant shape data, and monitor the relevant temperature change data of each area on the surface of the permanent magnet rotor during the heat treatment and cooling processes during the manufacturing stage of the permanent magnet rotor, and monitor the relevant surface defect data of the permanent magnet rotor before the magnetization operation of the permanent magnet rotor; The preliminary analysis module is used to construct a magnetic field response ability coefficient Cxxs based on the relevant temperature change data, relevant shape data, and relevant surface defect data, and preliminarily predict the uniformity of the permanent magnet rotor during the magnetization operation based on the magnetic field response ability coefficient Cxxs; The stability comprehensive analysis module is used to monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and analyze the alignment of the magnetic domains in the permanent magnet rotor and the magnetic field stability of the permanent magnet rotor under the action of an external magnetic field according to the relevant statistical data and relevant magnetic data, and combine the magnetic field response ability coefficient Cxxs and the trained magnetization state prediction model to construct a comprehensive magnetic performance stability index Zwzs; The level feedback module is used to preset a risk threshold R, and combine the value of the comprehensive magnetic performance stability index Zwzs to perform a level scoring on the magnetization state of the permanent magnet rotor. Based on the level scoring, corresponding control measures are taken for the permanent magnet rotor.

[0015] The present invention provides a method and device for controlling the magnetization uniformity of a permanent magnet rotor, and has the following beneficial effects: (1) By setting fixed points at different positions of the permanent magnet rotor to collect data on shape regularity, heat treatment temperature change and surface defects, the state of the permanent magnet rotor can be comprehensively grasped, providing detailed input data for the subsequent magnetization process. The accurate acquisition of these data helps to identify potential defects or irregularities, thus avoiding uneven magnetic field distribution during the magnetization process. By analyzing the temperature change data, shape data and surface defect data, a magnetic field response ability coefficient Cxxs is constructed, which can pre-evaluate the uniformity risk during the magnetization of the permanent magnet rotor, providing an effective early warning for the magnetization operation, and further avoiding the decline in equipment performance or failure caused by uneven magnetic field. Magnetic domain alignment and magnetic field stability analysis: Based on the monitoring of the internal magnetic domain data and magnetic data of the permanent magnet rotor in step S3, combined with the influence of the external magnetic field generated by the magnetization source on the magnetic domain alignment, the magnetic field stability of the rotor can be dynamically analyzed. By comprehensively considering the magnetic field response ability coefficient and the magnetization state prediction model, a comprehensive magnetic performance stability index Zwzs is generated, providing a scientific basis for the control and optimization in the subsequent magnetization process. Intelligent control and optimization: By setting a risk threshold R and performing level scoring in combination with the comprehensive magnetic performance stability index Zwzs, the effect of the magnetization operation can be accurately evaluated, and corresponding control measures are taken at different levels. This intelligent control strategy can not only ensure the stability of the magnetization process, but also realize the dynamic adjustment of the magnetization process in actual production, improve production efficiency, and reduce quality problems caused by magnetic field non-uniformity. In summary, this method can effectively improve the magnetic field uniformity, magnetic field stability and controllability of the magnetic properties of the permanent magnet rotor, ensure an efficient and accurate magnetization process, and further improve the overall performance and service life of the permanent magnet rotor and its applied equipment.

[0016] (2) Preset fixed points on the cross-sectional edge of the permanent magnet rotor and collect shape data. By extracting the maximum radius and the minimum radius, the shape regularity degree of the permanent magnet rotor at different positions can be accurately reflected. By calculating the flatness factor Bpyz between the fixed points, the shape irregularity is further quantified, enabling more refined control of the magnetic field distribution during the magnetization process. This technology can effectively avoid the influence of shape irregularity on the magnetization uniformity. Optimization of the heat treatment and cooling processes: During the manufacturing stage of the permanent magnet rotor, by monitoring the temperature change data in each region during the heat treatment and cooling processes, the temperature difference and the cooling rate difference can be captured in real time. After dimensionless processing of these data, the heat treatment factor Rcyz is obtained, which can accurately reflect the magnetic change risk in different regions during the heat treatment process. By analyzing the heat treatment factor, the cooling process can be optimized to avoid inconsistent magnetic properties caused by uneven temperature, thereby improving the stability of the magnetization uniformity of the permanent magnet rotor. Surface defect monitoring: Before the magnetization operation, real-time monitoring of the surface defects of the permanent magnet rotor, especially the collection of surface roughness data, can help analyze the potential negative impact on the magnetization uniformity due to surface irregularity. Surface defects directly affect the magnetic field distribution on the rotor surface. Timely monitoring and correction of surface defects can effectively improve the uniformity of the magnetization process and the magnetic field stability. Precise control and optimization: By combining the data analysis in steps S21 and S22, an accurate preliminary prediction and optimization plan can be provided for the magnetization process. The flatness factor Bpyz and the heat treatment factor Rcyz provide the shape regularity and the potential risks during the heat treatment process. These data are comprehensively utilized in the model, enabling the magnetization operation to be carried out more precisely and uniformly, further avoiding the efficiency loss and the equipment performance degradation caused by uneven magnetic fields in the traditional method.

