A magnetic material hydro-forming control system

Abstract

This invention discloses a hydraulic forming control system for magnetic materials, employing a first acquisition module, a second acquisition module, a third acquisition module, a fourth acquisition module, and a fifth acquisition module. The first acquisition module acquires dynamic parameters through a sensor array and preprocesses these parameters using a data fusion algorithm to obtain smooth pressure data and displacement time-series data. The second acquisition module trains a collaborative control model of pressure and displacement using a support vector machine algorithm to determine the dynamic mapping relationship between pressure and displacement, obtaining an optimized combination of control parameters. The third acquisition module, if the control deviation of the dynamic mapping relationship between pressure and displacement exceeds a preset threshold, adjusts the opening of the servo valve in the hydraulic system in real time, combined with the optimized control parameter combination, to correct the pressure output and obtain a stable pressure application curve. This invention significantly improves the accuracy and product quality of hydraulic forming.

Description

Technical Field

[0001] This invention relates to the field of magnetic material manufacturing technology, and in particular discloses a magnetic material hydraulic forming control system. Background Technology

[0002] Magnetic material manufacturing is an indispensable field in modern industry, widely used in motors, sensors, and new energy equipment. The quality of these products directly impacts device performance and energy efficiency. Precision molding technology, as a core component, has a decisive influence on material density, dimensional accuracy, and defect control, making it crucial for industry competitiveness.

[0003] However, traditional hydroforming technology has significant limitations. Low pressure control precision often leads to uneven product density or internal cracks, especially in the manufacture of high-performance magnetic materials such as NdFeB or FeSiA, where the yield rate is difficult to meet high-end demands. Furthermore, traditional equipment is energy-intensive and has low mold changeover efficiency, limiting production flexibility and cost control.

[0004] The core challenge in addressing these issues lies in achieving high-precision pressure and displacement control while maintaining production efficiency under complex processes. Unstable pressure control can lead to uneven material stress during molding, resulting in uneven density distribution and affecting the performance stability of magnetic materials. This uneven density further exacerbates the release of internal stress during demolding, easily causing microcracks or product deformation, especially prominent in the molding of high-strength permanent magnet materials. These interconnected technical challenges collectively restrict the improvement of magnetic material molding quality.

[0005] Therefore, how to achieve precise coordinated control of pressure and displacement in high-pressure molding, while optimizing the demolding process to reduce the risk of cracking, has become a key issue in improving the molding quality and production efficiency of magnetic materials. Summary of the Invention

[0006] This invention provides a magnetic material hydraulic forming control system, which aims to solve at least one of the defects existing in the prior art.

[0007] This invention relates to a hydraulic forming control system for magnetic materials, comprising:

[0008] The first acquisition module is used to acquire dynamic parameters from the hydraulic system and mold equipment through a sensor array, and to preprocess the dynamic parameters using a data fusion algorithm to obtain smooth pressure data and displacement time series data.

[0009] The second acquisition module is used to train a coordinated control model of pressure and displacement using a support vector machine algorithm based on smoothed pressure data and displacement time series data, determine the dynamic mapping relationship between pressure and displacement, and obtain an optimized combination of control parameters.

[0010] The third acquisition module is used to adjust the opening of the servo valve of the hydraulic system in real time and, in combination with the optimized control parameter combination, correct the pressure output to obtain a stable pressure application curve if the control deviation of the dynamic mapping relationship between pressure and displacement exceeds the preset threshold.

[0011] The fourth acquisition module is used to predict the distribution of material density based on stable pressure application curves and displacement time series data using neural network algorithms, determine whether there are density non-uniform regions, and obtain the predicted density distribution results.

[0012] The fifth acquisition module is used to adjust the demolding speed and mold temperature at the beginning of the demolding process based on the predicted density distribution, and to optimize the demolding parameters using a fuzzy control algorithm to obtain a demolding control strategy that reduces the risk of cracking.

[0013] Furthermore, the first acquisition module includes:

[0014] The first acquisition unit is used to collect dynamic parameters from the hydraulic system and mold equipment in real time through a sensor array to obtain an initial parameter dataset;

[0015] The generation unit is used to perform noise reduction processing on the collected pressure and displacement information based on the initial parameter dataset and a data fusion algorithm to generate a smoothed parameter set after preliminary processing.

[0016] The first determining unit is used to detect outliers in the pressure sequence and displacement sequence using a preset threshold range for the smoothing parameter set. If outlier data exceeding the threshold is detected, it is interpolated and corrected to determine the corrected parameter sequence.

[0017] The second acquisition unit is used to acquire the corrected parameter sequence, and apply time series analysis methods to extract the trend of the pressure sequence and displacement sequence to obtain parameter change trend data.

[0018] The first judgment unit is used to monitor the operating status of the hydraulic system and mold equipment in real time through parameter change trend data. If the trend data deviates from the preset normal range, an abnormal status flag is triggered to judge potential operating risks.

[0019] The third acquisition unit is used to predict the pressure and displacement trends in the future period based on the state anomaly markers and historical time series data, using the Kalman filter algorithm to obtain the predicted trend dataset.

[0020] The second determining unit is used to generate a dynamic adjustment strategy for the predicted trend dataset. If the predicted trend shows that the abnormality continues, the control parameters of the hydraulic system are fine-tuned to determine the optimized operating parameter configuration.

[0021] Furthermore, the dynamic parameters in the first acquisition unit cover multi-dimensional information, including pressure information and displacement information.

[0022] Furthermore, in the first determining unit, the anomaly detection of pressure data points is calculated using the following formula:

[0023]

[0024] in, Indicates the first Anomaly detection flags for individual pressure data points. Indicates the first One pressure measurement value, This represents the mean of the pressure series. The standard deviation of the pressure series is represented. Indicates the threshold multiplier coefficient;

[0025] The displacement value after interpolation correction is:

[0026]

[0027] in, Indicates the first Corrected displacement values ​​at each location This represents the previous normal displacement data point. This represents the next normal displacement data point. Indicates the interpolation correction factor;

[0028] Smoothing of the parameter sequence by weighted averaging is achieved using the following formula:

[0029]

[0030] in, Indicates the first The smoothing parameter value at each position. Indicates the length of the smooth window. Indicates the first Each weighting coefficient This represents the data points in the original parameter sequence.

[0031] Furthermore, in the first judgment unit, the overall deviation of the hydraulic system and mold equipment parameters from the normal range is calculated using the following formula:

[0032]

[0033] in, Indicates the degree of deviation of the parameter trend. Indicates the number of monitoring parameters. Indicates the first The current values ​​of the parameters, This indicates the preset normal mean value of the parameter;

[0034] The risk level of a single parameter deviating from the normal range is calculated using the following formula:

[0035]

[0036] in, Indicates the first Risk assessment index of each parameter, Indicates the first Real-time monitoring values ​​of each parameter This indicates the reference standard value for this parameter. This represents the upper limit of the parameter's normal range. This indicates the lower limit of the parameter's normal range.

[0037] The status exception flag is triggered using the following formula:

[0038]

[0039] in, Indicates an abnormal state marker signal. This indicates the number of sampling points for the trend data. Indicates the first Trend data values ​​at each moment. This represents the baseline value for the trend under normal conditions. This indicates the threshold for triggering an anomaly.

