Data analysis-based digital optimization method for forming parameters of flat wire motor coil

Abstract

The present application relates to the technical field of data analysis, and more particularly to a flat wire motor hairpin forming parameter digital optimization method based on data analysis. The method comprises the following steps: performing digital unit conversion processing of process parameters of each stage of hairpin deformation based on the process parameters of the flat wire motor hairpin process, to generate hairpin deformation digital unit data; performing production behavior characteristic analysis processing of the hairpin deformation digital unit data based on the control equipment operation data and the corresponding hairpin change data, to generate production behavior characteristic data of the deformation unit; establishing a forming deviation parameter-control equipment adjustment optimization tree model; and performing hairpin forming control equipment optimization adjustment parameter analysis on the production behavior characteristic data of the deformation unit by using the forming deviation parameter-control equipment adjustment optimization tree model, to generate control equipment optimization adjustment parameters. The present application realizes intelligent optimization of accurate and stable hairpin forming parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, and in particular to a flat wire motor hairpin forming parameter digital optimization method based on data analysis. BACKGROUND

[0002] With the rapid development of new energy vehicles and other high-end equipment fields, flat wire motors gradually become the mainstream development direction of the motor industry due to their high efficiency, high power density, miniaturization and other significant advantages. As the core component of the stator winding of the flat wire motor, the forming precision of the hairpin directly determines the assembly quality of the stator winding, and further affects the electromagnetic performance, heat dissipation efficiency and long-term operation reliability of the motor. However, the existing flat wire motor hairpin forming parameter optimization technology has many defects that are difficult to overcome for the complex characteristics of multi-factor coupling and dynamic change of the flat wire motor hairpin forming process, and the systematic analysis of the hairpin deformation mechanism cannot realize the precise matching of process parameters and deformation process. The parameters set by experience often have subjectivity and limitations, and it is difficult to adapt to individualized needs under different material specifications and different forming targets, which easily leads to size deviation, irregular shape and other quality problems of hairpin forming, increases the waste rate and production cost. On the other hand, the monitoring and adjustment of the existing forming process are mostly passive responses, which are difficult to collect multi-source dynamic data (such as equipment operation parameters and hairpin deformation data) in the production process in real time, and also cannot establish the correlation between process parameter changes and forming deviations. When forming quality problems occur, it is difficult to quickly locate the deviation source and make targeted parameter adjustment, resulting in low production efficiency and being unable to meet the needs of large-scale high-precision production. SUMMARY

[0003] Therefore, the present application provides a flat wire motor hairpin forming parameter digital optimization method based on data analysis to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a flat wire motor hairpin forming parameter digital optimization method based on data analysis includes the following steps:

[0005] Step S1: Obtain the flat wire motor hairpin forming process parameters to be formed; perform digital unit conversion processing on the process parameters of each stage of hairpin deformation based on the flat wire motor hairpin forming process parameters, and generate hairpin deformation digital unit data;

[0006] Step S2: Use the multi-source sensors built-in in the flat wire motor hairpin forming control equipment to monitor the control equipment operation data and the corresponding hairpin change data of the forming production, and obtain the control equipment operation data and the corresponding hairpin change data, respectively;

[0007] Step S3: Based on the control device operation data and the corresponding card changing data, the production behavior characteristic analysis processing of the card changing deformation unit data is performed to generate the deformation unit production behavior characteristic data;

[0008] Step S4: The process parameter change and deviation response correlation analysis processing of the changing unit is performed through the deformation unit production behavior characteristic data to generate the changing unit parameter change-deviation response correlation data;

[0009] Step S5: Based on the preset decision tree algorithm and the changing unit parameter change-deviation response correlation data, a forming deviation parameter-control device adjustment optimization tree model is established; the control device optimization adjustment parameter is generated by using the forming deviation parameter-control device adjustment optimization tree model to perform the card forming control device optimization adjustment parameter analysis of the deformation unit production behavior characteristic data.

[0010] Further, the flat wire motor card forming process parameters in step S1 include card forming material physical property parameters, control device setting parameters, and card forming target data.