[0017] (3) By combining the flatness factor Bpyz between each fixed point, the heat treatment factor Rcyz of each region, and the surface roughness data, a comprehensive analysis of the risk of non-uniform magnetic flux distribution caused by material defects in the permanent magnet rotor can be carried out. This step can identify in advance the potential magnetic problems caused by shape irregularity, uneven heat treatment, and surface defects, thereby providing a more accurate risk prediction for the magnetization process. By constructing the magnetic field response ability coefficient Cxxs, all influencing factors are comprehensively processed to obtain a dimensionless comprehensive index. This coefficient can intuitively reflect the risk of the magnetic flux distribution uniformity in the permanent magnet rotor caused by defects, providing a scientific basis for the control strategy during the magnetization process. Using this coefficient, a scientific evaluation and prediction of the magnetization uniformity of the rotor can be made before the magnetization operation, effectively improving the accuracy of the magnetization process.

[0018] (4) Real-time monitoring of magnetic domain alignment: By monitoring the relevant statistical data of magnetic domains in the permanent magnet rotor, including the number of magnetic domains per unit volume and the volume of the permanent magnet rotor, the distribution and number density of each magnetic domain in the permanent magnet rotor can be accurately obtained. This helps to judge the alignment state of magnetic domains under the external magnetic field, timely grasp the uniformity during the rotor magnetization process, and provide early warning for possible non-uniformity during the magnetization process. Precise calculation of magnetization intensity and energy storage capacity: By calculating the magnetization intensity Chq and magnetic energy density Cnmd, the magnetic performance of the permanent magnet rotor under the action of the external magnetic field can be comprehensively evaluated. These data reflect the degree of magnetic domain alignment and the energy storage situation, thus providing a basis for optimizing the magnetization process. Specifically, the magnetization intensity Chq can accurately reflect the sum of magnetic moments contributed by magnetic domains per unit volume, and the magnetic energy density Cnmd reflects the energy stored by the rotor in the external magnetic field, which helps to evaluate the magnetic stability of the rotor and ensure a uniform magnetic field distribution during the magnetization process. Improving the uniformity and efficiency of the magnetization process: By analyzing the statistical data of magnetic domains and magnetic data, combined with the action of the external magnetic field, dynamic monitoring and adjustment of the magnetization process of the permanent magnet rotor can be achieved. This method can timely detect areas where magnetic domains are not fully aligned during the magnetization process and take appropriate control measures for supplementation, making the magnetic field distribution during the magnetization process more uniform, thereby improving the magnetization efficiency and the magnetic performance stability of the permanent magnet rotor. Description of the Drawings

[0019] Figure 1 is a schematic flow chart of a method for controlling the magnetization uniformity of a permanent magnet rotor according to the present invention; Figure 2 is a block diagram of a device for controlling the magnetization uniformity of a permanent magnet rotor according to the present invention. Detailed Embodiments

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0021] Embodiment 1 Please refer to Figure 1 , the present invention provides a method for controlling the magnetization uniformity of a permanent magnet rotor, including the following steps, S1. Set fixed points at different positions of the permanent magnet rotor in advance to collect the shape regularity of the permanent magnet rotor between the fixed points, obtain relevant shape data, and during the manufacturing stage of the permanent magnet rotor, monitor the relevant temperature change data of each area on the surface of the permanent magnet rotor during the heat treatment and cooling processes, and before performing the magnetization operation on the permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor; S2. Based on the relevant temperature change data, relevant shape data, and relevant surface defect data, preliminarily analyze the risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor, construct the magnetic field response ability coefficient Cxxs, and based on the magnetic field response ability coefficient Cxxs, preliminarily predict the uniformity of the permanent magnet rotor during the magnetization operation; S3. On the basis of step S2, monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and according to the relevant statistical data and relevant magnetic data, analyze the alignment of the magnetic domains in the permanent magnet rotor and the magnetic field stability of the permanent magnet rotor under the action of an external magnetic field (the magnetic field generated by the magnetization source, and the magnetization source is a device used to apply an external magnetic field to the material, and it can generate a magnetic field through current, thereby affecting the alignment of the magnetic domains inside the material), and combine the magnetic field response ability coefficient Cxxs and the trained magnetization state prediction model to construct the comprehensive magnetic performance stability index Zwzs; S4. Set a risk threshold R in advance, and combine the value of the comprehensive magnetic performance stability index Zwzs to grade the magnetization state of the permanent magnet rotor, and based on the grade score, take corresponding control measures for the permanent magnet rotor.

