A pulse magnetic field monitoring method and system for a magnetizing process

Abstract

The application relates to the technical field of magnetic charging detection, in particular to a pulse magnetic field monitoring method and system for a magnetic charging process, which comprises the following steps: a magnetic field data acquisition step, a magnetic charging processing signal is synchronized to a preset pulse magnetic field detector by configuring a time sequence synchronization strategy, so that multi-source sensing data acquisition in a magnetic charging area is triggered; a data fusion modeling step, a magnetic field three-dimensional virtual model is constructed according to the multi-source sensing data, so as to feedback the dynamic change of the magnetic field in the magnetic charging process; a magnetic charging real-time evaluation step, whether key indexes conform to preset qualified index thresholds is analyzed based on the magnetic field three-dimensional virtual model, and a feedback evaluation result is generated; a magnetic charging feedback adjustment step, the feedback evaluation result and a preset magnetic charging parameter analysis model are calculated to determine magnetic charging output parameters and adjust the magnetic charging device. The application has the effect that a workpiece to be processed can be timely adjusted in the magnetic charging process, so that the processing qualified rate of the magnet is improved.

Description

Technical Field

[0001] This application relates to the field of magnetization detection technology, and in particular to a pulsed magnetic field monitoring method and system for the magnetization process. Background Technology

[0002] In the field of magnet manufacturing and processing, magnets of special shapes are usually magnetized to form a permanent magnetic field on the workpiece. Detecting the magnetic field change of the magnet is an important basis for judging whether the magnetization effect has achieved the expected result.

[0003] In related technologies, when testing the magnetic field of a magnet, a magnetic flux measuring instrument is used to detect and analyze the magnetic flux of the magnet. This allows the magnetic field change parameters of the workpiece after magnetization to be determined and compared with the set reference magnetic field parameters to determine whether the magnet meets the requirements for a qualified product. Since the workpiece has already formed a relatively stable magnetic field, it is difficult to adjust the magnetic field. At this time, only unqualified workpieces can be screened.

[0004] Regarding the aforementioned technologies, post-processing detection and analysis of the magnetic field of magnets makes it difficult or challenging to perform secondary processing on magnets under stable magnetic fields, which is not conducive to improving the processing qualification rate of magnets. Summary of the Invention

[0005] In order to enable timely magnetization adjustment of the workpiece during the magnetization process and improve the processing qualification rate of the magnet, this application provides a pulse magnetic field monitoring method and system for the magnetization process.

[0006] Firstly, this application provides a method for monitoring the pulsed magnetic field during the magnetization process: A method for monitoring the pulsed magnetic field during a magnetization process, comprising: The magnetic field data acquisition step is configured with a timing synchronization strategy to detect the magnetization processing signal of the workpiece and synchronize it to the pulse magnetic field detector to trigger the acquisition of multi-source sensor data in the magnetization area. The data fusion modeling step involves constructing a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data. This model is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. The real-time magnetization evaluation step uses a three-dimensional virtual model of the magnetic field for comparative analysis to determine whether the set key indicators meet the preset qualified indicator thresholds and generates feedback evaluation results. The magnetization feedback adjustment step calculates the magnetization output parameters based on the feedback evaluation results and the preset magnetization parameter analysis model, and adjusts the magnetization device to keep the key indicators within the preset qualified indicator threshold range.

[0007] By adopting the above technical solution, the magnetic field data acquisition step triggers multi-source sensor data acquisition through a time-series synchronization strategy, avoiding electromagnetic interference signals generated at other times. This constructs a visualized three-dimensional virtual model that reflects the magnetic field distribution during the magnetization process, which can accurately and effectively reflect the magnetic field distribution. This allows the subsequent magnetization feedback adjustment step to effectively adjust the magnetization output parameters, which helps to improve the magnetization quality of the workpiece and increase the workpiece's processing qualification rate.

[0008] Optionally, the magnetization parameter analysis model calculates the magnetization output parameters based on preset key magnetization coefficients, and the magnetization parameter analysis model uses the following formula for calculation: ; ; ; in, This is the adjustment amount of the magnetization voltage. This is the adjustment amount of the magnetizing current. This is the adjustment amount for the magnetization pulse width. The target magnetic field peak value is preset. This represents the peak magnetic field measured by the sensor array at the current moment. This is the output voltage of the current magnetization device. This represents the current operating current of the magnetizing equipment. The duration of the current magnetization pulse. The preset rise time of the target magnetic field, This represents the currently measured rise time of the magnetic field. This is the preset voltage regulation coefficient. This is the current regulation coefficient. This is the pulse width adjustment coefficient.

[0009] By adopting the above technical solution, the magnetization parameter analysis model quantifies the adjustment of magnetization voltage, current, and pulse width using specific formulas. It combines the preset target magnetic field peak value and rise time with the current measured data, and introduces voltage, current, and pulse width adjustment coefficients. This ensures that the parameter adjustment directly corresponds to the deviation of key magnetic field indicators, ensuring that the adjustment direction and amplitude accurately match the magnetization requirements. This further guarantees that key indicators stably meet the qualified threshold requirements during the magnetization process, improving the magnetization accuracy and the consistency of the magnet's magnetic field parameters.

[0010] Optionally, in the magnetic field data acquisition step, a temperature interference compensation sub-strategy is also configured when data acquisition within the magnetization area is triggered: The temperature output value is corrected using a preset temperature compensation model, which is calculated using the following formula: ; in, The magnetic field strength after temperature compensation. This represents the peak magnetic field measured by the sensor array at the current moment. The preset temperature compensation coefficient for the Hall sensor. This is the currently detected sensor temperature value. This is the preset calibration temperature value.

[0011] By adopting the above technical solution, the magnetic field data acquisition step is configured with a temperature interference compensation sub-strategy. The measured magnetic field strength is corrected by the temperature compensation model formula. The temperature compensation coefficient of the Hall sensor, the current sensor temperature and the calibration temperature are introduced to effectively counteract the interference of temperature changes on magnetic field measurement. This solves the problem of distortion of measured magnetic field data caused by temperature drift, making the compensated magnetic field strength closer to the true value. This provides reliable basic data for subsequent data fusion modeling, real-time magnetization evaluation and feedback adjustment.

[0012] Optionally, the magnetization feedback adjustment step is further configured with a magnetization segmentation adjustment strategy, which divides the magnetization stage into a pre-magnetization stage, a main magnetization stage, and a supplementary magnetization stage. The corresponding magnetic field magnetization interval is set according to the corresponding magnetization stage, and a magnetic field stability analysis strategy is configured to analyze the magnetic field stability coefficient of the workpiece in different magnetization stages. When the magnetic field stability coefficient is lower than the preset benchmark stability coefficient, a magnetization learning database for the workpiece to be processed is constructed based on the magnetization output parameters. The optimal magnetization output parameters are determined by analyzing the magnetization learning database and the pre-set magnetic field stability analysis model.

[0013] By adopting the above technical solution, the magnetization feedback adjustment step is configured with a segmented magnetization adjustment strategy, which divides magnetization into pre-magnetization, main magnetization, and supplementary magnetization stages and sets corresponding magnetic field ranges. Combined with the magnetic field stability analysis strategy, the stability coefficient of each stage is evaluated. When it is lower than the benchmark, a magnetization learning database is constructed and the optimal parameters are determined. Compared with the traditional magnetization process, which directly performs rapid magnetization, it helps to reduce the temperature rise caused by a large magnetic field acting on the workpiece, making the workpiece less prone to magnetic deterioration and improving the processing stability.