[0011] Further, step S1 includes the following steps:

[0012] Step S11: Obtain the flat wire motor card forming process parameters to be formed;

[0013] Step S12: Perform card deformation mechanism simulation processing according to the flat wire motor card forming process parameters to generate card deformation mechanism simulation data;

[0014] Step S13: Divide the card deformation mechanism simulation data into stages to generate card deformation mechanism stage simulation data;

[0015] Step S14: Perform unit architecture conversion processing of each card deformation mechanism stage according to the card deformation mechanism stage simulation data to generate card deformation mechanism unit architecture data;

[0016] Step S15: Perform card forming process parameter attribute type analysis processing according to the flat wire motor card forming process parameters to generate card forming process attribute type analysis data;

[0017] Step S16: Based on the card forming process attribute type analysis data, design the dependent mapping relationship of each deformation unit and process factor characteristic to generate the deformation unit-process factor characteristic dependent relationship model;

[0018] Step S17: Transmit the card deformation mechanism unit architecture data to the deformation unit-process factor characteristic dependent relationship model to perform digital process characteristic mapping processing of each deformation unit to generate card deformation digital unit data.

[0019] Further, the card sending process attribute type analysis data in step S15 includes material attribute analysis data, control equipment analysis data, and card sending mold restriction analysis data.

[0020] Further, step S16 includes the following steps:

[0021] Step S161: analyzing material deformation response factor characteristics of card sending deformation according to the material attribute analysis data, to generate material deformation response factor characteristic data;

[0022] Step S162: analyzing control equipment driving factor characteristics of card sending deformation according to the control equipment analysis data, to generate control equipment driving factor characteristic data;

[0023] Step S163: analyzing geometric restriction factor characteristics of each local area of card sending according to the card sending mold restriction analysis data, to generate card sending geometric restriction factor characteristic data;

[0024] Step S164: designing a dependent mapping relationship of each deformation unit and process factor characteristics by using the material deformation response factor characteristic data, the control equipment driving factor characteristic data, and the card sending geometric restriction factor characteristic data, to generate a deformation unit-process factor characteristic dependent relationship model.

[0025] Further, step S3 includes the following steps:

[0026] Step S31: analyzing control equipment running state characteristics by using the control equipment running data, to generate control equipment running state characteristic data;

[0027] Step S32: analyzing a control target local area according to the control equipment running state characteristic data, to generate control target local area data;

[0028] Step S33: analyzing card sending local geometric change characteristics by using the card sending change data through the control target local area data, to generate card sending local geometric change characteristic data;

[0029] Step S34: designing a card sending deformation digital unit data timing window according to the card sending deformation digital unit data, and performing window division processing on the control equipment running state characteristic data and the corresponding card sending local geometric change characteristic data through the card sending deformation digital unit timing window, to respectively generate window control equipment running state characteristic data and corresponding window card sending local geometric change characteristic data;

[0030] Step S35: mapping the window control equipment running state characteristic data and the corresponding window card sending local geometric change characteristic data to the card sending deformation digital unit data to perform deformation unit production behavior characteristic analysis processing, to generate deformation unit production behavior characteristic data.

[0031] Further, step S4 comprises the following steps:

[0032] Step S41: Perform production behavior trend state analysis on the deformation unit production behavior characteristic data to generate deformation unit production behavior trend state data;

[0033] Step S42: Perform production behavior quality deviation analysis according to the deformation unit production behavior trend state data to generate production behavior quality deviation data;

[0034] Step S43: Perform quality deviation attribution relationship analysis of the deformation unit according to the card deformation digital unit data and the production behavior quality deviation data to generate deformation unit quality deviation attribution relationship data;

[0035] Step S44: Perform process parameter change and deviation response correlation analysis processing of the deformation unit through the deformation unit quality deviation attribution relationship data to generate deformation unit parameter change-deviation response correlation data.

[0036] Further, step S41 comprises the following steps:

[0037] Step S411: Perform card local deformation autocorrelation trend feature analysis according to the window card local geometric change characteristic data to generate card local deformation autocorrelation trend feature data;

[0038] Step S412: Perform card local deformation trend influence feature analysis according to the window control equipment running state characteristic data to generate card local deformation influence feature data;

[0039] Step S413: Perform deformation unit production behavior trend state analysis on the deformation unit production behavior characteristic data through the card local deformation autocorrelation trend feature data and the card local deformation influence feature data to generate deformation unit production behavior trend state data.