[0040] Furthermore, the second acquisition module includes:

[0041] The fourth acquisition unit is used to match the changes in the pressure data and displacement time series data over time using a data comparison tool based on the acquired pressure data and displacement time series data. For the matching results, a preset threshold range is used for comparison. If the deviation exceeds the threshold range, the corresponding time point is marked to obtain the marked deviation dataset.

[0042] The third determining unit is used to perform numerical correction on the marked time points using a data interpolation tool for the marked deviation dataset, obtain the corrected pressure data and displacement data set, and use a data integration tool to rearrange the corrected pressure data and displacement data set according to the time series to determine the integrated time series dataset.

[0043] The fifth acquisition unit is used to construct a dynamic correspondence between pressure data and displacement time series data based on the integrated time series dataset using a data mapping tool. For the nonlinear changes in the correspondence, a feature extraction tool is used to process them to obtain a feature-based mapping dataset.

[0044] The fourth determining unit is used to compare the control parameters item by item with the parameter adjustment tool for the feature-based mapping relationship dataset. If the parameter combination does not match the preset operating state, the parameter values ​​are iteratively updated to determine the adjusted control parameter combination.

[0045] Furthermore, the third acquisition module includes:

[0046] The sixth acquisition unit is used to acquire real-time pressure data and displacement time series data from the hydraulic system, and to perform point-by-point matching of the pressure data and displacement time series data using time series tools to determine the dynamic correspondence between the pressure data and displacement time series data in the time series, thereby obtaining the initial mapping dataset.

[0047] The seventh acquisition unit is used to calculate the control deviation between pressure data and displacement data using a deviation calculation tool for the initial mapping dataset. If the deviation exceeds a preset threshold, the time points exceeding the threshold are marked using a marking tool to obtain the marked deviation dataset.

[0048] The fifth determining unit is used to adjust the opening degree of the hydraulic system servo valve through the servo valve control tool based on the marked deviation dataset, and to iteratively update the control parameters in combination with the parameter optimization tool to determine the adjusted servo valve opening parameter combination.

[0049] The eighth acquisition unit is used to process the adjusted servo valve opening parameter combination through the pressure output tool, generate corrected pressure output data, and use a curve fitting tool to smooth the corrected pressure output data to obtain a stable pressure application curve.

[0050] Furthermore, the fourth acquisition module includes:

[0051] The ninth acquisition unit is used to acquire corresponding data points from the stable pressure application curve and displacement sequence, and to perform point-by-point correspondence processing on the time axis using a data mapping tool to generate an initial distribution association dataset, determine the dynamic relationship between pressure application and displacement sequence, and obtain preliminary distribution characteristic data.

[0052] The tenth acquisition unit is used to extract material properties in segments based on the preliminary distribution characteristic data using state analysis tools, and compare the characteristic values ​​of each region with the region judgment tool. If the characteristic value of a certain region deviates from the preset threshold range, the region is marked as a potential uneven region, and the marked region dataset is obtained.

[0053] The eleventh acquisition unit is used to perform detailed calculations on the density distribution of the labeled region dataset using a depth prediction tool, generate a corresponding density distribution prediction map, determine whether there are areas with uneven density, and obtain the predicted distribution state data.

[0054] The sixth determination unit is used to compare the predicted distribution status data in multiple dimensions using distribution assessment tools, obtain the assessment indicators for each region, and verify them one by one using unevenness detection tools to determine the final density distribution prediction results.

[0055] Furthermore, the fifth acquisition module includes:

[0056] The twelfth acquisition unit is used to analyze the density values ​​of the regional data point by point using a data comparison tool based on the regional data obtained from the density distribution prediction results. If the density value exceeds the preset threshold range, it is marked as a high-risk area, and the marked distribution dataset is obtained.

[0057] The seventh determination unit is used to associate and match the density data of high-risk areas with the initial values ​​of demolding speed and mold temperature using a parameter mapping tool for the labeled distribution dataset, generate a preliminary parameter adjustment plan, and determine the adjustment range for each area.

[0058] The thirteenth acquisition unit is used to dynamically optimize the preliminary parameter adjustment scheme using fuzzy control tools, extract the optimal demolding speed and mold temperature combination from the adjustment range, and obtain the optimized parameter configuration data.

[0059] The second judgment unit is used to match the optimized parameter configuration data with the actual demolding process in real time through parameter application tools, dynamically adjust the speed and temperature according to the density distribution characteristics, determine whether the preset conditions for reducing crack risk are met, and obtain the final control strategy data.

[0060] Furthermore, in the thirteenth acquisition unit, the parameter mapping tool uses the MATLAB / Simulink Fuzzy Logic Toolbox.

[0061] The beneficial effects achieved by this invention are as follows:

[0062] This invention discloses a hydraulic forming control system for magnetic materials. It employs a first, second, third, fourth, and fifth acquisition module to acquire pressure and displacement data via sensors. A dynamic mapping model is established using data fusion and machine learning algorithms to adjust hydraulic system parameters in real time to obtain a stable pressure curve. Simultaneously, a neural network is used to predict material density distribution and optimize demolding process parameters accordingly. This method effectively solves problems such as unstable pressure control, uneven material density, and cracking during demolding in hydraulic forming, achieving intelligent control of the forming process. Through the synergistic application of multiple algorithms, this invention significantly improves the accuracy and product quality of hydraulic forming, which is of great significance for enhancing the automation and intelligence level of manufacturing. The specific beneficial effects of the magnetic material hydraulic forming control system disclosed in this invention are as follows:

[0063] I. Data preprocessing improves the accuracy of basic control systems

[0064] 1. Real-time and accurate acquisition of dynamic parameters: By covering the hydraulic system and mold equipment with a sensor array, key parameters such as pressure and displacement are acquired in real time, avoiding the error of single-point sampling and ensuring data integrity.

[0065] 2. Data fusion, denoising, and smoothing: The data fusion algorithm is used to preprocess dynamic parameters, eliminate signal fluctuations and noise interference, and generate smooth pressure-displacement time series data, providing reliable input for subsequent control models and improving the reliability of basic data (for example, pressure fluctuation errors can be reduced by more than 30%).

[0066] II. Pressure-Displacement Co-control Optimizes Molding Process

[0067] 1. Support Vector Machine Modeling for Dynamic Mapping: By training a pressure-displacement collaborative control model through algorithms, the nonlinear dynamic relationship between the two is accurately captured, breaking the limitations of traditional fixed parameter control and adapting to the needs of different material properties and forming stages (such as improving the pressure-displacement matching accuracy of complex magnetic core structures by 40%).

[0068] 2. Real-time deviation correction for stable pressure output: When the control deviation exceeds the threshold, the parameter combination is adjusted and optimized in real time through the servo valve opening to generate a stable pressure application curve, avoiding uneven material deformation caused by sudden pressure changes and improving molding consistency (pressure fluctuation range can be controlled within ±2%).

[0069] III. Density Distribution Prediction for Internal Defect Prevention

[0070] 1. Neural network predicts areas of uneven density: Based on pressure-displacement data and material properties, the spatial distribution of material density after molding is predicted by a neural network algorithm, and local loose or over-compressed areas are identified in advance (such as axial density deviation early warning accuracy of over 90%).

[0071] 2. Proactive quality control: By predicting the density distribution, process parameters can be adjusted during the molding process (rather than after testing), reducing differences in magnetic properties or structural defects caused by uneven density, and improving the uniformity of material properties.

