Prediction control system for wrinkling of automobile bearing stamping part

Abstract

The invention discloses a prediction control system for wrinkling of an automobile bearing stamping part, and relates to the technical field of model prediction, and the system comprises a data collection unit which is composed of a fluctuation collection matrix and a variable collection matrix, obtains the fluctuation change data of a plate in the stamping process through the fluctuation collection matrix, and transmits the fluctuation change data to a control unit; variable data of the plate in the stamping process is obtained through the variable collection matrix; and the stamping model is established based on historical stamping data, fluctuation change data are obtained for differential analysis, abnormal data segments of the historical stamping data are determined, abnormal sound wave and vibration wave characteristics of the plate are determined, and the corrugation of the stamping part is predicted through a corrugation prediction algorithm in combination with variable data, so that a prediction result is generated. In the plate stamping process, the state change of the plate in the mold is determined by monitoring the vibration wave and sound wave change of the plate, so that the purpose of predicting wrinkling is achieved, and the stamping quality of the plate is further improved.

Description

Technical Field

[0001] The invention relates to the technical field of model prediction, in particular to a prediction control system for wrinkling of automobile bearing stamping parts. Background Art

[0002] Stamped bearings are widely used in engines, transmissions, and steering systems to support and guide rotating components, ensuring smooth operation under high speeds and high loads. Their lightweight and high-precision design contributes to vehicle weight reduction and fuel economy, while also meeting high speed and wear resistance requirements.

[0003] The physical changes in sheet metal during the stamping process involve both elastic and plastic deformation. When external stamping pressure is applied to the sheet, the material first undergoes elastic deformation, manifesting as a linear relationship between stress and strain. After exceeding the yield point, the sheet metal enters the plastic deformation phase, undergoing permanent deformation. During this plastic deformation, the material's internal lattice structure adjusts, producing irreversible changes, often accompanied by localized work hardening.

[0004] Announcement No. CN113221402B discloses a method, system and storage medium for predicting and monitoring the springback of stamping parts. The method for predicting and monitoring the springback of stamping parts includes obtaining first data corresponding to a target mold profile and second data corresponding to a single stamping profile, determining a target compensation value based on the first data and the second data, constructing a springback prediction model and / or monitoring the sheet metal state based on the target compensation value, obtaining second data of a single stamping profile during the actual stamping process of the stamping part, determining the target compensation value by combining the first data corresponding to the target mold profile with the second data, guiding the stamping part processing according to the springback prediction model, so that the processing method of the stamping part is more in line with the actual processing requirements, and / or monitoring the sheet metal state is also conducive to ensuring the processing quality of the stamping part.

[0005] However, due to the complex shape and typical thin-wall structure of bearing stamping parts, the stamping process is prone to problems such as local overpressure and uneven force, which can cause wrinkling of the sheet metal. Therefore, it is difficult to solve the wrinkling problem by simply predicting springback.

[0006] When the sheet metal is subjected to locally excessively high blanking force or punching force during the stamping process, the sheet metal is subjected to a large local compressive stress that exceeds the yield limit of the material, which is prone to wrinkles. If the friction between the sheet metal and the mold is not effectively controlled, it will also lead to local deformation restriction, stress concentration, and increased wrinkling risk.

[0007] To this end, this application proposes a predictive control system that monitors the stress distribution and deformation of the sheet metal during the stamping process in real time. Summary of the Invention

[0008] One of the purposes of the present invention is to provide a predictive control system for wrinkling of automobile bearing stamping parts. During the process of sheet metal stamping, the changes in the vibration waves and sound waves of the sheet metal are monitored to determine the state changes of the sheet metal inside the mold, thereby achieving the purpose of wrinkling prediction and further improving the quality of sheet metal stamping processing.