[0040] Further, step S44 comprises the following steps:

[0041] Step S441: Perform grouping processing of process parameter value change conditions on the card deformation digital unit data to generate grouped card deformation digital unit data;

[0042] Step S442: Perform process parameter value change orientation identification on the grouped card deformation digital unit data to generate identified grouped card deformation digital unit data;

[0043] Step S443: According to the deformation unit quality deviation attribution relationship data and the identification grouping card deformation digital unit data, the process parameter change and the deviation response of the change unit are associated and analyzed, and change unit parameter change-deviation response correlation data is generated.

[0044] Further, step S5 includes the following steps:

[0045] Step S51: According to the change unit parameter change-deviation response correlation data, the parameter influence characteristic analysis of the forming deviation is performed, and forming deviation parameter influence characteristic data is generated;

[0046] Step S52: The forming deviation parameter influence characteristic data is designed by a preset decision tree algorithm to establish a forming deviation parameter-control equipment adjustment relationship tree model;

[0047] Step S53: The forming deviation parameter constraint characteristic data is generated by the forming deviation parameter influence characteristic data, and the forming deviation parameter-control equipment adjustment constraint tree model is obtained by the forming deviation parameter constraint characteristic data for the tree node control equipment adjustment constraint processing of the forming deviation parameter-control equipment adjustment relationship tree model;

[0048] Step S54: The parameter sensitivity weight of the forming deviation is analyzed by the forming deviation parameter influence characteristic data, and the forming deviation parameter-control equipment adjustment constraint tree model is obtained by the parameter sensitivity weight for the tree node weight optimization processing of the forming deviation parameter-control equipment adjustment constraint tree model;

[0049] Step S55: The forming deviation parameter-control equipment adjustment optimization tree model is used to analyze the control equipment optimization adjustment parameter of the card forming of the deformation unit production behavior characteristic data, and the control equipment optimization adjustment parameter is generated.