[0072] IV. Optimizing demolding parameters to reduce the risk of cracking

[0073] 1. Dynamic adjustment of demolding strategy based on density prediction: During the demolding stage, the demolding speed (such as reducing the speed in high-density areas to avoid pulling) and mold temperature are adjusted in a targeted manner according to the density distribution results (such as reducing the speed in high-density areas to avoid pulling) to achieve "demolding on demand".

[0074] 2. Fuzzy control algorithm improves demolding flexibility: By optimizing the combination of demolding parameters through fuzzy logic, it can adapt to the complex needs of different density distribution scenarios, effectively reduce the risk of cracks caused by demolding stress (the crack occurrence rate can be reduced by more than 50%), and improve the yield of finished products.

[0075] V. System-level benefits: Intelligent processes and improved production efficiency

[0076] 1. Full closed-loop control improves automation level: From parameter acquisition and model optimization to demolding control, a complete closed loop is formed, reducing manual intervention, lowering the operation threshold, and making it suitable for mass automated production.

[0077] 2. Material and energy consumption optimization: Precise control reduces material waste (e.g., scrap material loss is reduced by 25%), and stable pressure curves and demolding strategies can reduce equipment energy consumption (energy consumption is reduced by 15-20% year-on-year), combining quality and cost advantages. Attached Figure Description

[0078] Figure 1 This is a functional block diagram of an embodiment of a magnetic material hydraulic forming control system according to the present invention;

[0079] Figure 2 for Figure 1 A functional module diagram of an embodiment of the first acquisition module shown in the figure;

[0080] Figure 3 for Figure 1 A functional module diagram of one embodiment of the second acquisition module shown in the figure;

[0081] Figure 4 for Figure 1 A functional module diagram of one embodiment of the third acquisition module shown;

[0082] Figure 5 for Figure 1 A functional module diagram of one embodiment of the fourth acquisition module shown;

[0083] Figure 6 for Figure 1 The diagram shows a functional module schematic of one embodiment of the fifth acquisition module.

[0084] Explanation of icon numbers:

[0085] 10. First Acquisition Module; 20. Second Acquisition Module; 30. Third Acquisition Module; 40. Fourth Acquisition Module; 50. Fifth Acquisition Module; 11. First Acquisition Unit; 12. Generation Unit; 13. First Determination Unit; 14. Second Acquisition Unit; 15. First Judgment Unit; 16. Third Acquisition Unit; 17. Second Determination Unit; 21. Fourth Acquisition Unit; 22. Third Determination Unit; 23. Fifth Acquisition Unit; 24. Fourth Determination Unit; 31. Sixth Acquisition Unit; 32. Seventh Acquisition Unit; 33. Fifth Determination Unit; 34. Eighth Acquisition Unit; 41. Ninth Acquisition Unit; 42. Tenth Acquisition Unit; 43. Eleventh Acquisition Unit; 44. Sixth Determination Unit; 51. Twelfth Acquisition Unit; 52. Seventh Determination Unit; 53. Thirteenth Acquisition Unit; 54. Second Judgment Unit. Detailed Implementation

[0086] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0087] like Figure 1As shown, the first embodiment of the present invention proposes a hydraulic forming control system for magnetic materials, including a first acquisition module 10, a second acquisition module 20, a third acquisition module 30, a fourth acquisition module 40, and a fifth acquisition module 50. The first acquisition module 10 is used to acquire dynamic parameters from the hydraulic system and mold equipment through a sensor array, and preprocess the dynamic parameters using a data fusion algorithm to obtain smooth pressure data and displacement time series data. The second acquisition module 20 is used to train a collaborative control model of pressure and displacement using a support vector machine algorithm based on the smooth pressure data and displacement time series data, determine the dynamic mapping relationship between pressure and displacement, and obtain an optimized combination of control parameters. The third acquisition module 50... The first module 30 is used to adjust the opening of the servo valve of the hydraulic system in real time and correct the pressure output by combining optimized control parameters if the control deviation of the dynamic mapping relationship between pressure and displacement exceeds a preset threshold, so as to obtain a stable pressure application curve. The second module 40 is used to predict the distribution state of material density by using a neural network algorithm based on the stable pressure application curve and displacement time series data, to determine whether there are areas of uneven density, and to obtain the predicted density distribution. The third module 50 is used to adjust the demolding speed and mold temperature at the beginning of the demolding process based on the predicted density distribution, and to optimize the demolding parameters by using a fuzzy control algorithm to obtain a demolding control strategy that reduces the risk of cracking.

[0088] In the hydraulic forming of soft magnetic material cores, the sensor array in the first acquisition module 10 collects dynamic data of hydraulic cylinder pressure (10-200MPa) and pressure head displacement (0-50mm) in real time. After processing by the Kalman filter fusion algorithm, the measurement noise caused by hydraulic system vibration (amplitude ±5MPa) and mechanical clearance (±0.1mm) can be eliminated, and a smooth pressure-displacement curve is generated, providing reliable input parameters for the subsequent pressure-displacement collaborative control model, ultimately achieving a core density uniformity improvement of more than 15%.

[0089] In the second acquisition module 20, the support vector machine algorithm uses the regression characteristics (SVR, Support Vector Regression) of SVM (Support Vector Machine) to fit the dynamic relationship between pressure and displacement, capturing the nonlinear coupling characteristics that traditional linear models cannot describe. The collaborative control model achieves collaborative regulation of pressure and displacement during the forming process by modeling the time-varying mapping relationship between pressure and displacement, avoiding forming defects caused by single parameter control (such as insufficient displacement due to insufficient pressure or material damage due to overpressure).

[0090] The third acquisition module 30 is used to determine whether the control deviation of the dynamic mapping relationship between pressure and displacement exceeds a preset threshold. If the control deviation of the dynamic mapping relationship between pressure and displacement exceeds the preset threshold, the pressure output is corrected by adjusting the opening of the servo valve of the hydraulic system in real time and combining the optimized control parameter combination to obtain a stable pressure application curve.

[0091] In the fourth acquisition module 40, the neural network algorithm refers to a computational model constructed using multi-layer nonlinear mapping units (such as fully connected layers, convolutional layers, and recurrent layers) to achieve nonlinear mapping prediction by learning the implicit relationship between pressure-displacement data and material density.

[0092] Density distribution refers to the spatial distribution of material density within the mold cavity, typically represented by a three-dimensional matrix (e.g., density values ​​along the mold axis and radial direction). Uneven density regions refer to areas where the density of a magnetic material deviates from the overall average during hydroforming due to factors such as pressure distribution, displacement rate, or material filling. When the density value in this region deviates from the overall average density of the material beyond a preset threshold, it directly affects the uniformity of the magnetic properties and mechanical strength of the finished product. Uneven density regions are generally characterized by localized density deviations from the average by more than ±5%.

[0093] In the fifth acquisition module 50, the demolding control strategy for reducing crack risk refers to a systematic method that, during the demolding stage of hydraulic forming of magnetic materials, dynamically adjusts the demolding speed and mold temperature by analyzing the density distribution prediction results in real time (such as the location of local high-density areas and low-density areas), and optimizes the parameter combination using a fuzzy control algorithm to eliminate the risk of cracking caused by stress concentration. This strategy uses density deviation data as input and generates optimal demolding parameters through fuzzy logic mapping to achieve a closed-loop adjustment of "prediction-control-feedback".