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A predictive control system for wrinkling of automobile bearing stamping parts, comprising: The data acquisition unit is composed of a fluctuation acquisition matrix and a variable acquisition matrix. The data acquisition unit obtains the control parameters of the control system for the stamping equipment, obtains the fluctuation change data of the sheet during the stamping process through the fluctuation acquisition matrix, and obtains the variable data of the sheet during the stamping process through the variable acquisition matrix; The pre-processing unit suppresses the noise of the fluctuating data and the variable data through noise filtering and then splits them into data segments; The stamping model is established based on historical stamping data. It obtains fluctuation change data for differential analysis, identifies abnormal data segments compared with historical stamping data, and determines the abnormal acoustic and vibration wave characteristics of the sheet metal. Combined with variable data, the wrinkling prediction algorithm is used to predict the wrinkling of stamping parts and generate prediction results. The feedback control unit receives the prediction results, provides feedback on the abnormal state of the sheet, and adjusts the stamping parameters according to the feedback results.

[0010] In one or more embodiments of the present invention, the wave acquisition matrix is ​​composed of multiple acoustic emission sensors and vibration sensors, which are distributed at different positions of the mold with the middle of the plate as the reference point to capture the acoustic wave and vibration wave signals of the plate during the stamping process; A three-dimensional coordinate space is established and the coordinate point of each sensor in the coordinate space is determined. The data collected by the sensor is mapped to the coordinate point, and a three-dimensional distribution map of the sound waves and vibration waves is constructed. The wrinkling position of the sheet is determined by comparing and analyzing the signal strength, frequency change and timing change at different positions.

[0011] In one or more embodiments of the present invention, the variable acquisition matrix includes a plurality of pressure sensors and displacement sensors, wherein the pressure sensors are located at high stress or weak areas within the mold, and the displacement sensors are located outside the mold to monitor the displacement distance of the mold during the stamping process; Both the pressure sensor and the displacement sensor have a unified clock synchronization function. Based on the pressure data, a pressure distribution surface diagram with the synchronization time as the benchmark is constructed, and it is integrated with the displacement data to generate a dynamic stress-displacement model to map the compressive state of the sheet metal.

[0012] In one or more embodiments of the present invention, the variable acquisition matrix includes multiple temperature sensors and strain sensors. The temperature sensors are arranged in the key hot areas of the mold, and the strain sensors are attached to the surface of the plate to monitor temperature changes and strain distribution in real time. The coordinate points of the temperature sensors and strain sensors are mapped in a three-dimensional coordinate space to construct a temperature-strain distribution diagram.

[0013] In one or more embodiments of the present invention, a vibration sensor acquires vibration change data, and an acoustic emission sensor acquires sound wave change data. Noise suppression is performed on the vibration change data and the sound wave change data through a preprocessing unit. The vibration change data and the sound wave change data are split into data segments based on the complete vibration cycle in the vibration change data, and the periodic characteristics of the signal are retained.

[0014] In one or more embodiments of the present invention, the steps for determining the abnormal acoustic and vibration wave characteristics of the plate are as follows: Extract features from each vibration change data segment and sound wave change data segment through time series and frequency domain analysis; Based on historical stamping data, a feature distribution model in the normal stamping process is established to form a feature baseline, describing the feature range, distribution and evolution trend under normal stamping conditions; Comparing the characteristics of the vibration and sound wave change data segment with the characteristic distribution model to determine whether the characteristics of the vibration and sound wave change data segment are within a normal range; By setting a threshold or anomaly detection algorithm, it is determined whether the data segment is abnormal and the abnormal data segment is marked.

[0015] In one or more embodiments of the present invention, the abnormal data segment is combined with the variable data to establish a wrinkle prediction algorithm: Determine the timestamp of the abnormal data segment, determine the variable data range based on the timestamp, form a multidimensional feature vector for each group of corresponding abnormal data segment and variable data, and generate a state label; Extracting signal features from abnormal data segments and variable features from variable data; Construct a complete feature vector X by combining signal features and variable features; Using historical stamping data, a learning model is established based on the feature vector X and labels. Cross-validation and feature importance analysis are used to further optimize the learning model and screen sensitive features.