[0050] The application has the advantages that the application solves the core defects of the traditional method, such as lack of hairpin deformation mechanism system analysis and low matching degree of process parameters and deformation process, through systematic and refined process design. The core process parameters including hairpin material physical properties, control equipment settings and forming targets are comprehensively obtained to ensure the completeness of parameter coverage. The whole process stage of hairpin deformation is clearly disassembled through deformation mechanism simulation and stage division to lay a foundation for subsequent accurate analysis. Subsequently, key analysis data such as material properties, control equipment and mold restrictions are accurately refined through process parameter attribute type analysis, and a correlation mapping model of deformation unit and material deformation response, equipment driving and geometric restriction is further constructed to realize digital feature mapping of deformation mechanism unit architecture data and generate standardized and structured hairpin deformation digital unit data. Not only the depth analysis of the deformation mechanism is realized, but also the accurate correlation between the deformation unit and the multi-dimensional process factors is established to flexibly adapt to the personalized needs of different material specifications and different forming targets. With the help of the multi-source sensors built-in the flat wire motor hairpin forming control equipment, the synchronous real-time monitoring of the control equipment operation data and the hairpin change data is realized. This monitoring mode can dynamically capture the subtle fluctuations of the equipment running state (such as pressure, speed and temperature) and the geometric shape change of the hairpin in the corresponding stage in a real and dynamic manner. The two types of data obtained have strong correlation and timeliness. At the same time, the application of multi-source sensors ensures the comprehensiveness of data acquisition and avoids the one-sidedness of single data dimension, providing detailed and reliable raw data support for subsequent steps such as production behavior feature analysis and parameter change and deviation response correlation, ensuring the objectivity and effectiveness of the subsequent analysis results and providing key data support for accurately grasping the dynamic law of the forming process. The state feature analysis of the control equipment operation data is combined with the local area data analysis of the control target to realize the hierarchical extraction of the two types of data, i.e. the equipment operation and the hairpin deformation. Then, the two types of feature data are accurately windowed and matched through the design of the deformation unit time window to ensure the correspondence of the data and the deformation stage and the deformation unit. The matched data is mapped to the hairpin deformation digital unit data to generate production behavior feature data that can accurately reflect the dynamic behavior law of each deformation unit. Not only can the subtle changes in the forming process be quickly captured, but also the deep correlation between the equipment running state, the hairpin deformation state and the deformation unit is realized to provide core data support for the subsequent establishment of the correlation between parameter change and deviation response and the accurate positioning of the quality deviation source, further ensuring the pertinence and effectiveness of the subsequent parameter optimization.Through the analysis of the self-correlation trend of the local deformation of the hairpin and the characteristics of the equipment influence, the production behavior trend state of the deformation unit is accurately judged, which provides accurate basis for deviation identification; based on the trend state data, quality deviation analysis is carried out, and the attribution relationship between the deviation and the deformation unit is clarified combined with the digital unit data of the hairpin deformation, so that the problem of fuzzy deviation positioning is solved; through the grouping and guiding identification of process parameter change conditions, the accurate correlation between the process parameter change of the change unit and the deviation response is established, and the generated change unit parameter change-deviation response correlation data clearly reveals the quantitative corresponding relationship between different process parameter fluctuations and forming quality deviation, which improves the pertinence and scientificity of deviation optimization. Based on the decision tree algorithm combined with the change unit parameter change-deviation response correlation data, the core influencing factors are first determined through parameter influence characteristic analysis, and then the basic correlation model is constructed through tree relationship node design, and subsequently the deviation parameter constraint processing and sensitivity weight optimization are carried out, forming a forming deviation parameter-control equipment adjustment optimization tree model with constraint and precision. The model can accurately analyze and output the optimal adjustment parameters of the control equipment based on the production behavior characteristic data of the deformation unit, which on the one hand realizes the transformation from qualitative experience judgment to quantitative and accurate calculation, greatly improves the scientificity and accuracy of the adjustment parameters; on the other hand, the model has clear logical link and optimization mechanism, and can quickly respond to the deviation problem in different forming scenarios, output the optimal adjustment scheme with the best adaptability, effectively improve the precision and stability of the hairpin forming, shorten the parameter adjustment cycle, improve the production efficiency, fully meet the large-scale high-precision production demand, and provide core technical support for the high-precision manufacturing of flat wire motors.

[0051] Therefore, the data analysis-based digital optimization method for hairpin forming parameters of the flat wire motor of the present invention achieves precise matching between process parameters and deformation process by digitally converting the process parameters of each stage of hairpin deformation and conducting system analysis in conjunction with hairpin deformation mechanism simulation. This effectively overcomes the subjectivity and limitations of traditional empirical parameter setting, and can flexibly adapt to personalized needs under different material specifications and different forming objectives. It reduces quality problems such as dimensional deviation and irregular shape from the root, and lowers scrap rate and production cost. Relying on the multi-source sensors built into the flat wire motor hairpin forming control equipment, it realizes real-time and comprehensive control equipment operation data and hairpin change data during the forming process. By monitoring and establishing a mechanism for analyzing the production behavior characteristics of deformation units and the correlation between process parameter changes and deviation responses, the root cause of molding quality deviations can be quickly located. This solves the technical problems of difficulty in accurately correlating parameter changes and deviations and the inability to adjust parameters in a targeted manner. Based on the decision tree algorithm, the molding deviation parameter-control equipment adjustment optimization tree model can realize the accurate analysis and output of control equipment optimization adjustment parameters, promote the transformation of quantitative and precise control of hairpin molding parameter optimization, significantly improve the accuracy and stability of hairpin molding of flat wire motors, and improve production efficiency. This meets the needs of large-scale high-precision production and provides key process guarantees for the development of flat wire motors towards high precision and high reliability. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the steps of a digital optimization method for hairpin forming parameters of a flat wire motor based on data analysis according to the present invention.

[0053] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S5.