[0014] Optionally, the magnetic field stability analysis strategy includes: The magnetic field cloud map, local magnetic field intensity distribution and hysteresis loss map at each moment during the magnetization process are collected and analyzed to determine the magnetization stability level in each magnetization stage. The magnetization stability level is calculated using the following evaluation model: ; Where S represents the magnetization stability level, The weights for the preset magnetic flux density deviation, The weights for the attenuation characteristics reflect the convergence ability of the deviation during the magnetization process. The weighting of the set temperature rise rate reflects the importance of the magnet's thermal stability. The maximum deviation of local magnetic flux density is derived from magnetic field contour plot analysis. It refers to the maximum difference between the magnetic flux density at any measuring point and the target value within a certain pulse cycle during the magnetization process. The limit value for magnetic flux density deviation is determined based on the characteristics of the magnet material. This is the normalization coefficient for the magnetic flux density deviation. The decay time constant of the deviation with respect to the pulse period is obtained from the dynamic analysis of the local magnetization distribution. As a reference decay time constant, This is the normalization coefficient for the attenuation characteristics. The local temperature rise rate, calculated from the hysteresis loss diagram combined with infrared thermography data, refers to the rate of change of the highest local temperature of the magnet over time during the magnetization process. This is the limit value for the rate of temperature rise, determined based on the heat resistance of the magnet. This is the normalization coefficient for the rate of temperature rise.

[0015] By adopting the above technical solution, the magnetic field stability analysis strategy collects magnetic field cloud maps, local magnetic field intensity distribution and hysteresis loss maps during the magnetization process. The stability of each stage is quantitatively evaluated from three key dimensions: magnetic flux density uniformity, deviation convergence ability and thermal stability, through the stability level formula. This makes the stability level classification more scientific and accurate, and provides a reliable basis for segmented adjustment and determination of optimal parameters.

[0016] Optionally, the magnetic field data acquisition step is further configured with a detection model adaptation strategy, including: The magnetization pulse signal is analyzed to determine the magnetization intensity and magnetization stage, and the preferred detection model corresponding to the magnetization stage is matched with the preset detection model database. Based on the priority detection model, the power supply current waveform data of different magnetization stages are collected, and the data is analyzed according to the preset environmental interference analysis strategy to determine the stability coefficient of current data acquisition. When the stability of current data acquisition is greater than the preset steady-state acquisition coefficient, noise reduction of the power supply current waveform data is performed using the preset filtering model.

[0017] By adopting the above technical solution, the magnetic field data acquisition step configures the detection model adaptation strategy, selects the priority detection model based on the magnetization intensity and stage, collects the power supply current waveform data, and determines the current acquisition stability coefficient by combining the environmental interference analysis strategy. The priority detection model ensures the targeted acquisition of current data in different magnetization stages, while environmental interference analysis and noise reduction reduce the impact of current fluctuations on the data, making the current data more stable and accurate.

[0018] Optionally, the preset environmental interference analysis strategy includes: The corresponding environmental interference data is collected based on the pulse intensity during the magnetization phase, and the waveform curve changes of the current data are analyzed to determine the current fluctuation amplitude. When the current fluctuation amplitude exceeds the preset allowable amplitude range, the current detection interference source is determined by analyzing the environmental interference data. The current detection interference source includes temperature drift interference source, electromagnetic radiation interference source and mechanical vibration interference source. The stability coefficient of current data acquisition is determined by analyzing the sources of interference in current detection. The stability coefficient of current data acquisition is calculated using the following formula: ; in, The stability coefficient for current data acquisition. The standard deviation of the current sampling points of the continuous magnetization output parameter N during the current magnetization phase is given. This represents the average value of N current sampling points during the current magnetization phase, representing the continuous magnetization output parameters. This is a relative value of current fluctuation, reflecting the proportion of the current fluctuation amplitude to the average current. The preset interference source influence weighting coefficient is assigned a value based on the type of current detection interference source.

[0019] By adopting the above technical solution, the preset environmental interference analysis strategy adapts the interference data acquisition to the pulse intensity during the magnetization stage, determines the current fluctuation amplitude and locates the interference source, and quantifies the stability through the current data acquisition stability coefficient formula. This not only clarifies the source of interference for subsequent targeted prevention and control, but also objectively evaluates the reliability of the current data through the quantification coefficient, avoiding data deviation caused by unknown interference, further improving the accuracy of the basic data of magnetic field monitoring, and supporting the effectiveness of subsequent magnetization assessment and adjustment.

[0020] Optionally, the magnetic field stability analysis model is configured with the following analysis strategies: The magnetization output parameters stored in the magnetization learning database are divided according to the magnetization stage to form sub-magnetization output parameter libraries corresponding to the pre-magnetization stage, main magnetization stage, and supplementary magnetization stage. Based on the analysis of each magnetization output parameter in the sub-magnetization output parameter library and the magnetic field strength formed by the workpiece in the corresponding magnetization stage, the target magnetization output parameter that is closest to the target magnetic field strength required to be formed in the magnetization stage is determined. The optimal magnetization output parameters are dynamically updated based on the target output parameters to construct the magnetization output parameters required for adjusting the magnetization output parameters in different magnetization stages.

[0021] By adopting the above technical solutions, the sub-parameter library ensures that parameter analysis focuses on the characteristics of each stage, the target parameter screening makes the adjustment direction accurately point to the target magnetic field strength of each stage, and the dynamic update allows the optimal parameters to be continuously optimized with the accumulation of learning data, avoiding the decrease in adaptability caused by parameter solidification, ensuring that the parameters of each stage are always in the optimal state during long-term magnetization, and improving the consistency and pass rate of magnet processing.

[0022] Secondly, this application provides a pulsed magnetic field monitoring system for the magnetization process, employing the following technical solution: A pulsed magnetic field monitoring system for a magnetization process, comprising: The magnetic field data acquisition module is configured with a timing synchronization strategy to detect and synchronize the magnetization processing signal of the workpiece to a preset pulse magnetic field detector, so as to trigger the acquisition of data in the magnetization area to obtain multi-source sensor data. The data fusion modeling module constructs a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data, which is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. The real-time magnetization evaluation module uses a three-dimensional virtual model of the magnetic field for comparative analysis to determine whether the set key indicators meet the preset qualified indicator thresholds and generates feedback evaluation results. The magnetization feedback adjustment module calculates based on the feedback evaluation results and the preset magnetization parameter analysis model to determine the magnetization output parameters and adjust the magnetization device to keep key indicators within the preset qualified indicator threshold range.

[0023] By adopting the above technical solution, the magnetic field data acquisition module triggers multi-source data acquisition with a time-series synchronization strategy, the data fusion modeling module constructs a three-dimensional virtual model of the magnetic field, the real-time magnetization evaluation module compares key indicators to generate feedback, and the magnetization feedback adjustment module calculates and adjusts the magnetization output parameters, thus realizing the integration and automation of magnetization process monitoring and adjustment, ensuring that the magnetization process is controllable in real time, and greatly improving the magnet processing efficiency and pass rate.