[0016] In one or more embodiments of the present invention, during the stamping process, the vibration and sound wave change data and the corresponding process parameter data collected are preprocessed and feature extracted through the fluctuation acquisition matrix and the variable acquisition matrix to generate a new feature vector, which is input into the prediction model to determine the prediction result and risk probability.

[0017] In one or more embodiments of the present invention, the steps of determining the wrinkle position of the sheet material based on the three-dimensional coordinate space are as follows: The sound wave and vibration signals collected by each acoustic emission sensor and vibration sensor during the stamping process will be associated with their corresponding coordinate points; Using the spatial interpolation method, the data of each discrete coordinate point is expanded to the entire surface area of ​​the plate to form a three-dimensional distribution map of the sound and vibration signals; For the abnormal features of each point in the entire area, cluster analysis is used to group coordinate points with similar abnormal characteristics to determine whether there are local abnormal areas; Combining the spatial distribution map with the clustering results, the location of the abnormal area is identified in the three-dimensional coordinate system.

[0018] In one or more embodiments of the present invention, for each sensor coordinate point, the pre-processed fluctuation intensity, timing information and frequency characteristics are integrated into a multi-dimensional feature vector F i , where the multidimensional feature vector F i represents the feature vector of sensor i.

[0019] Through the above technical solution, the present invention has the following beneficial effects: 1. During the sheet metal stamping process, the present application determines the state changes of the sheet metal inside the mold by monitoring the changes in vibration waves and sound waves of the sheet metal, thereby achieving the purpose of wrinkling prediction and further improving the quality of sheet metal stamping processing.

[0020] 2. The sound waves and vibration waves generated by the normal state of sheet metal stamping and the wrinkled state after stamping are inconsistent. The acquired fluctuation change data is segmented to form data segments. Each data segment is analyzed, and the corresponding timestamp of the data segment can be easily combined with variable data to verify the wrinkling state.

[0021] 3. During the stamping process of the plate, the positions of deformation and stress distribution are different, resulting in different vibration locations and sound wave points, thus forming a unique acoustic vibration characteristic map. By analyzing the position of the abnormal point through the coordinate space, the location of the wrinkle can be determined.

[0022] 4. By comprehensively analyzing the fluctuation intensity, timing changes, and frequency characteristics collected by each sensor in the three-dimensional coordinate space, we can accurately construct the sound wave and vibration distribution map of the sheet during the stamping process. By comparing the data at different locations, when abnormal aggregation occurs in a local area, we can preliminarily determine that there is a risk of wrinkling in that area.

[0023] Other features and advantages of the present invention will be described in the following description, and part of them will become obvious from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of the predictive control system of the present invention. DETAILED DESCRIPTION

[0025] The following drawings illustrate various embodiments of the present invention. For clarity, many practical details are included in the following description. However, it should be understood that these practical details are not intended to limit the present invention. In other words, in some embodiments of the present invention, these practical details are optional. Furthermore, features from different embodiments may be interchangeably applicable, where practically possible.

[0026] Unless otherwise defined, all terms used herein (including technical and scientific terms) have their ordinary meanings as understood by those skilled in the art. Furthermore, the definitions of the aforementioned terms in commonly used dictionaries should be interpreted in the context of this specification as consistent with the meanings in the art relevant to the present invention. Unless otherwise explicitly defined, these terms should not be interpreted as having idealized or overly formal meanings.

[0027] When the sheet metal is subjected to locally excessively high blanking force or punching force during the stamping process, the sheet metal is subjected to locally large compressive stress, which exceeds the yield limit of the material and is prone to wrinkles. If the friction between the sheet metal and the mold is not effectively controlled, it will also lead to local deformation restriction, stress concentration, and increased wrinkling risk.

[0028] See Figure 1 As shown, the present invention provides a predictive control system for wrinkling of automobile bearing stamping parts, which predicts the wrinkling of stamping parts based on the changes of the sheet material during the stamping process, determines the changes of the sheet material after stamping, and adjusts the stamping based on the changing state of the stamping parts.