[0054] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0055] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0056] In addition, the accompanying drawings are included to provide a further understanding of the application, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the application and, together with the description, serve to explain the principles of the application. In the drawings:

[0057] To achieve the above object, the present application provides a data analysis-based digital optimization method for forming parameters of a flat wire motor coil, and in an embodiment of the present application, please refer to Figures 1 to 2 The present application provides a data analysis-based digital optimization method for forming parameters of a flat wire motor coil, and in an embodiment of the present application, please refer to Figure 1 The present application provides a data analysis-based digital optimization method for forming parameters of a flat wire motor coil, and in an embodiment of the present application, please refer to

[0058] Step S1: Obtain the forming process parameters of the flat wire motor coil; based on the forming process parameters of the flat wire motor coil, perform digital unit conversion processing on the process parameters of each stage of the coil forming process, and generate digital unit data of the coil forming process;

[0059] In the embodiment of the present application, comprehensive collection of forming process parameters of the to-be-formed flat wire motor hairpin is carried out, and the collected process parameters explicitly include hairpin material physical property parameters, control equipment setting parameters and hairpin forming target data, wherein the hairpin material physical property parameters cover core physical property indexes such as material elastic modulus, yield strength and thermal conductivity coefficient, the control equipment setting parameters cover equipment operation reference parameters such as forming pressure reference value, driving speed reference value and heating temperature reference value, and the hairpin forming target data cover forming quality standards such as hairpin final geometric size, shape accuracy tolerance and surface roughness. Based on the above-mentioned collected process parameters, hairpin deformation mechanism simulation is carried out, the mechanical properties and thermodynamic properties of the hairpin forming process are simulated and analyzed, and complete hairpin deformation mechanism simulation data are generated; then, according to the physical process of hairpin forming, the hairpin deformation mechanism simulation data are divided into three continuous deformation stages of initial positioning stage, bending deformation stage and shaping stage, and hairpin deformation mechanism stage simulation data are obtained; the unit architecture conversion is carried out for each deformation stage simulation data, each deformation stage is divided into a plurality of independent deformation units according to the hairpin structure part, the unit architecture corresponding to each stage is constructed, and hairpin deformation mechanism unit architecture data are generated; at the same time, attribute type analysis is carried out according to the collected process parameters, material class, equipment class and mold class parameter attributes are explicitly distinguished, and hairpin process attribute type analysis data containing material attribute analysis data, control equipment analysis data and hairpin mold restriction analysis data are generated, wherein the material attribute analysis data correspond to the attribute classification and quantitative range of the material physical property parameters, the control equipment analysis data correspond to the attribute classification and control dimension of the equipment setting parameters, and the hairpin mold restriction analysis data correspond to the attribute classification of the mold geometric parameters, positioning accuracy and other restriction conditions; feature analysis is respectively carried out based on the above-mentioned three types of analysis data, material deformation response factors affecting hairpin deformation are extracted through the material attribute analysis data, material deformation response factor feature data are generated, equipment driving factors driving hairpin deformation are extracted through the control equipment analysis data, control equipment driving factor feature data are generated, and local geometric restriction factors restricting hairpin deformation are extracted through the hairpin mold restriction analysis data, hairpin geometric restriction factor feature data are generated; the correlation mapping relationship between the deformation unit and the process factor features is constructed based on the three types of feature data, the corresponding correlation rules of each deformation unit and the material deformation response factor, the equipment driving factor and the geometric restriction factor are explicitly defined, and the deformation unit-process factor feature subordination relationship model is generated; finally, the hairpin deformation mechanism unit architecture data are input into the correlation model, the digital mapping of each deformation unit and the corresponding process factor features is completed through the correlation rules built in the model, and the hairpin deformation digital unit data containing the deformation unit number, the deformation stage to which the deformation unit belongs, the associated process factor features and the quantitative indexes are generated.

[0060] Step S2: The control equipment operation data of the forming production and the corresponding change data of the hairpin are monitored by using the multi-source sensor built in the flat wire motor hairpin forming control equipment, and the control equipment operation data and the corresponding change data of the hairpin are obtained respectively;