[0094] Furthermore, the magnetic material hydraulic forming control system provided in this embodiment includes a first acquisition module 10 comprising a first acquisition unit 11, a generation unit 12, a first determination unit 13, a second acquisition unit 14, a first judgment unit 15, a third acquisition unit 16, and a second determination unit 17. The first acquisition unit 11 is used to collect dynamic parameters in real time from the hydraulic system and mold equipment via a sensor array to obtain an initial parameter dataset. The generation unit 12 is used to perform noise reduction processing on the collected pressure and displacement information using a data fusion algorithm based on the initial parameter dataset, generating a pre-processed smooth parameter set. The first determination unit 13 is used to perform outlier detection on the pressure and displacement sequences using a preset threshold range for the smooth parameter set. If outlier data exceeding the threshold is detected, it is interpolated and corrected to determine the corrected value. The system consists of three parts: a first acquisition unit 14 and a second determination unit 17. The first acquisition unit 15 is used to acquire the corrected parameter sequence and extract the trend of the pressure and displacement sequences using time series analysis to obtain parameter change trend data. The second determination unit 16 is used to monitor the operating status of the hydraulic system and mold equipment in real time using the parameter change trend data. If the trend data deviates from the preset normal range, an abnormal status marker is triggered to determine potential operating risks. The third acquisition unit 16 is used to predict the pressure and displacement trends in the future period based on the abnormal status marker and historical time series data using the Kalman filter algorithm to obtain a predicted trend dataset. The second determination unit 17 is used to generate a dynamic adjustment strategy for the predicted trend dataset. If the predicted trend shows a continuous abnormality, the control parameters of the hydraulic system are fine-tuned to determine the optimized operating parameter configuration.

[0095] In real-time monitoring scenarios of hydraulic systems and mold equipment, the sensor array in the first acquisition unit 11 is used to collect multi-dimensional dynamic parameters such as pressure and displacement. Assuming a hydraulic stamping equipment is running, the sensor collects data once per second to obtain an initial dataset, where the pressure value ranges from 50 to 200 MPa and the displacement value ranges from 0.5 to 5.0 mm.

[0096] In generation unit 12, the data fusion algorithm can perform denoising on multi-sensor data using a weighted averaging method, reducing the impact of noise caused by vibration or electromagnetic interference, and generating a smooth parameter set. This effectively improves data reliability and lays the foundation for subsequent analysis.

[0097] In the first determining unit 13, for the smoothing parameter set, the preset pressure threshold range is 60 to 180 MPa, and the displacement threshold is 0.8 to 4.5 mm. If a pressure value of 190 MPa is detected at a certain moment, exceeding the threshold, a correction value is estimated using two consecutive normal data points through linear interpolation, ultimately adjusting the outlier value to 175 MPa. This correction method avoids interference from outlier data on subsequent trend analysis, ensuring the continuity and accuracy of the parameter sequence.

[0098] Anomaly detection of pressure data points is calculated using the following formula:

[0099] (1)

[0100] In formula (1), Indicates the first Anomaly detection flags for individual pressure data points. Indicates the first One pressure measurement value, This represents the mean of the pressure series. The standard deviation of the pressure series is represented. This represents the threshold multiplier. When the pressure value deviates from the mean by more than [a certain amount], [the threshold multiplier is applied]. Values ​​exceeding one standard deviation are marked as outliers.

[0101] The displacement value after interpolation correction is:

[0102] (2)

[0103] In formula (2), Indicates the first Corrected displacement values ​​at each location This represents the previous normal displacement data point. This represents the next normal displacement data point. This represents the interpolation correction coefficient. Formula (2) corrects abnormal displacement data using a linear interpolation method.

[0104] Smoothing of the parameter sequence by weighted averaging is achieved using the following formula:

[0105] (3)

[0106] In formula (3), Indicates the first The smoothing parameter value at each position. Indicates the length of the smooth window. Indicates the first Each weighting coefficient This represents the data points in the original parameter sequence.

[0107] The second acquisition unit 14, after acquiring the corrected parameter sequence, can use the sliding window method to extract the trend data of pressure and displacement for time series analysis. For example, assuming the pressure value gradually increased from 70 MPa to 90 MPa over the past hour, this might reflect an increase in the load on the hydraulic system.

[0108] The first judgment unit 15 monitors the equipment's operating status in real time through parameter change trend data. If the pressure trend exceeds the normal range of 80 to 100 MPa, an abnormal status marker is triggered, indicating a potential overload risk. This monitoring method helps to detect problems in a timely manner and reduce the probability of equipment damage.

[0109] The overall deviation of the hydraulic system and mold equipment parameters from the normal range is calculated using the following formula:

[0110] (4)

[0111] In formula (4), Indicates the degree of deviation of the parameter trend. Indicates the number of monitoring parameters. Indicates the first The current values ​​of the parameters, This indicates the default normal mean value of the parameter.

[0112] The risk level of a single parameter deviating from the normal range is calculated using the following formula:

[0113] (5)

[0114] In formula (5), Indicates the first Risk assessment index of each parameter, Indicates the first Real-time monitoring values ​​of each parameter This indicates the reference standard value for this parameter. This represents the upper limit of the parameter's normal range. This indicates the lower limit of the normal range of the parameter.

[0115] The status exception flag is triggered using the following formula:

[0116] (6)

[0117] In formula (6), Indicates an abnormal state marker signal. This indicates the number of sampling points for the trend data. Indicates the first Trend data values ​​at each moment. This represents the baseline value for the trend under normal conditions. This indicates the threshold for triggering an anomaly. When the average deviation of the trend data exceeds the preset threshold, output 1 to trigger an anomaly flag; otherwise, output 0 to indicate a normal state.

[0118] The third acquisition unit 16 combines anomaly markers and historical data, using a Kalman filter algorithm to predict pressure and displacement trends for the next 30 minutes. Assuming the current pressure is 95 MPa, historical data shows relatively small fluctuations, but the prediction indicates the pressure may continue to rise to 105 MPa, exceeding the safe range. The Kalman filter, by dynamically adjusting the prediction model, effectively addresses data uncertainty, improves prediction accuracy, and provides a reliable basis for subsequent decision-making.

[0119] The second determining unit 17, based on the predicted trend dataset, generates a dynamic adjustment strategy if the abnormal pressure persists. Assuming the predicted pressure remains above 100 MPa, the pressure can be controlled within a safe range by reducing the hydraulic pump output power by 5%, thus determining the optimized operating parameter configuration. This fine-tuning strategy effectively extends equipment life, reduces maintenance costs, and simultaneously ensures production efficiency and safety. Through the collaborative work of these multiple stages, from data acquisition to predictive adjustment, a closed-loop monitoring system is formed to ensure the stable operation of the hydraulic system and mold equipment.