[0022] In this embodiment, real-time monitoring and data feedback: By setting fixed points at different positions of the permanent magnet rotor, shape data, temperature change data, and surface defect data are collected in real time, which can accurately understand the possible defects, shape irregularities on the surface and inside of the rotor, and temperature fluctuations during the heat treatment process. These data provide a reliable basis for subsequent magnetization operations, ensuring the magnetic field uniformity during the magnetization operation. Risk prediction and prevention: Based on the temperature changes, shape data, and surface defect data obtained from the monitoring, by constructing a magnetic field response ability coefficient, it is possible to preliminarily analyze and predict the risk of uneven magnetic flux distribution that may occur during the magnetization of the permanent magnet rotor, thereby performing precise preliminary evaluation and risk control on the magnetization operation. This analysis can detect potential problems before magnetization, reducing performance losses and equipment failures caused by uneven magnetic fields. Magnetic domain alignment monitoring and optimization: During the magnetization process, the magnetic domain alignment situation and magnetic field stability inside the permanent magnet rotor are monitored in real time, and combined with the magnetic field response ability coefficient and the trained magnetization state prediction model, a comprehensive magnetic performance stability index is constructed. This index can comprehensively evaluate the magnetic domain alignment degree and magnetic field stability of the permanent magnet rotor under the action of an external magnetic field, providing data support for optimizing the magnetization process and improving the magnetic field uniformity. Magnetization state optimization and control: By setting a risk threshold and combining the comprehensive magnetic performance stability index to grade the magnetization state, targeted control measures can be taken according to the grading results. This can not only ensure the efficiency of the magnetization process but also timely adjust the magnetization strategy to further avoid abnormal situations during the magnetization process, thereby further improving the magnetization uniformity and the magnetic stability of the rotor. In summary, the control method of the present invention can significantly improve the magnetization uniformity of the permanent magnet rotor through the coordinated action of multiple links such as data acquisition, risk prediction, magnetic domain alignment analysis, and magnetization state control, effectively improving the performance stability and reliability of the permanent magnet rotor, and is widely applied to permanent magnet motors, power generation equipment, and other magnetic systems.

[0023] Embodiment 2 Please refer to Figure 1 , specifically: The specific steps of S1 include: S11. At the edges of the cross-section of the permanent magnet rotor, fixed points are evenly set, and the shape regularity of the permanent magnet rotor is collected between each fixed point to obtain relevant shape data. The relevant shape data includes the radius values between each fixed point, and the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor is extracted from it and the minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor ; S12. During the manufacturing stage of the permanent magnet rotor, the surface of the permanent magnet rotor is evenly divided into several groups of regions to monitor the relevant temperature change data of each region on the surface of the permanent magnet rotor during the heat treatment and cooling processes. Among them, the relevant temperature change data includes the temperature difference of each region on the surface of the permanent magnet rotor and the cooling rate difference ; S13. Before magnetizing the manufactured permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor. Among them, the relevant surface defect data includes the surface roughness of the permanent magnet rotor .

[0024] The specific steps of S2 include: S21. According to the relevant shape data obtained in step S11, analyze the shape regularity degree between each fixed point in the permanent magnet rotor to obtain the flatness factor Bpyz between each fixed point. Specifically, the flatness factor Bpyz between the corresponding fixed points is obtained through the following formula: ; In the formula, represents the minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor, represents the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor; The above-mentioned minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor and the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor can be monitored and obtained by using a laser scanning rangefinder and an optical displacement sensor; S22. According to the relevant temperature change data obtained in step S12, analyze the temperature change difference of each region on the surface of the permanent magnet rotor, and after dimensionless processing, obtain the heat treatment factor Rcyz of each region. Specifically, the heat treatment factor Rcyz of the corresponding region is obtained through the following formula: ; In the formula, and are both weight values, represents the temperature difference, represents the cooling rate difference. Among them, , and The specific values are set by the user according to the situation.

[0025] The above-mentioned temperature difference can be monitored and obtained by a temperature sensor; The cooling rate difference refers to the difference in the temperature change rate of different regions of the permanent magnet rotor during heat treatment or cooling. The cooling rate difference can reflect the non-uniformity during heat treatment and cooling, and further affect the magnetic distribution on the surface and inside of the permanent magnet rotor. It can be monitored and obtained by an infrared temperature sensor or a thermocouple sensor.

[0026] In this embodiment, precise shape monitoring and analysis: By evenly setting fixed points at the edge of the cross-section of the permanent magnet rotor and collecting shape data, the shape regularity of the permanent magnet rotor at different positions can be detailedly grasped. By extracting the maximum radius and minimum radius data and calculating the flatness factor Bpyz between the fixed points, the shape irregularity of the cross-section of the permanent magnet rotor can be analyzed from multiple aspects, thereby providing an accurate basis for the magnetic field distribution in the subsequent magnetization process. In this way, the risk of uneven magnetic field distribution caused by shape irregularity can be identified and controlled in advance. Optimization of heat treatment and cooling processes: During the manufacturing stage of the permanent magnet rotor, monitor the temperature change data in the heat treatment and cooling processes, and obtain the heat treatment factor Rcyz through dimensionless processing. By analyzing the temperature difference and cooling rate difference in each region on the surface of the permanent magnet rotor, the magnetic inhomogeneity that may be caused during the heat treatment process can be accurately identified. This method effectively reduces the magnetic field anomalies caused by heat treatment inhomogeneity and improves the magnetic performance stability of the overall permanent magnet rotor. Surface defect monitoring and correction: Before the magnetization operation, monitor the surface defect data of the permanent magnet rotor, especially the collection of surface roughness, which helps to timely detect defects that may affect the magnetic field uniformity. Through this monitoring, the potential impact of surface defects on the magnetization process can be pre-warned, so as to be corrected before magnetization to ensure a more uniform magnetic field distribution during the magnetization process and improve the service life and performance of the permanent magnet rotor. In summary, through scientific monitoring and analysis methods, the present invention predicts the uniformity of the permanent magnet rotor during magnetization from multiple dimensions, effectively improves the magnetic field stability of the permanent magnet rotor and the long-term reliability of the equipment, and has broad application prospects.