[0024] In summary, this application includes at least one of the following beneficial technical effects: 1. The magnetic field data acquisition step uses a timing synchronization strategy to trigger multi-source sensor data acquisition, avoiding electromagnetic interference signals generated at other times, thereby constructing a visualized three-dimensional virtual model that reflects the magnetic field distribution during the magnetization process. This model can accurately and effectively reflect the magnetic field distribution, enabling the subsequent magnetization feedback adjustment step to effectively adjust the magnetization output parameters, which helps to improve the magnetization quality of the workpiece and increase the workpiece's processing qualification rate. 2. The magnetization parameter analysis model quantifies the adjustment of magnetization voltage, current and pulse width through specific formulas. It combines the preset target magnetic field peak value and rise time with the current measured data, and introduces voltage, current and pulse width adjustment coefficients so that the parameter adjustment directly corresponds to the deviation of key magnetic field indicators, ensuring that the adjustment direction and amplitude are accurately matched to the magnetization requirements. 3. The magnetic field data acquisition step is configured with a temperature interference compensation sub-strategy. The measured magnetic field strength is corrected by the temperature compensation model formula. The temperature compensation coefficient of the Hall sensor, the current sensor temperature and the calibration temperature are introduced to effectively counteract the interference of temperature changes on the magnetic field measurement. This solves the problem of distortion of measured magnetic field data caused by temperature drift, and makes the compensated magnetic field strength closer to the true value. This provides reliable basic data for subsequent data fusion modeling, real-time magnetization evaluation and feedback adjustment. Attached Figure Description

[0025] Figure 1 This is a flowchart of steps S100 to S400 in this application.

[0026] Figure 2 This is a flowchart of steps S401 to S404 in this application.

[0027] Figure 3 This is a flowchart of steps S101 to S103 in this application.

[0028] Figure 4 This is a flowchart of steps S1021 to S1023 in this application.

[0029] Figure 5 This is a flowchart of steps S4041 to S4043 in this application. Detailed Implementation

[0030] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1-5 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0031] The embodiments of the present invention will now be described in further detail with reference to the accompanying drawings.

[0032] This application discloses a pulsed magnetic field monitoring method during the magnetization process. The method involves a magnetic field data acquisition step that triggers multi-source sensor data acquisition using a time-synchronous strategy; a data fusion and modeling step that constructs a three-dimensional virtual model reflecting the dynamic changes of the magnetic field; a real-time magnetization evaluation step that compares key indicators to generate feedback results; and a magnetization feedback adjustment step that adjusts the magnetization output parameters based on a parameter analysis model. This achieves real-time monitoring and dynamic adjustment of the magnetization process, effectively solving the problems of difficult secondary processing and low processing pass rates associated with post-magnet magnetic field detection. Instead of relying on static detection after magnetization, this method captures the dynamic changes of the magnetic field in real time during magnetization and stabilizes key indicators within the acceptable threshold range by adjusting magnetization parameters in real time. This significantly reduces the probability of defective products and improves the pass rate of magnet processing, while also providing technical support for the controllability of the magnetization process.

[0033] Reference Figure 1 The method for monitoring the pulsed magnetic field during the magnetization process includes the following steps: Step S100: Magnetic field data acquisition step, configured with a timing synchronization strategy to detect the magnetization processing signal of the workpiece and synchronize it to the pulse magnetic field detector to trigger the acquisition of multi-source sensor data in the magnetization area. In step S100, the magnetic field data acquisition step, the timing synchronization strategy is the core strategy to ensure precise alignment between the magnetization processing signal and the multi-source sensor data acquisition time. This is achieved through a signal-triggered synchronization module. When the magnetization device starts and outputs a magnetization processing signal, such as a magnetization pulse start signal, the synchronization module captures this signal in real time and transmits it to the pulse magnetic field detector. Upon receiving the synchronization signal, the detector immediately triggers the multi-source sensors within the magnetization area to start acquisition, avoiding data misalignment caused by a time difference between data acquisition and the magnetization process.

[0034] Multi-source sensor data covers key parameters directly related to the magnetization effect, including the magnetic field strength of the magnetization area, the output current and voltage of the magnetization device, the sensor's own temperature, and the magnet surface temperature. Specifically, magnetic field strength data is acquired through a Hall sensor array, current and voltage data are acquired through a high-precision current transformer and voltage sensor, and temperature data is acquired collaboratively through an infrared temperature sensor and a contact temperature sensor. This complementary use of multiple sensors avoids the limitations of a single data dimension; for example, measuring only magnetic field strength cannot determine the impact of temperature on the measurement.

[0035] In actual operation, the data acquisition area is first divided according to the size of the workpiece and the range of the magnetization area. Taking a cylindrical permanent magnet with a diameter of 50 mm and a circular magnetization cavity with a diameter of 80 mm as an example, six sets of Hall sensors are evenly arranged on the inner wall of the magnetization cavity, spaced 60 degrees apart along the circumference. Each set of sensors is 5 mm away from the surface of the workpiece to collect the magnetic field strength in different directions. A current transformer is connected in series and a voltage sensor is connected in parallel in the power supply circuit of the magnetization device to collect the magnetization current and voltage in real time. Three contact temperature sensors are attached to the surface of the workpiece, one each at the top, middle and bottom, and an infrared temperature sensor is arranged on the outside of the magnetization cavity to simultaneously monitor the magnet temperature and the sensor ambient temperature.

[0036] The specific execution logic of the timing synchronization strategy is as follows: set the trigger threshold of the magnetization processing signal, the synchronization module monitors the magnetization voltage in real time, and immediately sends a synchronization command to the pulse magnetic field detector after the threshold is reached. The detector triggers all sensors to start acquisition with a delay of no more than one millisecond. The acquisition frequency is set to 1 kilohertz, that is, 1,000 sets of data are acquired per second to ensure the capture of the dynamic changes of the magnetization pulse.

[0037] After data acquisition, the data is preprocessed: obvious outliers are removed, such as data exceeding the normal range by 10% due to instantaneous sensor interference. All data are uniformly converted to a time-series coordinate system with the magnetization start-up time as the time zero point to avoid time reference deviations between different sensors, and finally a time-aligned multi-source sensor dataset is formed.

[0038] Step S200: Data fusion modeling step, constructing a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data, which is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. In step S200, the data fusion modeling step, the three-dimensional virtual model of the magnetic field is a digital model that can intuitively present the spatial distribution and dynamic changes of the magnetic field during the magnetization process of the workpiece. It can not only show the spatial distribution of the magnetic field intensity at a certain moment, such as the magnetic field gradient inside the magnet, the surface and the surrounding area, but also dynamically replay the changing trend of the magnetic field during the entire magnetization process through the time axis, such as the entire process of the magnetic field from nothing to something, rising to the peak and then stabilizing.

[0039] Data fusion is the process of integrating and processing the multi-source sensor data collected in step S100. By eliminating conflicts between different sensor data, such as the magnetic field strength deviation of different Hall sensors at the same location, and supplementing data gaps, such as the magnetic field value of the area not covered by the sensor, a complete magnetic field data field is formed through interpolation calculation, providing coherent data support for model construction.

[0040] In the actual construction process, finite element modeling software such as ANSYS Maxwell is used to build a 3D virtual model. The core process consists of three steps: Step 1: Construct a geometric model. According to the actual dimensions of the workpiece to be processed and the structure of the magnetizing device, such as the number of turns, inner diameter, and length of the magnetizing coil, establish a one-to-one geometric model in the software, clarify the spatial position relationship among the workpiece to be processed, the magnetizing coil, and the sensor, and ensure that the geometric model is consistent with the actual magnetizing scenario.