[0061] In the embodiment of the application, the multi-source sensor built in the flat wire motor hairpin forming control equipment is used to carry out monitoring processing of the whole process of forming production. The sensor combination of the monitoring system includes a torque sensor, a pressure sensor, a laser displacement sensor, an infrared temperature sensor, a laser profile sensor and an industrial vision sensor. The technical parameters of each sensor are uniformly calibrated as follows: the torque sensor has a range of 0-50 N·m, a measurement accuracy of ±0.01 N·m and a sampling frequency of 100 Hz; the pressure sensor has a range of 0-20 MPa, a measurement accuracy of ±0.01 MPa and a sampling frequency of 100 Hz; the laser displacement sensor has a measurement range of 0-50 mm, an accuracy of ±0.001 mm and a sampling frequency of 100 Hz; the infrared temperature sensor has a measurement range of 0-300℃, an accuracy of ±0.5℃ and a sampling frequency of 50 Hz; the laser profile sensor has a scanning frequency of 50 frames / s, a measurement accuracy of ±2 μm and a sampling point spacing of 0.01 mm; and the industrial vision sensor has a resolution of 1920×1080, a frame rate of 30 frames / s and a measurement accuracy of ±0.003 mm. During the monitoring operation, the torque sensor is arranged at the end of the driving shaft of the equipment to collect the dynamic change data of the driving torque in the forming process in real time; the pressure sensor is integrated in the forming pressure head to synchronously collect the time sequence change data of the forming pressure; the laser displacement sensor is installed beside the transmission mechanism of the equipment to accurately collect the displacement change data corresponding to the servo driving speed; the infrared temperature sensor array is arranged around the forming area to comprehensively collect the temperature distribution data of each point in the forming area, thereby realizing multi-dimensional synchronous collection of the control equipment operation data; the laser profile sensor and the industrial vision sensor are cooperatively arranged above the forming station, the laser profile sensor scans the profile shape change in the hairpin forming process in real time, and the industrial vision sensor synchronously captures the size change of the key parts of the hairpin. After the data fusion processing of the two types of sensors, the change data of the hairpin are obtained. During the monitoring process, the accurate alignment of the two types of data is realized through the time sequence synchronization module built in the equipment, the time synchronization accuracy is ±0.001 s, and it is ensured that the control equipment operation data of each time node corresponds to unique change data of the hairpin. At the same time, the collected original data are subjected to noise elimination processing, and the abnormal data points generated due to sensor jitter are eliminated, and finally the structured control equipment operation data and the corresponding change data of the hairpin are generated, wherein the control equipment operation data includes time sequence quantitative indexes such as driving torque, forming pressure, servo driving displacement and speed, and forming area temperature, and the change data of the hairpin includes geometric change quantitative indexes such as profile shape coordinates, key size values and shape deviation of each part of the hairpin.

[0062] Step S3: Based on the control equipment operation data and the corresponding card change data, the card deformation digital unit data is processed to analyze the production behavior characteristics of the deformation unit, and the deformation unit production behavior characteristic data is generated;

[0063] In the embodiment of the application, based on the control equipment operation data and the corresponding card change data, first, the control equipment operation data is processed to extract the core operation state indicators such as the driving torque fluctuation value, the power stability coefficient, the displacement deviation, the speed fluctuation rate, and the temperature uniformity. The control equipment operation state characteristic data is generated through quantitative analysis of the indicators. Based on the operation state characteristic data, in combination with the key control nodes of the card forming, the local area of the control target that needs to be monitored is located, which covers the key deformation areas such as the card bending part and the end forming part. The control target local area data containing the area coordinates and the monitoring priority is generated. According to the control target local area data, the image data and the size data in the card change data are extracted, the deformation amount, the deformation rate, and the shape deviation of each key area are analyzed, and the card local geometric change characteristic data is generated. Based on the card deformation digital unit data, the card deformation unit time sequence window with equal time interval is designed according to the forming time length of each deformation unit. The time length of each window is accurately matched with the forming period of the corresponding deformation unit. The control equipment operation state characteristic data and the card local geometric change characteristic data are divided into synchronous windows through the time sequence window, so that the divided window control equipment operation state characteristic data corresponds to the equipment operation state of a specific deformation unit at a specific time period, and the window card local geometric change characteristic data corresponds to the card deformation state of the same deformation unit at the same time period. The two types of window data are accurately mapped to the card deformation digital unit data according to the deformation unit number, the equipment operation state characteristics and the card geometric change characteristics of the same deformation unit at the same time period are matched, the dynamic correlation between the two is analyzed, and the deformation unit production behavior characteristic data containing the time sequence operation characteristics, the time sequence deformation characteristics, and the correlation of each deformation unit is generated.