[0027] Embodiment 3 Please refer to Figure 1 , specifically: The specific steps of S2 further include: S23. According to the flatness factor Bpyz between each fixed point and the heat treatment factor Rcyz of each region, and in combination with the relevant surface defect data of the permanent magnet rotor, preliminarily analyze the risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor, and after dimensionless processing, construct a magnetic field response ability coefficient Cxxs. The magnetic field response ability coefficient Cxxs is obtained through the following formula: ; In the formula, represents the flatness factor between the th fixed points, represents the average flatness factor, represents the heat treatment factor of the th region, represents the average heat treatment factor represents the surface roughness, represents the number of fixed points between, represents the number of regions, , , and are all weight values. Among them, , and The specific values are set by the user according to the situation.

[0028] The above-mentioned surface roughness refers to the degree of microscopic irregularity on the surface of the permanent magnet rotor. When the surface roughness is large, it may cause uneven magnetic flux distribution, thus affecting the uniformity of the magnetizing process. Among them, a surface roughness gauge or a three-dimensional surface scanner (such as a laser scanner) can accurately measure the surface roughness of the permanent magnet rotor.

[0029] The specific steps of S2 also include: S24. By comparing the magnetic field response ability coefficient Cxxs with a preset threshold value, to preliminarily predict the uniformity of the permanent magnet rotor during the magnetizing operation. The specific content is as follows: If the magnetic field response ability coefficient Cxxs does not exceed the preset threshold value, it is preliminarily analyzed that there is a risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is unevenness in the permanent magnet rotor during the magnetizing operation, and the risk mechanism is triggered; If the magnetic field response ability coefficient Cxxs exceeds the preset threshold value, it is preliminarily analyzed that there is temporarily no risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is no unevenness in the permanent magnet rotor during the magnetizing operation, and the risk mechanism is not triggered temporarily.

[0030] In this embodiment, the risk of uneven magnetic flux distribution is accurately analyzed: By comprehensively considering the flatness factor Bpyz between each fixed point, the heat treatment factor Rcyz of each region, and the surface defect data of the permanent magnet rotor, the risk of uneven magnetic flux distribution caused by material defects in the permanent magnet rotor can be accurately evaluated. The dimensionless magnetic field response ability coefficient Cxxs, as a quantitative index, effectively reflects the potential problems of the magnetic uniformity of the permanent magnet rotor and provides a scientific basis for subsequent magnetization operation adjustment. Automatic risk prediction mechanism: According to the comparison between the constructed magnetic field response ability coefficient Cxxs and the preset threshold, the system can automatically judge the uniformity risk of the permanent magnet rotor during the magnetization operation. If Cxxs exceeds the set threshold, it indicates that the magnetic field distribution of the permanent magnet rotor during magnetization is basically uniform, and the magnetization operation can proceed smoothly; if Cxxs does not exceed the set threshold, the system can give a timely warning, identify the possible uniformity problems, and trigger the risk mechanism. This automatic warning mechanism further improves the safety and efficiency during the magnetization process. Intelligent control and real-time feedback: By calculating the magnetic field response ability coefficient Cxxs in real time and comparing it with the risk threshold, the control system can intelligently adjust the magnetization process to avoid magnetic field non-uniformity caused by factors such as material defects or irregular shapes. Combining with the real-time feedback risk assessment results, the system can optimize and adjust the permanent magnet rotor before magnetization to ensure the smooth progress of the magnetization process and effectively improve the final performance of the permanent magnet rotor. Optimize the accuracy and efficiency of the magnetization operation: By precisely controlling and comparing the magnetic field response ability coefficient Cxxs, the problems of energy waste and performance instability caused by non-uniformity during the magnetization process are avoided, thereby improving the efficiency and accuracy of the magnetization operation. Especially during the production process, it can predict risks based on the magnetization state and take necessary control measures in a timely manner to further ensure the stability and reliability of the permanent magnet rotor during final use. Effectively reduce production risks: Through real-time risk prediction and control of the magnetization process of the permanent magnet rotor, performance fluctuations and quality problems caused by uneven magnetization can be avoided, thereby reducing the defective product rate during the production process, improving the consistency and stability of product quality, and meeting the requirements of high-demand industrial applications.