[0029] The predictive control system includes: The data acquisition unit is composed of a fluctuation acquisition matrix and a variable acquisition matrix. The data acquisition unit obtains the control parameters of the control system for the stamping equipment, obtains the fluctuation change data of the sheet during the stamping process through the fluctuation acquisition matrix, and obtains the variable data of the sheet during the stamping process through the variable acquisition matrix; The pre-processing unit suppresses the noise of the fluctuating data and the variable data through noise filtering and then splits them into data segments; The stamping model is established based on historical stamping data. It obtains fluctuation change data for differential analysis, identifies abnormal data segments compared with historical stamping data, and determines the abnormal acoustic and vibration wave characteristics of the sheet metal. Combined with variable data, the wrinkling prediction algorithm is used to predict the wrinkling of stamping parts and generate prediction results. The feedback control unit receives the prediction results, provides feedback on the abnormal state of the sheet, and adjusts the stamping parameters according to the feedback results.

[0030] In one feasible method, during the stamping process of the sheet metal, when there is insufficient lubrication or uneven contact between the sheet metal and the mold, abnormal high-frequency friction sound or sudden impact sound will be generated, and local stress concentration and local sudden deformation will cause irregular vibration of the equipment or the sheet metal itself. These abnormal sound wave and vibration wave characteristics are captured by the data acquisition unit and analyzed to obtain the deformation state of the sheet metal inside the mold.

[0031] Among them, the characteristics of sound waves and vibration waves are used, combined with variable data, to further verify the sound waves and vibration waves to determine the changing state of the plate. Through multi-dimensional data cross-validation, the wrinkling risk is accurately identified, and the stamping process is adjusted in real time to ensure the quality of plate forming and improve production efficiency.

[0032] In one embodiment, the wave acquisition matrix is ​​composed of multiple acoustic emission sensors and vibration sensors, which are distributed at different positions of the mold with the middle of the plate as the reference point to capture the acoustic wave and vibration wave signals of the plate during the stamping process; A three-dimensional coordinate space is established and the coordinate point of each sensor in the coordinate space is determined. The data collected by the sensor is mapped to the coordinate point, and a three-dimensional distribution map of the sound waves and vibration waves is constructed. The wrinkling position of the sheet is determined by comparing and analyzing the signal strength, frequency change and timing change at different positions.

[0033] In one feasible method, during the stamping process of the plate, the different locations of deformation and stress distribution lead to different vibration generation locations and sound wave points, thus forming a unique acoustic vibration characteristic map. By monitoring these characteristic changes in real time, the wrinkling trend can be accurately judged.

[0034] Among them, parameters such as the wave transmission intensity, frequency, and timing changes are recorded in real time. When these data are mapped in spatial coordinates, the wave source point of the plate in the mold can be located and the wrinkling area of ​​the plate can be determined.

[0035] In one embodiment, the variable acquisition matrix includes a plurality of pressure sensors and displacement sensors, wherein the pressure sensors are located at high stress or weak areas within the mold, and the displacement sensors are located outside the mold to monitor the displacement distance of the mold during the stamping process; Both the pressure sensor and the displacement sensor have a unified clock synchronization function. Based on the pressure data, a pressure distribution surface diagram with the synchronization time as the benchmark is constructed, and it is integrated with the displacement data to generate a dynamic stress-displacement model to map the compressive state of the sheet metal.

[0036] In one feasible method, the wrinkling risk points can be accurately located by real-time comparison of acoustic vibration characteristics and stress-displacement models. During the stamping process of the plate, the pressure will vary with the degree of deformation. When the plate is normally pressurized, the pressure change of the plate is normal pressure change. When the plate shows signs of wrinkling, the pressure change will significantly deviate from the normal range.

[0037] By real-time monitoring of the deviation between pressure changes and acoustic vibration characteristics, the system can quickly identify and warn of wrinkling risks, ensuring stable sheet forming quality and effectively improving production efficiency.