[0041] Step 2: Data import and fusion. Import the multi-source sensing data preprocessed in step S200 into the software. Among them, the magnetic field intensity data of the six groups of Hall sensors are used as the measured constraint points, the current and voltage data are used to verify the rationality of the magnetizing energy input, and the temperature data are used to mark the environmental conditions of the magnetic field measurement; through a weighted fusion algorithm, such as weight assignment based on the sensor accuracy, the higher the accuracy of the sensor data, the greater the weight, to eliminate the magnetic field intensity deviation of different Hall sensors at the same moment. For example, at a certain moment, the measured value of sensor No. 3 is 1.2 Tesla, and the measured value of sensor No. 5 is 1.18 Tesla. After calculating according to the accuracy weight, 1.19 Tesla is taken. Then, through a spatial interpolation algorithm, such as Kriging interpolation, supplement the magnetic field data in the areas not covered by the sensors, such as the magnet center area and the outside area of the magnetizing coil, to form a three-dimensional magnetic field data field covering the entire magnetizing space.

[0042] Step 3: Dynamic model generation and verification. Based on the fused magnetic field data field, generate a dynamic magnetic field model at a time step consistent with the acquisition frequency, that is, one millisecond per step. In the model, the magnetic field intensity is represented by cloud maps of different colors. For example, red represents 1.2 to 1.5 Tesla, yellow represents 0.8 to 1.2 Tesla, and blue represents 0 to 0.8 Tesla, and add a time axis control to support dragging to view the magnetic field distribution at any moment; after the model is constructed, perform verification: select the sensor data that was not used as a constraint point in step S100, such as a reserved set of spare Hall sensors, compare the measured value of this sensor with the calculated value at the corresponding position of the model. If the deviation is less than 5%, the model is determined to be qualified; if the deviation exceeds, readjust the interpolation algorithm parameters or supplement the acquisition data until the model accuracy meets the standard.

[0043] Step S300: Real-time magnetizing evaluation step. Based on the three-dimensional virtual magnetic field model, conduct a comparative analysis to determine whether the set key indicators meet the preset qualified indicator thresholds, and generate a feedback evaluation result; In step S300, the real-time magnetization evaluation step, key indicators are the core parameters for measuring whether the magnetization effect meets the standards. These are set according to the application of the workpiece, such as permanent magnets for motors or magnets for sensors. They typically include: peak magnetic field (the maximum magnetic field strength on or inside the magnet's surface); magnetic field uniformity (the deviation rate of magnetic field strength at different locations on the magnet's surface); and magnetic field rise time (the time it takes for the magnetic field to rise from zero to its peak value). The acceptable threshold values ​​are based on pre-set ranges in the product standard. For example, the peak magnetic field threshold for permanent magnets used in motors is set to 1.2 to 1.4 Tesla, the magnetic field uniformity threshold is set to a deviation rate of no more than 8%, and the magnetic field rise time threshold is set to 5 to 15 milliseconds. These thresholds need to be determined in conjunction with the magnetic performance parameters of the magnet material, such as neodymium iron boron or ferrite.

[0044] The feedback evaluation results are a quantitative judgment report on whether the key indicators are qualified. They include the measured values ​​of each indicator, the threshold range, the qualified or unqualified judgment, and the problem description when unqualified, such as the magnetic field peak value of 1.15 Tesla being lower than the lower threshold limit of 1.2 Tesla, or the magnetic field uniformity deviation rate of 10% exceeding the threshold of 8%.

[0045] This step enables real-time quality monitoring of the magnetization process, breaking away from the traditional post-magnetization inspection model. By extracting and comparing indicators based on a 3D model, defects can be detected immediately during magnetization, providing clear direction for subsequent adjustments. For example, if the evaluation finds that the magnetic field uniformity is unsatisfactory, subsequent adjustments should focus on optimizing the magnetic field distribution balance, such as adjusting the current distribution of the magnetization coil. If the rise time is too long, the rise rate of the magnetization energy needs to be optimized, avoiding the problem of unrecoverable defects due to post-detection.

[0046] Step S400: Magnetization feedback adjustment step. Calculations are performed based on the feedback evaluation results and the preset magnetization parameter analysis model to determine the magnetization output parameters and adjust the magnetization device to keep key indicators within the preset qualified indicator threshold range.

[0047] In step S400, the magnetization feedback adjustment step, the magnetization output parameters are the core adjustable operating parameters of the magnetization device, including magnetization voltage, magnetization current, and magnetization pulse width. These parameters directly affect the energy input of the magnetization process, thus determining key indicators such as the peak value of the magnetic field and the rise time. The adjustment logic is based on the feedback evaluation results. If the evaluation results show that the indicators are qualified, the current magnetization output parameters remain unchanged; if the indicators are not qualified, the required parameter change is calculated through the magnetization parameter analysis model, and then the magnetization device is driven to perform adjustment, forming a closed-loop control of evaluation-calculation-adjustment.

[0048] In the magnetization feedback adjustment step, when the magnetization parameter analysis model calculates the magnetization output parameters based on the preset key magnetization coefficients, the analysis model uses the following formula for calculation: ; ; ; in, This is the adjustment amount of the magnetization voltage. This is the adjustment amount of the magnetizing current. This is the adjustment amount for the magnetization pulse width. The target magnetic field peak value is preset. This represents the peak magnetic field measured by the sensor array at the current moment. This is the output voltage of the current magnetization device. This represents the current operating current of the magnetizing equipment. The duration of the current magnetization pulse. The preset rise time of the target magnetic field, This represents the currently measured rise time of the magnetic field. This is the preset voltage regulation coefficient. This is the current regulation coefficient. This is the pulse width adjustment coefficient.

[0049] In the magnetic field data acquisition step, a temperature interference compensation sub-strategy is also configured when triggering data acquisition within the magnetization area: The temperature output value is corrected using a preset temperature compensation model, which is calculated using the following formula: ; in, The magnetic field strength after temperature compensation. This represents the peak magnetic field measured by the sensor array at the current moment. The preset temperature compensation coefficient for the Hall sensor. This is the currently detected sensor temperature value. This is the preset calibration temperature value.

[0050] Reference Figure 2 The magnetization feedback adjustment step is also equipped with a segmented magnetization adjustment strategy, including: Step S401: The magnetization stage is divided into a pre-magnetization stage, a main magnetization stage, and a supplementary magnetization stage; During implementation, the entire magnetization process is first divided into stages, and the core functions of each stage are clarified: the pre-magnetization stage focuses on establishing a basic magnetic field for the workpiece to be processed, so as to avoid damage to the internal structure of the magnet caused by the direct effect of subsequent high-intensity magnetization; the main magnetization stage is the core stage, and the goal is to quickly increase the magnetic field strength of the workpiece to be processed to close to the preset target value; the supplementary magnetization stage is to accurately correct any local magnetic field inhomogeneity or intensity deviation that may exist after the main magnetization, so as to ensure that the overall magnetic field meets the standard.

[0051] Step S402: Set the corresponding magnetic field magnetization interval according to the corresponding magnetization stage, and configure a magnetic field stability analysis strategy to analyze the magnetic field stability coefficient of the workpiece in different magnetization stages. Secondly, specific magnetic field magnetization intervals are set according to the functional positioning of each stage. For example, the interval for the pre-magnetization stage is set to 30%-50% of the target magnetic field strength, the interval for the main magnetization stage is set to 80%-100%, and the interval for the supplementary magnetization stage is set to 90%-100%, ensuring that the interval matches the functional positioning of each stage. At the same time, a magnetic field stability analysis strategy is configured. The core purpose of this strategy is to analyze the stability coefficient of the magnetic field formed on the workpiece in different magnetization stages in real time, determine whether the current magnetic field state meets the stage requirements, and provide a basis for judging whether parameters need to be adjusted in the future.