[0120] Furthermore, in the magnetic material hydraulic forming control system provided in this embodiment, the second acquisition module 20 includes a fourth acquisition unit 21, a third determination unit 22, a fifth acquisition unit 23, and a fourth determination unit 24. The fourth acquisition unit 21 is used to perform point-by-point matching of the pressure data and displacement time series data over time using a data comparison tool, based on the acquired pressure data and displacement time series data. For the matching results, a preset threshold range is used for comparison. If the deviation exceeds the threshold range, the corresponding time point is marked to obtain a marked deviation dataset. The third determination unit 22 is used to perform numerical correction on the marked time points using a data interpolation tool to obtain the corrected deviation dataset. The pressure and displacement data sets are then rearranged according to time series using a data integration tool to determine the integrated time series dataset. The fifth acquisition unit 23 is used to construct a dynamic correspondence between pressure data and displacement time series data using a data mapping tool based on the integrated time series dataset. For the nonlinear changes in the correspondence, a feature extraction tool is used to process them to obtain a characteristic mapping relationship dataset. The fourth determination unit 24 is used to compare the control parameters item by item using a parameter adjustment tool on the characteristic mapping relationship dataset. If the parameter combination does not match the preset operating state, the parameter values ​​are iteratively updated to determine the adjusted control parameter combination.

[0121] In the operation monitoring scenario of hydraulic systems and mold equipment, the fourth acquisition unit 21, for processing smoothed pressure and displacement data, first needs to perform point-by-point matching of the changing patterns over time. Assuming that during the operation of a hydraulic stamping equipment, pressure and displacement data are collected once per minute, with pressure values ​​ranging from 60 to 150 MPa and displacement values ​​ranging from 0.5 to 4.0 mm, data comparison tools can reveal that at a certain time point, the pressure value is 140 MPa, while the displacement value is 3.8 mm, a significant deviation compared to historical patterns. Using preset threshold ranges, such as a pressure deviation allowance of ±10 MPa and a displacement deviation allowance of ±0.3 mm, if these ranges are exceeded, the time point is marked, forming a deviation dataset. This point-by-point matching method helps to quickly locate abnormal fluctuations.

[0122] The third determining unit 22 uses a data interpolation tool to correct outliers in the labeled deviation dataset. For example, if the pressure value at a labeled time point is 155 MPa, exceeding the upper threshold, and the pressure values ​​at subsequent time points are 148 MPa and 150 MPa respectively, the interpolation tool estimates a correction value of 149 MPa. Displacement data can be adjusted similarly to ensure data continuity.

[0123] In the fifth acquisition unit 23, the corrected dataset is then rearranged according to time series using a data integration tool to form an integrated time series dataset, providing a complete foundation for subsequent analysis. For example, data mapping tools can be useful when constructing a dynamic correspondence between pressure and displacement data. Suppose that during a certain period, the pressure value gradually increases from 80 MPa to 120 MPa, and the displacement value changes from 1.2 mm to 2.5 mm, exhibiting a non-linear relationship. Through feature extraction tools, this non-linear change can be extracted into key feature points, such as inflection points or intervals with large rates of change, forming a characteristic mapping dataset. This processing method facilitates the rapid identification of key changing trends during subsequent parameter adjustments.

[0124] The dynamic mapping function between pressure data and displacement data is:

[0125] (7)

[0126] In formula (7), A function representing the dynamic mapping relationship between pressure data and displacement data. Indicates the first A stress time series at time 1 The value, Indicates the first The displacement time series at time 1 The value, Indicates the first Weighting coefficients at each time point This represents the linear coupling coefficient between pressure and displacement. Indicates the first The standard deviation parameter at each time point This indicates the total length of the time series.

[0127] The feature extraction function for the nonlinear variation part is:

[0128] (8)

[0129] In formula (8), The feature extraction function represents the nonlinear variation component. This represents the original input data variable. Indicates the first The amplitude coefficient of each hyperbolic tangent term, Indicates the first The slope parameter of each hyperbolic tangent term. Indicates the first The offset of each hyperbolic tangent term. This represents the amplitude coefficient of the m-th sine term. Indicates the first Frequency parameters of each sinusoidal term, Indicates the first The phase angle of a sinusoidal term, This represents the total number of hyperbolic tangent terms. This represents the total number of sine terms.

[0130] The characteristic mapping relationship dataset is as follows:

[0131] (9)

[0132] In formula (9), This represents a dataset of characteristic mapping relationships. Indicates the first The stress feature vector of each sample Indicates the first The displacement feature vector of each sample Indicates the first The feature correlation coefficient of each sample Indicates the first The original stress data, Indicates the first One original displacement data, This represents the weight of the basis function for extracting the l-th pressure feature. Indicates the first One pressure feature extraction basis function Indicates the first The weights of the basis function are extracted from each displacement feature. Indicates the first One displacement feature extraction basis function This represents the scaling factor used in the correlation calculation. This represents the total number of pressure characteristic basis functions. This represents the total number of displacement characteristic basis functions.

[0133] The fourth determining unit 24 uses a parameter adjustment tool to optimize control parameters for the characteristic mapping dataset. For example, assuming the current hydraulic system's operating parameter combination has a pressure control upper limit of 130 MPa, but according to characteristic data analysis, the pressure frequently approaches this value during actual operation, potentially leading to system instability. Through iterative updates, the upper limit is adjusted to 135 MPa, while related flow parameters are fine-tuned to ensure the operating state matches the preset target. The adjusted parameter combination better adapts to actual working conditions. Through the collaborative processing of these multiple stages, from data comparison to parameter adjustment, a complete analysis chain is formed. Each stage closely revolves around the operational needs of the hydraulic system and mold equipment, ensuring that each step of data processing provides reliable support for subsequent decision-making. This approach effectively improves the system's adaptability and stability in practical applications.

[0134] Furthermore, the magnetic material hydraulic forming control system provided in this embodiment includes a third acquisition module 30 comprising a sixth acquisition unit 31, a seventh acquisition unit 32, a fifth determination unit 33, and an eighth acquisition unit 34. The sixth acquisition unit 31 is used to acquire real-time pressure data and displacement time-series data from the hydraulic system, and to perform point-by-point matching of the pressure data and displacement time-series data using a time-series tool to determine the dynamic correspondence between the pressure data and displacement time-series data in the time series, thereby obtaining an initial mapping dataset. The seventh acquisition unit 32 is used to calculate the pressure data and displacement data using a deviation calculation tool for the initial mapping dataset. If the control deviation exceeds a preset threshold, the time point exceeding the threshold is marked by a marking tool to obtain a marked deviation dataset; the fifth determination unit 33 is used to adjust the opening of the hydraulic system servo valve by using a servo valve control tool based on the marked deviation dataset, and to iteratively update the control parameters by combining a parameter optimization tool to determine the adjusted servo valve opening parameter combination; the eighth acquisition unit 34 is used to process the adjusted servo valve opening parameter combination by using a pressure output tool to generate corrected pressure output data, and to smooth the corrected pressure output data by using a curve fitting tool to obtain a stable pressure application curve.

[0135] In the scenario of hydraulic system operation monitoring, the real-time acquisition of pressure and displacement time-series data in the sixth acquisition unit 31 is the foundation for building a dynamic control model. Hydraulic systems are commonly used in industrial stamping equipment, and the precise matching of pressure and displacement directly affects processing accuracy. Assuming that during the operation of a stamping machine, pressure and displacement data are collected once per second, with pressure values ​​ranging from 50 to 160 MPa and displacement values ​​ranging from 0.2 to 3.5 mm, time-series tools can map these data point-to-point to form an initial mapping dataset, facilitating subsequent analysis of the dynamic relationship between the two.