[0064] Step S4: The correlation analysis processing of the process parameter change and the deviation response of the deformation unit is performed through the deformation unit production behavior characteristic data, and the deformation unit parameter change-deviation response correlation data is generated.

[0065] In the embodiment of the present application, the production behavior characteristic data of the deformation unit is taken as the core analysis object, the time sequence analysis method is used to carry out the self-correlation trend characteristic analysis of the local deformation of the window card according to the local geometric change characteristic data of the window card, the correlation coefficient of the deformation amount of the same deformation unit in adjacent time periods is calculated, the change trend and continuity of the deformation amount are mined, and the self-correlation trend characteristic data of the local deformation of the window card is generated; meanwhile, the correlation between the equipment operation parameter change and the local deformation trend of the window card is analyzed according to the equipment operation state characteristic data of the window control device, the influence degree of the driving torque change, the speed change and the temperature change on the deformation trend is determined, and the local deformation influence characteristic data of the window card is generated; the trend of the deformation unit production behavior characteristic data is analyzed by combining the above two types of characteristic data, the deformation development direction, the stable state and the change rate of each deformation unit are determined, and the trend state data of the deformation unit production behavior is generated; the production behavior data exceeding the shape accuracy tolerance and the size deviation threshold is screened out by comparing and analyzing the trend state data and the card forming target data, the type, degree and occurrence period of the quality deviation are determined, and the production behavior quality deviation data is generated; according to the card deformation digital unit data, the production behavior quality deviation data is matched according to the deformation unit number, the specific deformation unit corresponding to each quality deviation is determined, and the quality deviation attribution relationship data of the deformation unit is generated; the card deformation digital unit data is grouped according to the process parameter value change condition, different value ranges of core parameters such as forming pressure, driving speed and heating temperature are taken as the grouping basis, and the grouped card deformation digital unit data is generated; the process parameter value change direction and the change amplitude interval corresponding to each group of data are determined by identifying the process parameter value change direction of each group of data, and the identified grouped card deformation digital unit data is generated; the deviation occurrence of different deformation units under the same parameter change condition is analyzed by combining the quality deviation attribution relationship data of the deformation unit and the identified grouped card deformation digital unit data, the quantitative corresponding relationship between the process parameter change and the deviation degree and the deviation type is established, and the change unit parameter change-deviation response correlation data is generated.

[0066] Step S5: a forming deviation parameter-control equipment adjustment optimization tree model is established based on the preset decision tree algorithm and the change unit parameter change-deviation response correlation data; the control equipment optimization adjustment parameter analysis of the deformation unit production behavior characteristic data is carried out by using the forming deviation parameter-control equipment adjustment optimization tree model, and the control equipment optimization adjustment parameter is generated.

[0067] In the embodiment of the present application, the model construction and parameter optimization are carried out based on the preset decision tree algorithm and the variation-unit parameter variation-deviation response correlation data. According to the variation-unit parameter variation-deviation response correlation data, the influence degree and influence range of each process parameter variation on the forming deviation are analyzed, the core parameters that play a leading role in the forming deviation are screened out, the influence weight and influence law of each core parameter are determined, and the forming deviation parameter influence characteristic data are generated. The parameter influence characteristic data are subjected to tree relationship node design through the preset decision tree algorithm, the variation-unit number is taken as the root node, the core forming deviation parameter is taken as the first branch node, the parameter variation range is taken as the second branch node, and the control equipment adjustment direction is taken as the leaf node, so as to construct the forming deviation parameter-control equipment adjustment relationship tree model. Based on the forming deviation parameter influence characteristic data, the reasonable value range of each core parameter, the limit threshold of the control equipment adjustment and the safe operation boundary are analyzed, and the forming deviation parameter constraint characteristic data are generated. The constraint characteristic data are used for node constraint processing of the forming deviation parameter-control equipment adjustment relationship tree model, the leaf nodes that exceed the reasonable range of the parameters or break through the limit of the equipment adjustment are eliminated, and the forming deviation parameter-control equipment adjustment constraint tree model is obtained. The sensitivity weight of each core parameter is calculated through the forming deviation parameter influence characteristic data, each branch node in the constraint tree model is given a corresponding weight value, the node weight corresponding to the high-sensitivity parameter is strengthened, and the forming deviation parameter-control equipment adjustment optimization tree model is obtained. The variation-unit production behavior characteristic data are input into the optimization tree model, the model matches the corresponding root node according to the variation-unit number, the first and second branch nodes are matched through the forming deviation parameter characteristic, and the optimal leaf node is screened out by combining the node weight calculation, so as to determine the adjustment parameter type and specific adjustment amount of the control equipment, and generate the control equipment optimization adjustment parameter including the adjustment parameter, adjustment time sequence and adjustment precision requirement of the control equipment corresponding to each variation unit.