[0052] Finally, the specific operations of the magnetic field stability analysis strategy were implemented: magnetic field cloud maps (reflecting the overall magnetic field distribution), local magnetic field intensity distribution (capturing detailed deviations), and hysteresis loss maps (correlated with the thermal stability of the magnet) were collected at each moment during the magnetization process. By comprehensively comparing and analyzing the three types of data, the uniformity of the magnetic field, the intensity fluctuation range, and the heating condition of the magnet at each stage were determined. Finally, the magnetization stability level corresponding to each magnetization stage was determined, providing a precise direction for parameter adjustment within the stage.

[0053] Strategies for magnetic field stability analysis include: The magnetic field cloud map, local magnetic field intensity distribution and hysteresis loss map at each moment during the magnetization process are collected and analyzed to determine the magnetization stability level in each magnetization stage. The magnetization stability level is calculated using the following evaluation model: ; Where S represents the magnetization stability level, The weights for the preset magnetic flux density deviation, The weights for the attenuation characteristics reflect the convergence ability of the deviation during the magnetization process. The weighting of the set temperature rise rate reflects the importance of the magnet's thermal stability. The maximum deviation of local magnetic flux density is derived from magnetic field contour plot analysis. It refers to the maximum difference between the magnetic flux density at any measuring point and the target value within a certain pulse cycle during the magnetization process. The limit value for magnetic flux density deviation is determined based on the characteristics of the magnet material. This is the normalization coefficient for the magnetic flux density deviation. The decay time constant of the deviation with respect to the pulse period is obtained from the dynamic analysis of the local magnetization distribution. As a reference decay time constant, This is the normalization coefficient for the attenuation characteristics. The local temperature rise rate, calculated from the hysteresis loss diagram combined with infrared thermography data, refers to the rate of change of the highest local temperature of the magnet over time during the magnetization process. This is the limit value for the rate of temperature rise, determined based on the heat resistance of the magnet. This is the normalization coefficient for the rate of temperature rise.

[0054] Step S403: When the magnetic field stability coefficient is lower than the preset reference stability coefficient, construct a magnetization learning database for the workpiece to be processed based on the magnetization output parameters; The magnetic field stability coefficient is a quantitative indicator calculated using the magnetic field stability analysis strategy mentioned earlier. It mainly reflects the stability of the magnetic field during the current magnetization stage. Specifically, it is calculated by combining data such as the magnetic field uniformity reflected in the magnetic field cloud map, the fluctuation range of the local magnetic field intensity distribution, and the thermal stability of the magnet reflected in the hysteresis loss diagram. Generally, the higher the coefficient value, the more stable the magnetic field state.

[0055] The baseline stability coefficient is a qualified critical value set in advance based on the material characteristics of the workpiece, such as the magnetic stability threshold of neodymium iron boron or ferrite, and the magnetization quality requirements. Once the actual monitored magnetic field stability coefficient is lower than this value, it indicates that the current magnetic field state does not meet the magnetization requirements of the corresponding stage and needs to be optimized through subsequent operations.

[0056] Magnetization output parameters are the core control parameters of the magnetization device during operation. They mainly include magnetization voltage, magnetization current, and magnetization pulse width. These parameters directly affect the formation process and final stable state of the magnetic field.

[0057] The magnetization learning database is a structured data set used to store the correspondence between magnetization output parameters and magnetic field state under unstable magnetic field conditions. Its core purpose is to accumulate parameter experience under various unstable scenarios and provide data support for the subsequent optimization of magnetization output parameters.

[0058] The implementation process of this step is as follows: real-time monitoring of the magnetic field stability coefficient at each magnetization stage. When the value is detected to be lower than the preset benchmark stability coefficient, the magnetization output parameters under the current working condition are immediately collected, such as the current magnetization voltage of 320V, the magnetization current of 520A, and the magnetization pulse width of 18ms. At the same time, the corresponding magnetic field instability characteristics are recorded in detail, such as the local magnetic field strength deviation reaching 15% and the magnet temperature rise rate exceeding the preset standard. Then, the corresponding data of these magnetization output parameters and magnetic field state are classified and stored according to the magnetization stage. As the data is continuously accumulated, the magnetization learning database of the workpiece to be processed is gradually constructed.

[0059] Step S404: Analyze the magnetization learning database and the preset magnetic field stability analysis model to determine the optimal magnetization output parameters.

[0060] The magnetic field stability analysis model is an algorithm model based on the magnetization principle, magnetization characteristics of magnets and historical magnetization data. It has the ability to analyze the correlation between magnetization output parameters and magnetic field stability coefficient. It can also identify the direction of adjusting magnetization output parameters to improve magnetic field stability through data mining.

[0061] The optimal magnetization output parameters refer to the combination of magnetization output parameters that can raise the magnetic field stability coefficient to above the reference stability coefficient, while meeting the requirements of the magnetic field magnetization range in the current magnetization stage. For example, in the main magnetization stage, the magnetic field strength needs to be close to the target value. Therefore, the optimal parameters need to take into account both magnetic field stability and magnetization efficiency.

[0062] The implementation process is as follows: Magnetization output parameters and magnetic field state data under unstable operating conditions stored in the magnetization learning database are input into a preset magnetic field stability analysis model. The model analyzes the influence of different magnetization output parameter adjustments on the magnetic field stability coefficient. For example, it finds that for every 10V increase in magnetization voltage, the magnetic field stability coefficient can increase by 5%, and when the magnetization current fluctuation is controlled within ±10A, the magnetic field uniformity deviation can be reduced by 8%. Then, it selects the parameter combination from numerous combinations that ensures the magnetic field stability of the current magnetization stage meets the requirements of the magnetization range. The final optimal magnetization output parameters may be a magnetization voltage of 330V, a magnetization current of 515A, and a magnetization pulse width of 19ms.

[0063] Reference Figure 3 The magnetic field data acquisition process also includes a detection model adaptation strategy, including: Step S101: Analyze the magnetization pulse signal to determine the magnetization intensity and magnetization stage, and match the priority detection model corresponding to the magnetization stage in the preset detection model database; The magnetization pulse signal is a periodic electrical signal output by the magnetization device to the workpiece. Its amplitude, frequency and duration can directly reflect the energy change during the magnetization process and are the basic signals for judging the magnetization status.

[0064] Magnetization intensity is the energy output level of the current magnetization process. It is usually classified according to the amplitude of the magnetization pulse signal. For example, high amplitude corresponds to high-intensity magnetization, and medium-low amplitude corresponds to medium-low-intensity magnetization. Different intensities are adapted to different signal acquisition requirements.

[0065] The pre-built detector model database is a structured collection of pre-constructed detector models that stores various detector models. Each detector model is designed for the signal characteristics of a specific magnetization stage, such as a slowly varying signal model for the pre-magnetization stage and a spike signal model for the main magnetization stage. The priority detector model is the detector model selected from the pre-built detector model database that best matches the signal characteristics of the current magnetization stage, minimizing signal acquisition errors and improving data accuracy.