[0031] Embodiment 4 Please refer to Figure 1 , specifically: The specific steps of S3 include: S31. According to the risk mechanism triggered in step S24, monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor. Among them, the relevant statistical data includes the number of magnetic domains per unit volume and the volume of the permanent magnet rotor ; the relevant magnetic data includes the magnetic moment of a single magnetic domain and the magnetic induction intensity ; S32. Analyze the alignment of magnetic domains in the permanent magnet rotor under the action of an external magnetic field and the energy storage capacity of the permanent magnet rotor in the external magnetic field based on relevant statistical data and relevant magnetic data, so as to calculate and obtain the magnetization intensity Chq and the magnetic energy density Cnmd: ; In the formula, represents the number of magnetic domains per unit volume; represents the magnetic moment of a single magnetic domain; represents the volume of the permanent magnet rotor; when an external magnetic field is applied, the magnetic domains start to align and the magnetization intensity will increase accordingly. If the magnetic domains are completely aligned, the magnetization intensity will reach the saturation state.

[0032] The number of magnetic domains per unit volume mentioned above refers to the number of magnetic domains in the permanent magnet rotor within a given volume range. Magnetic domains are regions with the same magnetization direction. The more the number, the finer the magnetic response of the material. Among them, a scanning electron microscope (SEM) or a high-resolution X-ray microscope can be used to observe and count the distribution of magnetic domains. In addition, combined with magnetic testing techniques, such as magnetization curve testing, the number of magnetic domains per unit volume can also be indirectly estimated.

[0033] The magnetic moment of a single magnetic domain refers to the total magnetic moment within each magnetic domain, which is usually related to the physical properties of the material (such as the magnetic permeability of the magnetic material, atomic arrangement, etc.) and factors such as electron spin. Among them, a vibrating sample magnetometer (VSM) or a Hall effect sensor can be used to measure the magnetic characteristics of the material, so as to deduce the magnetic moment of a single magnetic domain. In addition, magnetic resonance imaging (MRI) and magnetometers can also be used to indirectly evaluate the magnetic moment.

[0034] The volume of the permanent magnet rotor can be monitored and obtained through a three-dimensional laser scanner.

[0035] The magnetization intensity Chq reflects the sum of the magnetic moments contributed by all magnetic domains in each unit volume of the permanent magnet rotor. Each magnetic domain has a magnetic moment (depending on the characteristics of the material and the arrangement of the magnetic domains), and the magnetization intensity is the comprehensive manifestation of these magnetic moments. In other words, the sum of the magnetic moments of all magnetic domains per unit volume is the total magnetic moment of that unit volume, reflecting the magnetization degree of that part of the material.

[0036] In this embodiment, the alignment of magnetic domains is monitored in real time: By monitoring the number of magnetic domains per unit volume and the volume of the permanent magnet rotor, the magnetic domain distribution and quantity inside the permanent magnet rotor can be comprehensively understood, providing important basic data for further analyzing the alignment state of magnetic domains. This step can ensure the integrity of magnetic domain alignment during the magnetization process of the permanent magnet rotor, thereby improving the magnetic field uniformity and avoiding performance degradation caused by non-uniform magnetic flux. Calculate the magnetization intensity and magnetic energy density accurately: By precisely measuring the magnetic moment and magnetic induction intensity of each magnetic domain and combining with the volume of the permanent magnet rotor, the magnetization intensity Chq and magnetic energy density Cnmd can be calculated. These data provide a scientific basis for the magnetization process of the permanent magnet rotor and can accurately evaluate the magnetization degree and energy storage capacity of the rotor under the action of an external magnetic field. The magnetization intensity Chq reflects the sum of the magnetic moments contributed by all magnetic domains per unit volume, and the magnetic energy density Cnmd reflects the energy stored in the material under an external magnetic field. The combination of the two can provide a quantitative standard for the uniformity control during the magnetization operation. Optimize the magnetization process and improve the stability of magnetic properties: By analyzing the alignment of magnetic domains and the magnetization intensity in the permanent magnet rotor, potential magnetic field non-uniformity problems during the magnetization process can be identified in advance. The analysis results provide real-time feedback to the control system, ensuring that the permanent magnet rotor maintains stable magnetic properties during the magnetization operation and maximizing the magnetic energy density, thereby improving the working efficiency and service life of the permanent magnet rotor. Prevent magnetic non-uniformity problems in advance: Through comprehensive calculation and analysis of the magnetization intensity and magnetic energy density, this method can identify potential magnetic non-uniformity problems before the magnetization operation and make adjustments in advance. This step can not only reduce possible errors during the magnetization process but also improve the uniformity of the final magnetic field, avoid affecting the performance of the permanent magnet rotor due to magnetic field non-uniformity, and improve the quality stability of the overall product.

[0037] Embodiment 5 Please refer to Figure 1 , specifically: The specific steps of S3 also include: S33. The magnetic energy density Cnmd is obtained through the following formula: ; In the formula, represents the magnetic induction intensity, and Cxxs represents the magnetic field response ability coefficient, which reflects the magnetic permeability of the permanent magnet rotor.