[0136] In the seventh acquisition unit 32, the application of a deviation calculation tool can help identify outliers in the initial mapping dataset. For example, suppose at a certain point in time, the pressure is 158 MPa and the displacement is 3.4 mm, while historical data shows that the pressure at that displacement is usually around 150 MPa, exceeding a preset threshold of 10 MPa. In this case, the marking tool will mark this point in time as an outlier, forming a marked deviation dataset. This method facilitates the rapid location of aspects requiring adjustment.

[0137] In adjusting the hydraulic system, the servo valve control tool in the fifth determination unit 33 plays a crucial role. Based on the labeled deviation dataset, it is assumed that the current servo valve opening is 60%, leading to excessive pressure. By analyzing data from previous and subsequent time points, it is found that adjusting the opening to 55% might be more appropriate. Combined with the parameter optimization tool, the control parameters are iteratively updated to ultimately determine a new combination of opening parameters. This adjustment method effectively balances the system's operating state.

[0138] In the eighth acquisition unit 34, the pressure output tool processes the adjusted servo valve opening parameters when generating the corrected pressure output data. Assuming the adjusted pressure value decreases from 158 MPa to 152 MPa, approaching the target range, this processing helps reduce the risk of system overload while ensuring stability during processing. Specifically, the curve fitting tool smooths the corrected pressure output data, eliminating small fluctuations. Assuming the corrected pressure data exhibits slight oscillations over a period, the fitting tool generates a smooth pressure application curve, making pressure changes more continuous. This smoothing helps the equipment avoid shocks caused by sudden changes during operation, improving overall operational stability. In this embodiment, from initial data acquisition to final curve generation, each step closely revolves around the actual needs of the hydraulic system. For example, determining the dynamic correspondence provides a data foundation for subsequent deviation analysis, while deviation marking and parameter adjustment lay the foundation for pressure output correction, ultimately forming a stable control curve through smoothing. This progressive approach effectively improves the adaptability and reliability of the hydraulic system in industrial applications.

[0139] Furthermore, the magnetic material hydraulic forming control system provided in this embodiment includes a fourth acquisition module 40 comprising a ninth acquisition unit 41, a tenth acquisition unit 42, an eleventh acquisition unit 43, and a sixth determination unit 44. The ninth acquisition unit 41 is used to acquire corresponding data points from a stable pressure application curve and displacement sequence, perform point-by-point correspondence processing on the time axis using a data mapping tool, generate an initial distribution association dataset, determine the dynamic relationship between pressure application and displacement sequence, and obtain preliminary distribution characteristic data. The tenth acquisition unit 42 is used to extract material properties in segments based on the preliminary distribution characteristic data using a state analysis tool, and combine regional... The judgment tool compares the characteristic values ​​of each region. If the characteristic value of a certain region deviates from the preset threshold range, the region is marked as a potential uneven region, and the marked region dataset is obtained. The eleventh acquisition unit 43 is used to perform detailed calculations on the density distribution of the marked region dataset using a depth prediction tool, generate a corresponding density distribution prediction map, determine whether there are uneven density regions, and obtain the predicted distribution state data. The sixth determination unit 44 is used to perform multi-dimensional comparisons on the predicted distribution state data using a distribution evaluation tool, obtain the evaluation index of each region, and perform verification one by one with the unevenness detection tool to determine the final density distribution prediction result.

[0140] In the scenario of hydraulic system operation monitoring, the ninth acquisition unit 41 obtains data points from the stable pressure application curve and displacement sequence, which is the foundation for constructing dynamic relationships. Assuming that during the operation of an industrial stamping equipment, the pressure application curve has reached a stable state through prior adjustments, with the pressure value fluctuating between 50 and 160 MPa, and the displacement sequence varying within the range of 0.2 to 3.5 mm. The data mapping tool can map these data points one-to-one onto the time axis, forming an initial distributed correlation dataset. For example, within a certain time period, when the pressure value is 120 MPa, the corresponding displacement value is 2.5 mm. Through point-by-point mapping, the dynamic relationship between the two can be preliminarily determined, generating distribution characteristic data to provide a basis for subsequent analysis.

[0141] In the tenth acquisition unit 42, based on the preliminary distribution characteristic data, the state analysis tool can be used to extract material properties in segments. Assuming the material exhibits different stress responses in different displacement regions, the tool divides the displacement sequence into multiple intervals and extracts the characteristic data for each interval. Combined with a region judgment tool, if the characteristic value of a certain region deviates from a preset threshold—for example, at a displacement of 2.0 mm, if the characteristic value exceeds the normal range by 10%—it is marked as a potential non-uniform region. This segmented extraction and marking method helps to quickly locate regions that may affect processing accuracy, forming a marked region dataset.

[0142] In the eleventh acquisition unit 43, when processing the labeled region dataset, the depth prediction tool can perform detailed calculations on the density distribution. Assuming a potentially uneven region is within the labeled area, the tool compares historical and current data to predict possible trends in the density distribution and generates a density distribution prediction map. For example, the prediction map might show a 5% deviation in the density distribution within a displacement range of 2.0 to 2.5 millimeters, thus determining the existence of uneven density regions. This prediction method provides intuitive data support for subsequent verification.

[0143] The sixth determining unit 44 uses a distribution assessment tool to perform multi-dimensional comparisons of the predicted distribution data to obtain assessment indicators for each region. Assuming that in the predicted density unevenness region, the assessment indicators show that the stability of this region is 15% lower than other regions, the density distribution prediction result is finally determined by verifying each region using an unevenness detection tool. This multi-dimensional comparison and verification method improves the comprehensiveness and reliability of the analysis, providing a precise basis for subsequent adjustments to the hydraulic system.

[0144] Furthermore, the magnetic material hydraulic forming control system provided in this embodiment includes a fifth acquisition module 50 comprising a twelfth acquisition unit 51, a seventh determination unit 52, a thirteenth acquisition unit 53, and a second judgment unit 54. The twelfth acquisition unit 51 is used to analyze the density values ​​of the regional data obtained from the density distribution prediction results point-by-point using a data comparison tool. If the density value exceeds a preset threshold range, it is marked as a high-risk area, thus obtaining a marked distribution dataset. The seventh determination unit 52 is used to, for the marked distribution dataset, use a parameter mapping tool to convert the density data of the high-risk areas... The initial values ​​of demolding speed and mold temperature are correlated and matched to generate a preliminary parameter adjustment scheme and determine the adjustment range of each region. The thirteenth acquisition unit 53 is used to dynamically optimize the preliminary parameter adjustment scheme using fuzzy control tools, extract the optimal combination of demolding speed and mold temperature from the adjustment range, and obtain the optimized parameter configuration data. The second judgment unit 54 is used to match the optimized parameter configuration data with the actual demolding process in real time through parameter application tools, dynamically adjust the speed and temperature according to the density distribution characteristics, judge whether the preset conditions for reducing crack risk are met, and obtain the final control strategy data.

[0145] In the scenario of industrial stamping equipment operation, the density distribution prediction results in the twelfth acquisition unit 51 provide a crucial basis for subsequent optimization. By analyzing the density values ​​of regional data point by point using a data comparison tool, potential problem areas can be accurately identified. For example, suppose that during a certain stamping process, the prediction result shows that the density value of a certain area reaches 2.3 g / cm³, while the preset threshold is 2.0 g / cm³, exceeding the threshold by 15%. In this case, the data comparison tool will mark this area as a high-risk area, forming a marked distribution dataset. It should be noted that the density exceeding the standard may be due to uneven stress on the material during stamping or fluctuations in die temperature, leading to local density anomalies. The marked dataset lays the foundation for subsequent parameter adjustments.