[0066] The implementation process is as follows: the magnetization pulse signal is acquired in real time through the signal acquisition unit, the magnetization intensity is determined by analyzing the amplitude of the signal, and the current magnetization stage is determined by combining the signal change pattern. For example, when the signal amplitude is low and the change is gradual, it is determined to be the pre-magnetization stage, and when the signal amplitude rises sharply and the peak value is high, it is determined to be the main magnetization stage. Then, according to the determined magnetization stage, the corresponding priority detection model is matched in the preset detection model database. For example, the low amplitude gradually changing signal detection model is matched for the pre-magnetization stage, and the high amplitude spike signal detection model is matched for the main magnetization stage, so as to prepare for the subsequent current data acquisition.

[0067] Step S102: Based on the priority detection model, the power supply current waveform data of different magnetization stages are collected, and the data is analyzed according to the preset environmental interference analysis strategy to determine the current data acquisition stability coefficient. First, clarify the meaning of the core terms in the steps. The power supply current waveform data is the curve data formed by the change of current in the power supply circuit of the magnetization device over time. It can reflect the real-time input status of magnetization energy and is the key basis for calculating magnetization parameters.

[0068] Environmental interference analysis strategy is a method used to identify and assess the impact of external interference on current data acquisition. It mainly determines whether there is interference such as temperature drift, electromagnetic radiation or mechanical vibration by analyzing the fluctuation amplitude and abnormal peaks of the current waveform.

[0069] The stability coefficient of current data acquisition is an indicator that quantifies the quality of current data acquisition. It is obtained by calculating the ratio of the fluctuation amplitude of continuous sampling points to the average value. The closer the coefficient is to 1, the smaller the fluctuation of the current data and the more stable the acquisition.

[0070] The implementation process of this step is as follows: using the priority detection model matched in step S101 as the acquisition benchmark, the corresponding acquisition mode is activated for different magnetization stages. For example, in the pre-magnetization stage, a low sampling frequency is used to acquire a smooth current waveform, and in the main magnetization stage, a high sampling frequency is used to capture rapidly changing current peaks, thereby obtaining the power supply current waveform data for each stage. Then, according to the preset environmental interference analysis strategy, it is observed whether there are irregular fluctuations or abnormal spikes in the current waveform. For example, the appearance of high-frequency small amplitude fluctuations in the waveform may be caused by electromagnetic radiation interference. Then, by calculating the ratio of the standard deviation to the average value of N consecutive sampling points, the current data acquisition stability coefficient is obtained. For example, if the stability coefficient of 100 consecutive sampling points in a certain main magnetization stage is 0.88, it means that there is a certain fluctuation in the current data in this stage.

[0071] Step S103: When the current data acquisition stability is greater than the preset steady-state acquisition coefficient, noise reduction of the power supply current waveform data is performed using the preset filtering model.

[0072] The steady-state acquisition coefficient is a pre-set critical value used to determine whether current data needs noise reduction. It is usually determined based on the accuracy requirements of the magnetization parameters, for example, it is set to 0.85. If the stability coefficient is greater than this value, it means that the fluctuation of the current data exceeds the acceptable range and noise reduction is required.

[0073] The preset filtering model is an algorithmic model used to filter interference signals in current data. Common models include low-pass filtering models and Kalman filtering models. Different models are suitable for different types of interference. For example, low-pass filtering models can remove high-frequency electromagnetic interference, while Kalman filtering models can suppress random noise. Noise reduction of supply current waveform data removes interference components from the current waveform through the filtering model, making the waveform smoother and closer to the true current change trend, thus providing accurate data for subsequent parameter calculations.

[0074] In practice, the stability coefficient of the current data acquisition obtained in step S102 is compared with the preset steady-state acquisition coefficient. For example, if the stability coefficient of 0.92 is greater than the steady-state acquisition coefficient of 0.85, it is determined that there is significant interference in the current data. Then, a suitable preset filtering model is selected according to the type of interference. For example, a low-pass filtering model is used for high-frequency electromagnetic interference. The power supply current waveform data is processed by the filtering model to filter out the interference signals, such as removing high-frequency spikes in the waveform, so that the processed current waveform can accurately reflect the actual power supply status of the magnetization device and avoid interference data from affecting the calculation accuracy of subsequent magnetization parameters.

[0075] The environmental interference analysis strategies include: Step S1021: Based on the pulse intensity during the magnetization stage, collect corresponding environmental interference data and analyze the waveform curve changes of the current data to determine the current fluctuation amplitude. Pulse intensity is the energy level of the output pulse signal during the magnetization stage. It is usually determined by the peak voltage or current of the pulse signal. The pulse intensity varies in different magnetization stages. For example, the pulse intensity in the main magnetization stage is higher than that in the pre-magnetization stage.

[0076] Environmental interference data refers to external environmental parameters that may affect the accuracy of current acquisition, including changes in ambient temperature, the strength of surrounding electromagnetic signals, and equipment vibration frequency. Current fluctuation amplitude is the difference between adjacent peaks and valleys in the current data waveform curve, reflecting the degree of instability of the current signal; a larger fluctuation amplitude indicates more drastic current changes.

[0077] During implementation, the acquisition parameters for environmental interference data are adjusted based on the pulse intensity of the current magnetization phase. For example, during the main magnetization phase with high pulse intensity, the acquisition frequency of the ambient temperature and electromagnetic signals is increased to ensure comprehensive capture of interference data under strong interference conditions. Simultaneously, the waveform curve of the current data is continuously monitored, and the current fluctuation amplitude is calculated by analyzing the difference between adjacent peaks and troughs in the curve. For instance, during the main magnetization phase, the current waveform may exhibit significant fluctuations; by measuring the numerical difference between adjacent peaks and troughs, the current fluctuation amplitude at this time is determined to be 50A.

[0078] Step S1022: When the current fluctuation amplitude is greater than the preset allowable amplitude range, analyze the environmental interference data to determine the current detection interference source. The current detection interference source includes temperature drift interference source, electromagnetic radiation interference source and mechanical vibration interference source. The allowable amplitude range is the maximum permissible range of current fluctuations pre-set according to the magnetization accuracy requirements. For example, it is set to 0-20A in the pre-magnetization stage and 0-50A in the main magnetization stage. If the actual fluctuation amplitude exceeds this range, the current signal is considered to be affected by interference. Current detection interference sources are external factors that cause unstable current data acquisition.

[0079] Temperature drift interference sources refer to interference caused by changes in ambient temperature that lead to deviations in sensor performance. For example, high temperatures can reduce the measurement accuracy of a current sensor. Electromagnetic radiation interference sources refer to interference from electromagnetic signals generated by surrounding electronic equipment on the current acquisition circuit, such as electromagnetic radiation generated by a nearby motor. Mechanical vibration interference sources refer to interference caused by vibrations during equipment operation that result in poor contact in the acquisition circuit, such as vibrations caused by the operation of mechanical parts in a magnetizing device.

[0080] In practice, the current fluctuation amplitude determined in step S1021 is compared with the preset allowable amplitude range. If the fluctuation amplitude exceeds the range, for example, the fluctuation amplitude reaches 60A during the main magnetization stage, exceeding the allowable 0-50A range, then interference source analysis is initiated. Combined with the collected environmental interference data, if the current fluctuation intensifies synchronously when the ambient temperature rises sharply, it can be determined that there is a temperature drift interference source. If noise appears in the current waveform after the nearby high-frequency equipment is started, it can be determined that there is an electromagnetic radiation interference source. If the equipment vibration frequency is consistent with the current fluctuation frequency, it can be determined that there is a mechanical vibration interference source.