[0038] The above-mentioned magnetic induction intensity is the magnetic flux density per unit area, expressed as the magnetic flux passing through per unit area, and can be monitored and obtained through a Hall effect sensor or a fluxmeter; The specific steps of S3 also include: S34. Based on the magnetization intensity Chq and magnetic energy density Cnmd obtained in steps S32 and S33, and combined with deep learning technology, construct a magnetization state prediction model and dimensionless processing, and fit and output the comprehensive magnetic performance stability index Zwzs. The comprehensive magnetic performance stability index Zwzs is obtained through the following formula: ; In the formula, and are both weight values. represents the logarithmic function with the constant as the base. represents the Euler number; represents a correction constant, where , and The specific values are set by the user according to the situation.

[0039] The specific steps of S4 include: S41. By comparing the comprehensive magnetic performance stability index Zwzs with the risk threshold R, grade the magnetization state of the permanent magnet rotor. The specific content is as follows: If the comprehensive magnetic performance stability index Zwzs exceeds the risk threshold R, generate a first-level score. At this time, keep the current magnetization process setting, continue to monitor the magnetic field response and magnetic energy density, ensure that the magnetization process remains stable, and perform periodic monitoring to track the change of the comprehensive magnetic performance stability index Zwzs to ensure that any potential problems can be discovered as early as possible; If the comprehensive magnetic performance stability index Zwzs does not exceed the risk threshold R, generate a second-level score. At this time, according to the shape rule of the permanent magnet rotor, increase the magnetization time and adjust the magnetization current to ensure the uniformity of the magnetization process, and start the warning system to analyze various indicators and identify potential failure modes, such as local hot spots, magnetic field non-uniformity and other problems. According to the analysis results of the magnetic energy density Cnmd and magnetization intensity Chq, adaptively adjust the selection and layout of the permanent magnet material to optimize the magnetic permeability and magnetization uniformity.

[0040] In this embodiment, the magnetic energy density and the magnetic field response ability coefficient are accurately calculated: by obtaining the magnetic induction intensity, the magnetic field response ability coefficient Cxxs, and the magnetic energy density Cnmd, the present invention can accurately analyze the magnetic permeability and the energy storage ability of the permanent magnet rotor under the action of an external magnetic field. The magnetic energy density Cnmd reflects the energy stored in the material during the magnetization process, while the magnetic field response ability coefficient Cxxs reflects the magnetic permeability of the permanent magnet rotor, thereby providing a quantitative index for the uniformity control during the magnetization process. These data help to judge the magnetic field response situation and provide a basis for the optimization of the subsequent magnetization process. The combination of deep learning and dimensionless processing: by combining deep learning technology with dimensionless processing, the comprehensive magnetic performance stability index Zwzs can accurately reflect the magnetization state stability of the permanent magnet rotor. This index effectively improves the decision-making accuracy during the magnetization process by fitting the magnetization intensity Chq and the magnetic energy density Cnmd data and comprehensively considering various influencing factors, such as magnetic permeability, magnetic domain alignment state, etc. By continuously optimizing and training the magnetization state prediction model, the real-time evaluation of the magnetic stability and uniformity of the permanent magnet rotor can be achieved during the magnetization process. Real-time monitoring and adjustment of the magnetization process: by comparing the comprehensive magnetic performance stability index Zwzs with the preset risk threshold R, the state of the permanent magnet rotor during the magnetization process can be monitored in real time. If Zwzs exceeds the risk threshold R, the magnetization process maintains the current setting, continuously monitors and tracks the index changes; if Zwzs does not exceed the threshold, the magnetization parameters are automatically adjusted, such as increasing the magnetization time or adjusting the magnetization current, to ensure the uniformity of the magnetization process, avoid local magnetic field non-uniformity or hot spot phenomena, thereby improving the magnetization uniformity and the magnetic stability of the permanent magnet rotor. Intelligent early warning and adaptive adjustment: through the risk threshold and the grade scoring mechanism, potential problems, such as fault modes of local hot spots or magnetic field non-uniformity, can be identified in real time during the magnetization process. In summary, the method for controlling the magnetization uniformity of the permanent magnet rotor provided by the present invention can not only optimize the magnetization process through accurate physical data and deep learning technology, but also effectively improve the magnetic stability of the permanent magnet rotor, reduce production risks, and improve production efficiency and the final performance of the permanent magnet rotor.