[0146] In the seventh determination unit 52, the parameter mapping tool correlates and matches the density data of high-risk areas with demolding speed and mold temperature to generate a preliminary parameter adjustment plan. For example, in a high-risk area, the current demolding speed is 5 mm / s and the mold temperature is 180℃. Through historical data analysis, the mapping tool finds that when the demolding speed is reduced to 4 mm / s and the mold temperature is adjusted to 175℃, the density value tends to be within the normal range. The preliminary plan sets an adjustment range for this area: demolding speed between 3.8 and 4.2 mm / s and mold temperature between 170 and 180℃. This correlation and matching ensures the targeted nature of parameter adjustments and provides data support for optimization.

[0147] In the thirteenth acquisition unit 53, when the fuzzy control tool dynamically optimizes the preliminary scheme, it comprehensively considers the stability of multiple parameter combinations. Assuming a high-risk area, the tool analyzes various combinations of demolding speed and mold temperature, finding that the combination of 4 mm / s and 175℃ exhibits high stability in historical data, reducing density deviation to within 5%. The fuzzy control tool then extracts this combination as the optimal parameter configuration data through dynamic iteration.

[0148] In the second judgment unit 54, the parameter application tool maps the optimized parameter configuration data to the actual demolding process in real time, dynamically adjusting the demolding speed and mold temperature. For example, during the operation of the stamping equipment, if the tool detects that the density value in a certain area is still too high, it will adjust the demolding speed to 4 mm / s in real time, maintain the mold temperature at 175℃, and use sensors to report changes in density distribution. Assuming that after adjustment, the density value drops to 2.05 g / cm³, meeting the preset condition for reducing crack risk, i.e., the deviation is controlled within 5%, the final generated control strategy data provides precise guidance for subsequent production. If extended to more complex working conditions, the core solution can combine multiple sets of historical data and analyze the density distribution characteristics under different stamping speeds through the parameter mapping tool to further refine the adjustment range.

[0149] The magnetic material hydraulic forming control system disclosed in this embodiment has the following advantages compared with the prior art:

[0150] I. Data preprocessing improves the accuracy of basic control systems

[0151] 1. Real-time and accurate acquisition of dynamic parameters: By covering the hydraulic system and mold equipment with a sensor array, key parameters such as pressure and displacement are acquired in real time, avoiding the error of single-point sampling and ensuring data integrity.

[0152] 2. Data fusion, denoising, and smoothing: The data fusion algorithm is used to preprocess dynamic parameters, eliminate signal fluctuations and noise interference, and generate smooth pressure-displacement time series data, providing reliable input for subsequent control models and improving the reliability of basic data (for example, pressure fluctuation errors can be reduced by more than 30%).

[0153] II. Pressure-Displacement Co-control Optimizes Molding Process

[0154] 1. Support Vector Machine Modeling for Dynamic Mapping: By training a pressure-displacement collaborative control model through algorithms, the nonlinear dynamic relationship between the two is accurately captured, breaking the limitations of traditional fixed parameter control and adapting to the needs of different material properties and forming stages (such as improving the pressure-displacement matching accuracy of complex magnetic core structures by 40%).

[0155] 2. Real-time deviation correction for stable pressure output: When the control deviation exceeds the threshold, the parameter combination is adjusted and optimized in real time through the servo valve opening to generate a stable pressure application curve, avoiding uneven material deformation caused by sudden pressure changes and improving molding consistency (pressure fluctuation range can be controlled within ±2%).

[0156] III. Density Distribution Prediction for Internal Defect Prevention

[0157] 1. Neural network predicts areas of uneven density: Based on pressure-displacement data and material properties, the spatial distribution of material density after molding is predicted by a neural network algorithm, and local loose or over-compressed areas are identified in advance (such as axial density deviation early warning accuracy of over 90%).

[0158] 2. Proactive quality control: By predicting the density distribution, process parameters can be adjusted during the molding process (rather than after testing), reducing differences in magnetic properties or structural defects caused by uneven density, and improving the uniformity of material properties.

[0159] IV. Optimizing demolding parameters to reduce the risk of cracking

[0160] 1. Dynamic adjustment of demolding strategy based on density prediction: During the demolding stage, the demolding speed (such as reducing the speed in high-density areas to avoid pulling) and mold temperature are adjusted in a targeted manner according to the density distribution results (such as reducing the speed in high-density areas to avoid pulling) to achieve "demolding on demand".

[0161] 2. Fuzzy control algorithm improves demolding flexibility: By optimizing the combination of demolding parameters through fuzzy logic, it can adapt to the complex needs of different density distribution scenarios, effectively reduce the risk of cracks caused by demolding stress (the crack occurrence rate can be reduced by more than 50%), and improve the yield of finished products.

[0162] V. System-level benefits: Intelligent processes and improved production efficiency

[0163] 1. Full closed-loop control improves automation level: From parameter acquisition and model optimization to demolding control, a complete closed loop is formed, reducing manual intervention, lowering the operation threshold, and making it suitable for mass automated production.

[0164] 2. Material and energy consumption optimization: Precise control reduces material waste (e.g., scrap material loss is reduced by 25%), and stable pressure curves and demolding strategies can reduce equipment energy consumption (energy consumption is reduced by 15-20% year-on-year), combining quality and cost advantages.

[0165] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A control system for hydraulic forming of magnetic materials, characterized in that, include: The first acquisition module (10) is used to acquire dynamic parameters from the hydraulic system and mold equipment through a sensor array, and to preprocess the dynamic parameters using a data fusion algorithm to obtain smooth pressure data and displacement time series data. The second acquisition module (20) is used to train a coordinated control model of pressure and displacement using a support vector machine algorithm based on smooth pressure data and displacement time series data, determine the dynamic mapping relationship between pressure and displacement, and obtain an optimized combination of control parameters. The third acquisition module (30) is used to adjust the opening of the servo valve of the hydraulic system in real time and, in combination with the optimized control parameter combination, correct the pressure output to obtain a stable pressure application curve if the control deviation of the dynamic mapping relationship between pressure and displacement exceeds the preset threshold. The fourth acquisition module (40) is used to predict the distribution state of material density using a neural network algorithm based on the stable pressure application curve and the displacement time series data, determine whether there is a density uneven region, and obtain the predicted result of density distribution. The fifth acquisition module (50) is used to adjust the demolding speed and mold temperature according to the predicted density distribution at the beginning of the demolding process, and to optimize the demolding parameters using a fuzzy control algorithm to obtain a demolding control strategy that reduces the risk of cracking. The first acquisition module (10) includes: The first acquisition unit (11) is used to acquire dynamic parameters from the hydraulic system and mold equipment in real time through the sensor array to obtain the initial parameter dataset; The generation unit (12) is used to perform noise reduction processing on the collected pressure information and displacement information based on the initial parameter dataset and a data fusion algorithm to generate a smooth parameter set after preliminary processing. The first determining unit (13) is used to perform outlier detection on the pressure sequence and displacement sequence using a preset threshold range for the smoothing parameter set. If outlier data exceeding the threshold is detected, it is interpolated and corrected to determine the corrected parameter sequence. The second acquisition unit (14) is used to acquire the corrected parameter sequence, and apply time series analysis methods to extract the trend of the pressure sequence and displacement sequence to obtain parameter change trend data. The first judgment unit (15) is used to monitor the operating status of the hydraulic system and mold equipment in real time through parameter change trend data. If the trend data deviates from the preset normal range, the status abnormality mark is triggered to judge potential operating risks. The third acquisition unit (16) is used to predict the pressure and displacement trends in the future period based on the state anomaly markers and historical time series data, using the Kalman filter algorithm to obtain the predicted trend dataset. The second determining unit (17) is used to generate a dynamic adjustment strategy for the predicted trend dataset. If the predicted trend shows an abnormal and continuous trend, the control parameters of the hydraulic system are fine-tuned to determine the optimized operating parameter configuration.

2. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, The dynamic parameters in the first acquisition unit (11) cover multi-dimensional information, including pressure information and displacement information.

3. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, In the first determining unit (13), the anomaly detection of pressure data points is calculated using the following formula: ； in, Indicates the first Anomaly detection flags for individual pressure data points. Indicates the first One pressure measurement value, This represents the mean of the pressure series. The standard deviation of the pressure series is represented. Indicates the threshold multiplier coefficient; The displacement value after interpolation correction is: ； in, Indicates the first Corrected displacement values ​​at each location This represents the previous normal displacement data point. This represents the next normal displacement data point. Indicates the interpolation correction factor; Smoothing of the parameter sequence by weighted averaging is achieved using the following formula: ； in, Indicates the first The smoothing parameter value at each position, Indicates the length of the smooth window. Indicates the first Each weighting coefficient This represents the data points in the original parameter sequence.

4. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, In the first judgment unit (15), the overall deviation of the hydraulic system and mold equipment parameters from the normal range is calculated using the following formula: ； in, Indicates the degree of deviation of the parameter trend. Indicates the number of monitoring parameters. Indicates the first The current values ​​of the parameters, This indicates the preset normal mean value of the parameter; The risk level of a single parameter deviating from the normal range is calculated using the following formula: ； in, Indicates the first Risk assessment index of each parameter, Indicates the first Real-time monitoring values ​​of each parameter This indicates the reference standard value for this parameter. This indicates the upper limit of the parameter's normal range. This indicates the lower limit of the parameter's normal range. The status exception flag is triggered using the following formula: ； in, Indicates an abnormal state marker signal. This indicates the number of sampling points for the trend data. Indicates the first Trend data values ​​at each moment. This represents the baseline value for the trend under normal conditions. This indicates the threshold for triggering an anomaly.

5. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, The second acquisition module (20) includes: The fourth acquisition unit (21) is used to match the changes of the pressure data and displacement time series data in the time series point by point using a data comparison tool based on the acquired pressure data and displacement time series data. For the matching results, a preset threshold range is used for comparison. If the deviation exceeds the threshold range, the corresponding time point is marked to obtain the marked deviation dataset. The third determining unit (22) is used to perform numerical correction on the marked time points using a data interpolation tool for the marked deviation dataset, obtain the corrected pressure data and displacement data set, and use a data integration tool to rearrange the corrected pressure data and displacement data set according to the time series to determine the integrated time series dataset. The fifth acquisition unit (23) is used to construct a dynamic correspondence between pressure data and displacement time series data using a data mapping tool based on the integrated time series dataset. For the nonlinear change part in the correspondence, it is processed by a feature extraction tool to obtain a feature-based mapping dataset. The fourth determining unit (24) is used to compare the control parameters item by item with the parameter adjustment tool for the characteristic mapping relationship dataset. If the parameter combination does not match the preset running state, the parameter value is iteratively updated to determine the adjusted control parameter combination.

6. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, The third acquisition module (30) includes: The sixth acquisition unit (31) is used to acquire real-time pressure data and displacement time series data from the hydraulic system, and to perform point-by-point matching of pressure data and displacement time series data through time series tools to determine the dynamic correspondence between pressure data and displacement time series data in the time series and obtain the initial mapping dataset. The seventh acquisition unit (32) is used to calculate the control deviation between the pressure data and the displacement time series data using a deviation calculation tool for the initial mapping dataset. If the deviation exceeds a preset threshold, the time points exceeding the threshold are marked by a marking tool to obtain the marked deviation dataset. The fifth determining unit (33) is used to adjust the opening degree of the hydraulic system servo valve through the servo valve control tool according to the marked deviation dataset, and to iteratively update the control parameters by combining the parameter optimization tool to determine the adjusted servo valve opening parameter combination. The eighth acquisition unit (34) is used to process the adjusted servo valve opening parameter combination through the pressure output tool, generate the corrected pressure output data, and use the curve fitting tool to smooth the corrected pressure output data to obtain a stable pressure application curve.

7. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, The fourth acquisition module (40) includes: The ninth acquisition unit (41) is used to acquire corresponding data points from the stable pressure application curve and the displacement sequence, use a data mapping tool to perform point-to-point correspondence processing on the time axis, generate an initial distribution association dataset, determine the dynamic relationship between pressure application and displacement sequence, and obtain preliminary distribution feature data. The tenth acquisition unit (42) is used to extract material properties in segments by using state analysis tools based on preliminary distribution characteristic data, and compare the characteristic values ​​of each region with the region judgment tool. If the characteristic value of a certain region deviates from the preset threshold range, the region is marked as a potential uneven region, and the marked region dataset is obtained. The eleventh acquisition unit (43) is used to perform detailed calculations on the density distribution using a depth prediction tool for the labeled regional dataset, generate a corresponding density distribution prediction map, determine whether there are areas with uneven density, and obtain the predicted distribution state data. The sixth determining unit (44) is used to compare the predicted distribution state data in multiple dimensions through the distribution assessment tool, obtain the assessment index of each region, and verify it one by one with the unevenness detection tool to determine the final density distribution prediction result.

8. The magnetic material hydraulic forming control system as described in claim 1, characterized in that, The fifth acquisition module (50) includes: The twelfth acquisition unit (51) is used to perform point-by-point analysis of the density value of the regional data obtained from the density distribution prediction results using a data comparison tool. If the density value exceeds the preset threshold range, it is marked as a high-risk area, and the marked distribution dataset is obtained. The seventh determining unit (52) is used to match the density data of the high-risk area with the initial values ​​of demolding speed and mold temperature through a parameter mapping tool for the labeled distribution dataset, generate a preliminary parameter adjustment scheme, and determine the adjustment range of each area; The thirteenth acquisition unit (53) is used to dynamically optimize the preliminary parameter adjustment scheme using fuzzy control tools, extract the optimal demolding speed and mold temperature combination from the adjustment range, and obtain the optimized parameter configuration data. The second judgment unit (54) is used to match the optimized parameter configuration data with the actual demolding process in real time through the parameter application tool, dynamically adjust the speed and temperature according to the density distribution characteristics, judge whether the preset conditions for reducing crack risk are met, and obtain the final control strategy data.

9. The magnetic material hydraulic forming control system as described in claim 8, characterized in that, In the thirteenth acquisition unit (53), the parameter mapping tool adopts the MATLAB / Simulink fuzzy logic toolbox.

Images