[0038] In one embodiment, the variable acquisition matrix includes multiple temperature sensors and strain sensors. The temperature sensors are arranged in the key hot areas of the mold, and the strain sensors are attached to the surface of the plate to monitor temperature changes and strain distribution in real time. The coordinate points of the temperature sensors and strain sensors are mapped in a three-dimensional coordinate space to construct a temperature-strain distribution diagram.

[0039] In one feasible method, the temperature change of the plate during the stamping process is obtained through a temperature sensor and a strain sensor, and the temperature-strain distribution diagram is combined with a three-dimensional distribution diagram. The three-dimensional coordinate space can more intuitively determine the various changing states of the plate from stamping to deformation, including temperature, sound waves, and vibration waves, so that the analysis of the plate can be more comprehensive.

[0040] In another embodiment, the strain sensor is arranged inside the mold. Since the strain sensor is designed to be thin and soft, but it is necessary to avoid the influence of the strain sensor on the plate, the thickness and flexibility of the strain sensor are required to avoid affecting the plate. Therefore, the strain sensor is installed inside the mold to monitor the temperature changes at the contact position between the mold and the plate, reducing the difficulty of selecting the strain sensor.

[0041] In one embodiment, a vibration sensor acquires vibration change data, and an acoustic emission sensor acquires sound wave change data. Noise suppression is performed on the vibration change data and the sound wave change data through a preprocessing unit. The vibration change data and the sound wave change data are split into data segments based on the complete vibration cycle in the vibration change data, and the periodic characteristics of the signal are retained.

[0042] In one feasible method, the vibration change data and the sound wave change data are split by retaining the periodic characteristics in a complete manner, thereby retaining the periodic characteristics. This helps to extract key information related to the vibration mode, peak and valley distribution, and more accurately capture the vibration characteristics within each cycle, so as to facilitate the comparison and prediction of abnormal vibration characteristics.

[0043] In another embodiment, the vibration change data and the sound wave change data are split based on a fixed time window, without the need for additional detection of the period, so the segmentation is simpler and easier to implement. A stable segmentation strategy can be provided based on the time window to keep the length of each data segment consistent. The relatively robust and unified data division can be applicable to variable stamping conditions and can be used when there is less historical stamping data and the data fluctuations are complex.

[0044] In one embodiment, the steps for determining the abnormal acoustic and vibration wave characteristics of the plate are as follows: Extract features from each vibration change data segment and sound wave change data segment through time series and frequency domain analysis; Based on historical stamping data, a feature distribution model in the normal stamping process is established to form a feature baseline, describing the feature range, distribution and evolution trend under normal stamping conditions; Comparing the characteristics of the vibration and sound wave change data segment with the characteristic distribution model to determine whether the characteristics of the vibration and sound wave change data segment are within a normal range; By setting a threshold or anomaly detection algorithm, it is determined whether the data segment is abnormal and the abnormal data segment is marked.

[0045] In this embodiment, vibration data features include but are not limited to: maximum amplitude, mean, variance, spectrum peak, spectrum distribution, and energy distribution; Acoustic wave data features include but are not limited to: signal strength, frequency content, frequency band energy, and acoustic emission.

[0046] In one feasible method, historical stamping data records the vibration signal and acoustic wave signal of the entire stamping process. In each stamping cycle, the vibration signal and acoustic wave signal are preprocessed, and then the characteristic baseline of the normal stamping process is constructed through the historical stamping data.

[0047] Distance measurement is used to determine whether the real-time data is within the normal range. If the amplitude, spectrum distribution or acoustic wave energy in a certain data segment in the real-time data exceeds the normal range defined by the baseline, then the data segment is significantly different from the historical stamping data.

[0048] Among them, data segment characteristics that deviate from the normal range are marked as abnormal data segments.

[0049] In one embodiment, the abnormal data segments are combined with the variable data to establish a wrinkle prediction algorithm: Determine the timestamp of the abnormal data segment, determine the variable data range based on the timestamp, form a multidimensional feature vector for each group of corresponding abnormal data segment and variable data, and generate a state label; Extracting signal features from abnormal data segments and variable features from variable data; Construct a complete feature vector X by combining signal features and variable features; Using historical stamping data, a learning model is established based on the feature vector X and labels. Cross-validation and feature importance analysis are used to further optimize the learning model and screen sensitive features.