[0041] Embodiment 6 Please refer to Figure 2 , specifically: a device for controlling the magnetization uniformity of a permanent magnet rotor, including a data acquisition module, a preliminary analysis module, a stability comprehensive analysis module, and a grade feedback module; The data acquisition module is used to pre-set fixed points at different positions of the permanent magnet rotor to collect the shape regularity of the permanent magnet rotor between the fixed points, obtain relevant shape data, and during the manufacturing stage of the permanent magnet rotor, monitor the relevant temperature change data of each region on the surface of the permanent magnet rotor during the heat treatment and cooling processes, and before the magnetization operation of the permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor; The preliminary analysis module is used to construct the magnetic field response ability coefficient Cxxs based on relevant temperature change data, relevant shape data and relevant surface defect data, and preliminarily judge the uniformity of the permanent magnet rotor during the magnetizing operation based on the magnetic field response ability coefficient Cxxs; The stability comprehensive analysis module is used to monitor relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and analyze the alignment of the magnetic domains in the permanent magnet rotor and the magnetic field stability of the permanent magnet rotor under the action of an external magnetic field according to the relevant statistical data and relevant magnetic data, and combine the magnetic field response ability coefficient Cxxs and the trained magnetizing state prediction model to construct a comprehensive magnetic performance stability index Zwzs; The level feedback module is used to preset a risk threshold R, and combine the value of the comprehensive magnetic performance stability index Zwzs to grade the magnetizing state of the permanent magnet rotor, and take corresponding control measures for the permanent magnet rotor based on the grade score.

[0042] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling the magnetization uniformity of a permanent magnet rotor, characterized in that: including the following steps, S1. Set fixed points at different positions of the permanent magnet rotor in advance to collect the shape regularity of the permanent magnet rotor between the fixed points, obtain relevant shape data, and during the manufacturing stage of the permanent magnet rotor, monitor the relevant temperature change data of each area on the surface of the permanent magnet rotor during the heat treatment and cooling processes, and monitor the relevant surface defect data of the permanent magnet rotor before the magnetizing operation; S2. Based on the relevant temperature change data, relevant shape data, and relevant surface defect data, construct the magnetic field response ability coefficient Cxxs, and based on the magnetic field response ability coefficient Cxxs, preliminarily predict the uniformity of the permanent magnet rotor during the magnetizing operation; S3. On the basis of step S2, monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and according to the relevant statistical data and relevant magnetic data, analyze the alignment of the magnetic domains in the permanent magnet rotor under the action of an external magnetic field and the magnetic field stability of the permanent magnet rotor, and combine the magnetic field response ability coefficient Cxxs and the trained magnetizing state prediction model to construct a comprehensive magnetic performance stability index Zwzs; S4. Preset a risk threshold R, and combine the value of the comprehensive magnetic performance stability index Zwzs to grade the magnetizing state of the permanent magnet rotor, and based on the grade score, take corresponding control measures for the permanent magnet rotor.

2. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 1, characterized in that: The specific steps of S1 include: S11. At the edges on the cross-section of the permanent magnet rotor, set fixed points on average, and collect the shape regularity of the permanent magnet rotor between each fixed point to obtain relevant shape data. The relevant shape data includes the radius values between each fixed point, and the maximum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor is extracted therefrom. And the minimum radius between the corresponding fixed points on the cross-section of the permanent magnet rotor. ; S12. During the manufacturing stage of the permanent magnet rotor, the surface of the permanent magnet rotor is evenly divided into several groups of regions to monitor the relevant temperature change data of each region on the surface of the permanent magnet rotor during the heat treatment and cooling processes. Among them, the relevant temperature change data includes the temperature difference of each region on the surface of the permanent magnet rotor and the cooling rate difference ; S13. Before magnetizing the manufactured permanent magnet rotor, monitor the relevant surface defect data of the permanent magnet rotor, where the relevant surface defect data includes the surface roughness of the permanent magnet rotor .

3. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 2, characterized in that: The specific steps of S2 include: S21. According to the relevant shape data obtained in step S11, analyze the shape regularity between each fixed point in the permanent magnet rotor to obtain the flatness factor Bpyz between the fixed points. The flatness factor Bpyz between the corresponding fixed points is specifically obtained through the following formula: ； In the formula, represents the minimum radius between corresponding fixed points on the cross-section of the permanent magnet rotor, represents the maximum radius between corresponding fixed points on the cross-section of the permanent magnet rotor; S22. According to the relevant temperature change data obtained in step S12, analyze the temperature change difference of each area on the surface of the permanent magnet rotor, and after dimensionless processing, obtain the heat treatment factor Rcyz of each area. The heat treatment factor Rcyz of the corresponding area is specifically obtained through the following formula: ； In the formula, and are both weight values, represents the temperature difference, represents the difference in cooling rate, where , and The specific values are set by the user according to the situation.

4. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 3, characterized in that: The specific steps of S2 further include: S23. According to the flatness factor Bpyz between the fixed points and the heat treatment factor Rcyz of each area, and combined with the relevant surface defect data of the permanent magnet rotor, preliminarily analyze the risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor, and after dimensionless processing, construct the magnetic field response ability coefficient Cxxs. The magnetic field response ability coefficient Cxxs is obtained through the following formula: ； In the formula, represents the flattening factor between the fixed points, represents the average flattening factor, represents the heat treatment factor of the region, represents the average heat treatment factor represents the surface roughness, represents the number between the fixed points, represents the number of regions, , , and are all weight values. Among them, , and The specific values are set by the user according to the situation.

5. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 1, characterized in that: The specific steps of S2 further include: S24. By comparing the magnetic field response ability coefficient Cxxs with a preset threshold value, preliminarily predict the uniformity of the permanent magnet rotor during the magnetizing operation. The specific content is as follows: When the magnetic field response ability coefficient Cxxs does not exceed the preset threshold, it is preliminarily analyzed that there is a risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is unevenness during the magnetization operation of the permanent magnet rotor, triggering the risk mechanism; When the magnetic field response ability coefficient Cxxs exceeds the preset threshold, it is preliminarily analyzed that there is no risk of uneven magnetic flux distribution caused by permanent magnet material defects in the permanent magnet rotor. At this time, it is preliminarily predicted that there is no unevenness during the magnetization operation of the permanent magnet rotor, and the risk mechanism is not triggered temporarily.

6. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 5, characterized in that: The specific steps of S3 include: S31. Monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor according to the risk mechanism triggered in step S24, where the relevant statistical data includes the number of magnetic domains per unit volume and the volume of the permanent magnet rotor ; the relevant magnetic data includes the magnetic moment of a single magnetic domain and the magnetic induction intensity ; S32. According to relevant statistical data and relevant magnetic data, analyze the alignment of magnetic domains in the permanent magnet rotor under the action of an external magnetic field and the energy storage ability of the permanent magnet rotor in the external magnetic field, so as to calculate and obtain the magnetization intensity Chq and the magnetic energy density Cnmd: ； In the formula, represents the number of magnetic domains per unit volume; represents the magnetic moment of a single magnetic domain; represents the volume of the permanent magnet rotor.

7. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 6, characterized in that: The specific steps of S3 also include: S33. The magnetic energy density Cnmd is obtained through the following formula: ； In the formula, represents the magnetic induction intensity, and Cxxs represents the magnetic field response ability coefficient.

8. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 7, characterized in that: The specific steps of S3 also include: S34. Based on the magnetization intensity Chq and the magnetic energy density Cnmd obtained in steps S32 and S33, and combined with deep learning technology, construct a magnetization state prediction model and dimensionless processing, and fit and output the comprehensive magnetic performance stability index Zwzs. The comprehensive magnetic performance stability index Zwzs is obtained through the following formula: ； In the formula, and are both weight values, represents the logarithmic function with the constant as the base, represents the Euler number; represents a correction constant, where , and The specific values are set by the user according to the situation.

9. A method for controlling the magnetization uniformity of a permanent magnet rotor according to claim 1, characterized in that: The specific steps of S4 include: S41. By comparing the comprehensive magnetic performance stability index Zwzs with the risk threshold R, grade scoring is performed on the magnetization state of the permanent magnet rotor. The specific content is as follows: When the comprehensive magnetic performance stability index Zwzs exceeds the risk threshold R, a first-level score is generated. At this time, the current magnetization process setting is maintained, the magnetic field response and magnetic energy density are continuously monitored to ensure the stability of the magnetization process, and periodic monitoring is performed to track the change of the comprehensive magnetic performance stability index Zwzs; When the comprehensive magnetic performance stability index Zwzs does not exceed the risk threshold R, a second-level score is generated. At this time, according to the shape rule of the permanent magnet rotor, the magnetization time is increased and the magnetization current is adjusted, and the warning system is started to analyze various indicators, identify potential failure modes, and adaptively adjust the selection and layout of the permanent magnet material according to the analysis results of the magnetic energy density Cnmd and the magnetization intensity Chq to optimize the magnetic permeability and magnetization uniformity.

10. A permanent magnet rotor magnetization uniformity control device for implementing the permanent magnet rotor magnetization uniformity control method according to any one of claims 1 to 9 above, characterized in that: Including a data acquisition module, a preliminary analysis module, a stability comprehensive analysis module and a grade feedback module; The data acquisition module is used to set fixed points at different positions of the permanent magnet rotor in advance to collect the shape regularity of the permanent magnet rotor between the fixed points, obtain relevant shape data, and during the manufacturing stage of the permanent magnet rotor, monitor the relevant temperature change data of each area on the surface of the permanent magnet rotor during the heat treatment and cooling processes, and monitor the relevant surface defect data of the permanent magnet rotor before the magnetizing operation is performed on the permanent magnet rotor; The preliminary analysis module is used to construct the magnetic field response ability coefficient Cxxs based on the relevant temperature change data, relevant shape data, and relevant surface defect data, and based on the magnetic field response ability coefficient Cxxs, preliminarily predict the uniformity of the permanent magnet rotor during the magnetizing operation; The stability comprehensive analysis module is used to monitor the relevant statistical data and relevant magnetic data of the magnetic domains in the permanent magnet rotor, and based on the relevant statistical data and relevant magnetic data, analyze the alignment of the magnetic domains in the permanent magnet rotor and the magnetic field stability of the permanent magnet rotor under the action of an external magnetic field, and combine the magnetic field response ability coefficient Cxxs and the trained magnetizing state prediction model to construct a comprehensive magnetic performance stability index Zwzs; The grade feedback module is used to preset a risk threshold R, and combine the value of the comprehensive magnetic performance stability index Zwzs to grade the magnetizing state of the permanent magnet rotor, and based on the grade score, take corresponding control measures for the permanent magnet rotor.

Images