[0081] Step S1023: Analyze the interference sources of current detection to determine the stability coefficient of current data acquisition; The stability coefficient of current data acquisition is a quantitative indicator that comprehensively reflects the degree of interference affecting current data. Its value is related to the type and intensity of the interference source. The stronger the interference, the lower the coefficient, and vice versa. It is used to determine whether the current data meets the requirements of subsequent analysis.

[0082] The implementation process of this step is as follows: Based on the current detection interference sources determined in step S1022, analyze the degree of influence of various interference sources on the current data. For example, if there is a temperature drift interference source and the temperature change range reaches 10℃, the influence weight of this interference source on the current measurement is combined; if there is also an electromagnetic radiation interference source and the electromagnetic signal strength exceeds the standard by 20%, its influence weight is added; by comprehensively evaluating the superimposed influence of various interference sources, the current data acquisition stability coefficient is finally determined. There is no need to disclose the specific calculation formula; it is only necessary to clarify that the coefficient is derived from a comprehensive analysis of the type, intensity, and influence weight of the interference source.

[0083] The stability coefficient of current data acquisition is calculated using the following formula: ; in, The stability coefficient for current data acquisition. The standard deviation of the current sampling points of the continuous magnetization output parameter N during the current magnetization phase is given. This represents the average value of N current sampling points during the current magnetization phase, representing the continuous magnetization output parameters. This is a relative value of current fluctuation, reflecting the proportion of the current fluctuation amplitude to the average current. The preset interference source influence weighting coefficient is assigned a value based on the type of current detection interference source.

[0084] Reference Figure 4 The magnetic field stability analysis model is configured with the following analysis strategies: Step S4041: Divide the magnetization output parameters stored in the magnetization learning database according to the magnetization stage to form sub-magnetization output parameter libraries corresponding to the pre-magnetization stage, main magnetization stage and supplementary magnetization stage. The magnetization learning database is a structured data set constructed in the previous steps that stores the correspondence between magnetization output parameters and magnetic field state under unstable magnetic field conditions. It includes magnetization output parameters under different magnetization stages. Magnetization output parameters refer to the core control parameters of the magnetization device during operation, including magnetization voltage, magnetization current, magnetization pulse width, etc. The sub-magnetization output parameter library is a specialized parameter library formed by splitting according to the magnetization stage. Each sub-library only stores the magnetization output parameters of the corresponding stage, which is convenient for subsequent targeted analysis.

[0085] In the specific implementation process, all stored magnetization output parameters in the magnetization learning database are extracted and classified according to the magnetization stage labels corresponding to the parameters. For example, parameters labeled with pre-magnetization, such as voltage 280-300V, current 450-480A, and pulse width 15-18ms, are classified into the pre-magnetization stage sub-magnetization output parameter library; parameters labeled with main magnetization, such as voltage 320-350V, current 500-530A, and pulse width 18-22ms, are classified into the main magnetization stage sub-magnetization output parameter library. Similarly, the classification of the supplementary magnetization stage sub-magnetization output parameter library is completed to ensure that the parameters of each sub-library match the magnetization requirements of the corresponding stage.

[0086] Step S4042: Analyze each magnetization output parameter in the sub-magnetization output parameter library and the magnetic field strength formed by the workpiece in the corresponding magnetization stage to determine the target magnetization output parameter that is closest to the target magnetic field strength required to be formed in the magnetization stage. The sub-magnetizing output parameter library is a specialized parameter library for each stage as defined in step S4041. The target magnetic field strength is the preset magnetic field strength standard that the workpiece needs to reach in each magnetizing stage. For example, the target magnetic field strength in the pre-magnetizing stage is 0.4-0.6T, in the main magnetizing stage it is 1.2-1.4T, and in the supplementary magnetizing stage it is 1.3-1.4T. The target magnetizing output parameters are the combination of magnetizing output parameters in the sub-magnetizing output parameter library that makes the magnetic field strength formed on the workpiece closest to the target magnetic field strength of that stage.

[0087] The implementation process of this step is as follows: For each sub-magnetizing output parameter library, the magnetizing output parameters in the library are extracted one by one, and the magnetic field strength record of the workpiece to be processed corresponding to each parameter is retrieved. This record comes from the corresponding data of parameter-magnetic field state in the magnetizing learning database. The magnetic field strength corresponding to each parameter is compared with the target magnetic field strength of this stage, and the parameter with the smallest difference is selected as the target magnetizing output parameter.

[0088] Step S4043: Dynamically update the optimal magnetization output parameters required for adjusting the magnetization output parameters in different magnetization stages based on the target output parameters.

[0089] The target output parameters, determined in step S4042, are the magnetization output parameters that enable the magnetic field strength to approach the stage target. The optimal magnetization output parameters are a combination of parameters formed based on the target output parameters and the magnetic field stability requirements. For example, the magnetic field stability coefficient needs to be higher than the benchmark value to ensure that the workpiece reaches the target magnetic field strength and maintains magnetic field stability in the corresponding stage, and will be continuously optimized as new data accumulates.

[0090] The implementation process of this step is as follows: The target output parameters for each stage determined in step S4042 are combined and analyzed with the magnetic field stability data of the corresponding stage. For example, the magnetic field stability coefficient corresponding to the target output parameter. If the magnetic field stability coefficient corresponding to the target output parameter is higher than the benchmark stability coefficient, the parameter is directly used as the optimal magnetization output parameter. If the stability coefficient does not meet the standard, the target output parameter is finely adjusted, such as by slightly adjusting the voltage or current, until a parameter combination that balances the magnetic field strength and stability is found is used as the initial optimal magnetization output parameter.

[0091] As new data is added to the magnetization learning database, steps S4041-S4042 are repeated periodically to update the target output parameters, thereby dynamically optimizing the optimal magnetization output parameters. For example, if the initial optimal parameters in the magnetization stage correspond to a magnetic field stability coefficient of 0.82, after the target output parameters are updated following the addition of new data, the stability coefficient increases to 0.88. The new parameters then become the optimal magnetization output parameters for the updated magnetization stage.

[0092] Based on the same inventive concept, embodiments of the present invention provide a pulsed magnetic field monitoring system for a magnetization process, comprising: The magnetic field data acquisition module is configured with a timing synchronization strategy to detect and synchronize the magnetization processing signal of the workpiece to a preset pulse magnetic field detector, so as to trigger the acquisition of data in the magnetization area to obtain multi-source sensor data. The data fusion modeling module constructs a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data, which is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. The real-time magnetization evaluation module uses a three-dimensional virtual model of the magnetic field for comparative analysis to determine whether the set key indicators meet the preset qualified indicator thresholds and generates feedback evaluation results. The magnetization feedback adjustment module calculates based on the feedback evaluation results and the preset magnetization parameter analysis model to determine the magnetization output parameters and adjust the magnetization device to keep key indicators within the preset qualified indicator threshold range.