[0050] In one practicable method, the multidimensional feature vector label is 0 or 1, where label 0 represents that the plate is normal and has no wrinkles; label 1 represents that the plate is abnormal and has a risk of wrinkling.

[0051] For example, the characteristic vector X = [RMS vibration, peak vibration, acoustic wave energy, pressure change rate, punching speed]; The random forest algorithm process is: import numpy as np from sklearn.ensemble import RandomForestClassifier from sklearn.model_selection import train_test_split from sklearn.metrics import classification_report # Assume we have constructed dataset X and label y # X.shape = (n_samples, n_features) # y.shape = (n_samples) where 0 means no wrinkling and 1 means there is a risk of wrinkling # Split training set and test set X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) #Build a random forest classifier model = RandomForestClassifier(n_estimators=100, random_state=42) model.fit(X_train, y_train) # Make predictions on the test set y_pred = model.predict(X_test) print(classification_report(y_test, y_pred)) In the above example, the model is further optimized through cross-validation and feature importance analysis methods to screen out the features that are most sensitive to wrinkling risk.

[0052] In one embodiment, during the stamping process, the vibration and sound wave change data and the corresponding process parameter data collected are preprocessed and feature extracted through the fluctuation acquisition matrix and the variable acquisition matrix to generate a new feature vector, which is input into the prediction model to determine the prediction result and risk probability.

[0053] For example, # new_feature_vector is a one-dimensional feature vector generated in real time wrinkle_prob = model.predict_proba(new_feature_vector.reshape(1, -1))[0][1] # Get the probability of wrinkling threshold = 0.6 # preset threshold (can be tuned based on historical data) if wrinkle_prob>threshold: # Trigger feedback control, such as reducing punch speed, adjusting die pressure or increasing lubrication trigger_feedback_control() Real-time data is fed into the model. If the predicted risk exceeds a preset threshold, process parameters are adjusted automatically or manually, achieving closed-loop control to prevent sheet wrinkling. The stamping process can be dynamically monitored and predicted using both historical and real-time data, significantly reducing the defective product rate caused by wrinkling while enhancing the intelligence of the production process.

[0054] In one embodiment, the steps of determining the wrinkle position of the sheet material based on the three-dimensional coordinate space are as follows: The sound wave and vibration signals collected by each acoustic emission sensor and vibration sensor during the stamping process will be associated with their corresponding coordinate points; Using the spatial interpolation method, the data of each discrete coordinate point is expanded to the entire surface area of ​​the plate to form a three-dimensional distribution map of the sound and vibration signals; For the abnormal features of each point in the entire area, cluster analysis is used to group coordinate points with similar abnormal characteristics to determine whether there are local abnormal areas; Combining the spatial distribution map with the clustering results, the location of the abnormal area is identified in the three-dimensional coordinate system.

[0055] One feasible approach involves comprehensively analyzing the wave intensity, temporal variations, and frequency characteristics of the wave data collected by each sensor in a three-dimensional coordinate space. This allows for a precise reconstruction of the acoustic and vibration distribution of the sheet during the stamping process. By comparing data from different locations, if abnormal clustering occurs in a localized area, a preliminary assessment of wrinkling risk can be made. Using spatial clustering methods and multidimensional feature fusion technology, not only can wrinkling risk locations be predicted, but they can also provide a basis for real-time feedback control, enabling closed-loop regulation and ensuring stamping quality.

[0056] In one embodiment, for each sensor coordinate point, the pre-processed fluctuation intensity, timing information and frequency characteristics are integrated into a multi-dimensional feature vector F i , where the multidimensional feature vector F i represents the feature vector of sensor i.