[0093] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0094] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for monitoring the pulsed magnetic field during a magnetization process, comprising a pulsed magnetic field detector and a multi-source data sensor for detecting pulsed magnetic field data and sensing data, characterized in that, include: The magnetic field data acquisition step is configured with a timing synchronization strategy to detect the magnetization processing signal of the workpiece and synchronize it to the pulse magnetic field detector, so as to trigger the acquisition of multi-source sensor data in the magnetization area and ensure that the magnetization processing signal and the multi-source sensor data acquisition time are accurately aligned. The data fusion modeling step involves constructing a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data. This model is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. The real-time magnetization evaluation step involves comparative analysis based on a three-dimensional virtual magnetic field model to determine whether the set key indicators meet the preset qualification thresholds and to generate feedback evaluation results. The three-dimensional virtual magnetic field model is a digital model of the spatial distribution and dynamic temporal changes of the magnetic field during the magnetization process of the workpiece. The key indicators include magnetic field peak value, magnetic field uniformity, and magnetic field rise time. The magnetization feedback adjustment step calculates the magnetization output parameters based on the feedback evaluation results and the preset magnetization parameter analysis model, and adjusts the magnetization device to keep the key indicators within the preset qualified indicator threshold range. The magnetization parameter analysis model calculates the magnetization output parameters based on preset key magnetization coefficients. The magnetization parameter analysis model uses the following formula for calculation: ； ； ； in, This is the adjustment amount of the magnetization voltage. This is the adjustment amount of the magnetizing current. This is the adjustment amount for the magnetization pulse width. The target magnetic field peak value is preset. This represents the peak magnetic field measured by the sensor array at the current moment. This is the output voltage of the current magnetization device. This represents the current operating current of the magnetizing equipment. The duration of the current magnetization pulse. The preset rise time of the target magnetic field, This represents the currently measured rise time of the magnetic field. This is the preset voltage regulation coefficient. This is the current regulation coefficient. This is the pulse width adjustment coefficient.

2. The pulse magnetic field monitoring method for the magnetization process according to claim 1, characterized in that, In the magnetic field data acquisition step, a temperature interference compensation sub-strategy is also configured when data acquisition within the magnetization area is triggered: The temperature output value is corrected using a preset temperature compensation model, which is calculated using the following formula: ； in, The magnetic field strength after temperature compensation. This represents the peak magnetic field measured by the sensor array at the current moment. The preset temperature compensation coefficient for the Hall sensor. This is the currently detected sensor temperature value. This is the preset calibration temperature value.

3. The pulse magnetic field monitoring method for the magnetization process according to claim 1, characterized in that, The magnetization feedback adjustment step is also configured with a magnetization segmentation adjustment strategy, which divides the magnetization stage into a pre-magnetization stage, a main magnetization stage, and a supplementary magnetization stage. The corresponding magnetic field magnetization interval is set according to the corresponding magnetization stage, and a magnetic field stability analysis strategy is configured to analyze the magnetic field stability coefficient of the workpiece in different magnetization stages. When the magnetic field stability coefficient is lower than the preset benchmark stability coefficient, a magnetization learning database for the workpiece to be processed is constructed based on the magnetization output parameters. The optimal magnetization output parameters are determined by analyzing the magnetization learning database and the preset magnetic field stability analysis model. The magnetic field stability analysis strategy includes: The magnetic field cloud map, local magnetic field intensity distribution and hysteresis loss map at each moment during the magnetization process are collected and analyzed to determine the magnetization stability level in each magnetization stage. The magnetization stability level is calculated using the following evaluation model: ； Where S represents the magnetization stability level, The weights for the preset magnetic flux density deviation, The weights for the attenuation characteristics reflect the convergence ability of the deviation during the magnetization process. The weighting of the set temperature rise rate reflects the importance of the magnet's thermal stability. The maximum deviation of local magnetic flux density is derived from magnetic field contour plot analysis. It refers to the maximum difference between the magnetic flux density at any measuring point and the target value within a certain pulse cycle during the magnetization process. The limit value for magnetic flux density deviation is determined based on the characteristics of the magnet material. This is the normalization coefficient for the magnetic flux density deviation. The decay time constant of the deviation with respect to the pulse period is obtained from the dynamic analysis of the local magnetization distribution. As a reference decay time constant, This is the normalization coefficient for the attenuation characteristics. The local temperature rise rate, calculated from the hysteresis loss diagram combined with infrared thermography data, refers to the rate of change of the highest local temperature of the magnet over time during the magnetization process. This is the limit value for the rate of temperature rise, determined based on the heat resistance of the magnet. This is the normalization coefficient for the rate of temperature rise.

4. The pulse magnetic field monitoring method for the magnetization process according to claim 3, characterized in that, The magnetic field data acquisition step also includes a detection model adaptation strategy, including: The magnetization pulse signal is analyzed to determine the magnetization intensity and magnetization stage, and the preferred detection model corresponding to the magnetization stage is matched with the preset detection model database. Based on the priority detection model, the power supply current waveform data of different magnetization stages are collected, and the data is analyzed according to the preset environmental interference analysis strategy to determine the stability coefficient of current data acquisition. When the stability of current data acquisition is greater than the preset steady-state acquisition coefficient, noise reduction of the power supply current waveform data is performed using the preset filtering model. The preset environmental interference analysis strategy includes: The corresponding environmental interference data is collected based on the pulse intensity during the magnetization phase, and the waveform curve changes of the current data are analyzed to determine the current fluctuation amplitude. When the current fluctuation amplitude exceeds the preset allowable amplitude range, the current detection interference source is determined by analyzing the environmental interference data. The current detection interference source includes temperature drift interference source, electromagnetic radiation interference source and mechanical vibration interference source. The stability coefficient of current data acquisition is determined by analyzing the sources of interference in current detection. The stability coefficient of current data acquisition is calculated using the following formula: ； in, The stability coefficient for current data acquisition. The standard deviation of the current sampling points of the continuous magnetization output parameter N during the current magnetization phase is given. This represents the average value of N current sampling points during the current magnetization phase, representing the continuous magnetization output parameters. This is a relative value of current fluctuation, reflecting the proportion of the current fluctuation amplitude to the average current. The preset interference source influence weighting coefficient is assigned a value based on the type of current detection interference source.

5. The pulse magnetic field monitoring method for the magnetization process according to claim 3, characterized in that, The magnetic field stability analysis model is configured with the following analysis strategies: The magnetization output parameters stored in the magnetization learning database are divided according to the magnetization stage to form sub-magnetization output parameter libraries corresponding to the pre-magnetization stage, main magnetization stage, and supplementary magnetization stage. Based on the analysis of each magnetization output parameter in the sub-magnetization output parameter library and the magnetic field strength formed by the workpiece in the corresponding magnetization stage, the target magnetization output parameter that is closest to the target magnetic field strength required to be formed in the magnetization stage is determined. The optimal magnetization output parameters are dynamically updated based on the target output parameters to construct the magnetization output parameters required for adjusting the magnetization output parameters in different magnetization stages.

6. A pulsed magnetic field monitoring system for a magnetization process, used to implement the pulsed magnetic field detection method for a magnetization process as described in any one of claims 1-5, characterized in that, include: The magnetic field data acquisition module is configured with a timing synchronization strategy to detect and synchronize the magnetization processing signal of the workpiece to a preset pulse magnetic field detector, so as to trigger the acquisition of data in the magnetization area to obtain multi-source sensor data. The data fusion modeling module constructs a three-dimensional virtual model of the magnetic field formed during the magnetization process of the workpiece based on multi-source sensor data, which is used to reflect the dynamic changes of the magnetic field during the magnetization process of the workpiece. The real-time magnetization evaluation module uses a three-dimensional virtual model of the magnetic field for comparative analysis to determine whether the set key indicators meet the preset qualified indicator thresholds and generates feedback evaluation results. The magnetization feedback adjustment module calculates based on the feedback evaluation results and the preset magnetization parameter analysis model to determine the magnetization output parameters and adjust the magnetization device to keep key indicators within the preset qualified indicator threshold range.

Images