[0057] For example, during one stamping cycle, each sensor collects the following pre-processed data features (the example data units are hypothetical values): Signal strength (amplitude / RMS): A(0,0): 1.0, B(10,0): 1.1, C(-10,0): 2.5, D(0,10): 1.0, E(0,-10): 1.2; Main frequencies (Hz): A(0,0): 2000, B(10,0): 2050, C(-10,0): 2500, D(0,10): 1980, E(0,-10): 2010; Timing characteristics (such as signal start time or phase change rate): The phase changes at most points are relatively smooth, but point C shows a significant advance and large phase fluctuation.

[0058] After digital signal processing, a multi-dimensional feature vector is formed. The feature vector at sensor C is: F c =[Intensity=2.5, Main Frequency=2500 Hz, Phase Fluctuation=Large]; The data of the remaining sensors are relatively similar, reflecting normal conditions.

[0059] Compared to the historical baseline data under normal conditions, the signal strength collected at various locations during the normal stamping process ranged from 0.9 to 1.2, the frequency remained stable between 1900 and 2100 Hz, and the timing signal was stable. However, the data collected by sensor C was clearly abnormal—with an amplitude of 2.5, a main frequency of 2500 Hz, and large phase fluctuations.

[0060] Combined with the abnormal signal at the location of sensor C, it is determined that there is excessive vibration energy and frequency offset in the local area, indicating that the plate in this area may be at risk of wrinkling due to local abnormal force. An early warning will then be issued, and the information will be fed back to the stamping control unit through the feedback control unit to automatically adjust the local mold distribution pressure, reduce the stamping speed or improve the lubrication scheme to alleviate the stress concentration in this area.

[0061] It has the following technical effects: During the sheet metal stamping process, the present application determines the state changes of the sheet metal inside the mold by monitoring the changes in vibration waves and sound waves of the sheet metal, thereby achieving the purpose of wrinkling prediction and further improving the quality of the sheet metal stamping process.

[0062] The sound waves and vibration waves generated by the normal state of sheet metal stamping and the wrinkled state after stamping are inconsistent. The acquired fluctuation change data is segmented to form data segments, and each data segment is analyzed. The corresponding timestamp of the data segment can be easily combined with variable data to verify the wrinkling state.

[0063] During the stamping process of the plate, the different locations of deformation and stress distribution lead to different vibration locations and sound wave points, thus forming a unique acoustic vibration characteristic map. By analyzing the position of the abnormal points in the coordinate space, the location of the wrinkle can be determined.

[0064] Although the present invention is disclosed in conjunction with the above embodiments, it is not intended to limit the present invention. Anyone familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the definition of the attached claims.

Claims

1. A predictive control system for wrinkling of automobile bearing stamping parts, characterized in that: include: The data acquisition unit is composed of a fluctuation acquisition matrix and a variable acquisition matrix. The data acquisition unit obtains the control parameters of the control system for the stamping equipment, obtains the fluctuation change data of the sheet during the stamping process through the fluctuation acquisition matrix, and obtains the variable data of the sheet during the stamping process through the variable acquisition matrix; The pre-processing unit suppresses the noise of the fluctuating data and the variable data through noise filtering and then splits them into data segments; The stamping model is established based on historical stamping data. It obtains fluctuation change data for differential analysis, identifies abnormal data segments compared with historical stamping data, and determines the abnormal acoustic and vibration wave characteristics of the sheet metal. Combined with variable data, the wrinkling prediction algorithm is used to predict the wrinkling of stamping parts and generate prediction results. The feedback control unit receives the prediction results, provides feedback on the abnormal state of the sheet, and adjusts the stamping parameters according to the feedback results.

2. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 1, characterized in that: The wave acquisition matrix consists of multiple acoustic emission sensors and vibration sensors. With the middle of the sheet as the reference point, they are distributed at different positions of the die to capture the acoustic and vibration wave signals of the sheet during the stamping process. A three-dimensional coordinate space is established and the coordinate point of each sensor in the coordinate space is determined. The data collected by the sensor is mapped to the coordinate point, and a three-dimensional distribution map of the sound waves and vibration waves is constructed. The wrinkling position of the sheet is determined by comparing and analyzing the signal strength, frequency change and timing change at different positions.

3. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 2, characterized in that: The variable acquisition matrix includes multiple pressure sensors and displacement sensors. The pressure sensors are located at high stress or weak areas inside the mold, and the displacement sensors are set outside the mold to monitor the mold displacement distance during the stamping process. Both the pressure sensor and the displacement sensor have a unified clock synchronization function. Based on the pressure data, a pressure distribution surface diagram with the synchronization time as the benchmark is constructed, and it is integrated with the displacement data to generate a dynamic stress-displacement model to map the compressive state of the sheet metal.

4. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 3, characterized in that: The variable acquisition matrix includes multiple temperature sensors and strain sensors. The temperature sensors are arranged in the key hot areas of the mold, and the strain sensors are attached to the surface of the plate to monitor temperature changes and strain distribution in real time. The coordinate points of the temperature sensors and strain sensors are mapped in the three-dimensional coordinate space to construct a temperature-strain distribution diagram.

5. The predictive control system for wrinkling of automobile bearing stamping parts according to claim 3, characterized in that: The vibration sensor obtains vibration change data, and the acoustic emission sensor obtains sound wave change data. The vibration change data and the sound wave change data are subjected to noise suppression by the preprocessing unit. The vibration change data and the sound wave change data are split into data segments based on the complete vibration cycle in the vibration change data to retain the periodic characteristics of the signal.

6. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 4 or 5, characterized in that: The steps to determine the abnormal sound and vibration wave characteristics of the plate are as follows: Extract features from each vibration change data segment and sound wave change data segment through time series and frequency domain analysis; Based on historical stamping data, a feature distribution model in the normal stamping process is established to form a feature baseline, describing the feature range, distribution and evolution trend under normal stamping conditions; Comparing the characteristics of the vibration and sound wave change data segment with the characteristic distribution model to determine whether the characteristics of the vibration and sound wave change data segment are within a normal range; By setting a threshold or anomaly detection algorithm, it is determined whether the data segment is abnormal and the abnormal data segment is marked.

7. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 6, characterized in that: Abnormal data segments are combined with variable data to establish a wrinkle prediction algorithm: Determine the timestamp of the abnormal data segment, determine the variable data range based on the timestamp, form a multidimensional feature vector for each group of corresponding abnormal data segment and variable data, and generate a state label; Extracting signal features from abnormal data segments and variable features from variable data; Construct a complete feature vector X by combining signal features and variable features; Using historical stamping data, a learning model is established based on the feature vector X and labels. Cross-validation and feature importance analysis are used to further optimize the learning model and screen sensitive features.

8. The predictive control system for wrinkling of automobile bearing stamping parts according to claim 7, characterized in that: During the stamping process, the vibration and sound wave change data and the corresponding process parameter data collected are preprocessed and feature extracted through the fluctuation acquisition matrix and the variable acquisition matrix to generate a new feature vector. The new feature vector is input into the prediction model to determine the prediction result and risk probability.

9. The predictive control system for wrinkling of automobile bearing stamping parts according to claim 8, characterized in that: The steps to determine the wrinkle position of the sheet based on the three-dimensional coordinate space are as follows: The sound wave and vibration signals collected by each acoustic emission sensor and vibration sensor during the stamping process will be associated with their corresponding coordinate points; Using the spatial interpolation method, the data of each discrete coordinate point is expanded to the entire surface area of ​​the plate to form a three-dimensional distribution map of the sound and vibration signals; For the abnormal characteristics of each point in the entire area, cluster analysis is used to group coordinate points with similar abnormal characteristics to determine whether there are local abnormal areas; Combining the spatial distribution map with the clustering results, the location of the abnormal area is identified in the three-dimensional coordinate system.

10. A predictive control system for wrinkling of automobile bearing stamping parts according to claim 9, characterized in that: For each sensor coordinate point, the preprocessed fluctuation intensity, timing information and frequency characteristics are integrated into a multidimensional feature vector F i , where the multidimensional feature vector F i represents the feature vector of sensor i.

Images