[0068] Further, the flat wire motor coil forming process parameters in step S1 include coil forming material physical property parameters, control equipment setting parameters and coil forming target data.

[0069] Further, step S1 includes the following steps:

[0070] Step S11: obtaining the flat wire motor coil forming process parameters to be formed;

[0071] In the embodiment of the present application, full-dimensional accurate collection operation of the flat wire motor hairpin forming process parameters to be formed is carried out. In the hairpin material physical property parameter collection link, the relevant data of the elastic modulus, yield strength, tensile strength and Poisson's ratio of the flat wire material are obtained. For example, the elastic modulus value interval of the flat wire material iron-based alloy is 200GPa-210GPa, the yield strength value interval is 300MPa-350MPa, the tensile strength value interval is 350MPa-400MPa, and the Poisson's ratio value is fixed at 0.27. In the control equipment setting parameter collection link, the data collection is completed by using an industrial grade industrial control parameter collection technology. The technology is configured with a sampling frequency of 100Hz, a data resolution of 16 bits, and a signal collection error of ±0.1%. Through the technology, the forming pressure reference value, the servo drive speed reference value, the forming area heating temperature reference value, and the mold clamping gap reference value of the hairpin forming control equipment are collected. The forming pressure reference value interval is 2MPa-8MPa, the servo drive speed reference value interval is 0.5mm / s-4mm / s, the forming area heating temperature reference value interval is 80℃-150℃, and the mold clamping gap reference value interval is 0.02mm-0.08mm. In the hairpin forming target data collection link, the data collection is completed by using a laser profile measurement technology. The technology is configured with a scanning frequency of 50 frames / s, a measurement accuracy of ±2μm, and a sampling point spacing of 0.01mm. Through the technology, the target geometric size, the profile tolerance, the surface roughness, and the bending angle tolerance of the hairpin after forming (such as U-shaped bending, Z-shaped forming, and 3D bending three-stage process forming) are collected. The length tolerance of the target geometric size is ±0.01mm, the width tolerance is ±0.008mm, the profile tolerance is ±0.015mm, the surface roughness is Ra0.8μm, and the bending angle tolerance is ±0.1°. All collection operations are completed based on standardized physical testing and industrial collection means. All parameters collected are quantitative numerical data without qualitative description data. The parameter values collected are verified by multiple retests, the data error is controlled within the specified threshold range, and the collected process parameters can fully reflect the basic process requirements and physical constraint conditions of the flat wire motor hairpin forming, thereby providing comprehensive and accurate basic data support for subsequent deformation mechanism simulation and digital unit conversion.

[0072] Step S12: hairpin deformation mechanism simulation processing is performed according to the flat wire motor hairpin process parameters to generate hairpin deformation mechanism simulation data.

[0073] In the embodiment of the application, full-process simulation of the hairpin deformation mechanism is carried out, and the process is realized by using a coupled simulation analysis technology of mechanics and thermodynamics. The time step of the simulation calculation is 0.01 s, the mesh partitioning accuracy of the hairpin three-dimensional model is 0.05 mm, the convergence accuracy of the mechanical simulation is 1e-6, the temperature calculation accuracy of the thermodynamic simulation is ±0.5℃, and the simulation iteration number is set to 2000. During the simulation, the material property parameters of the hairpin are taken as the boundary conditions of the simulation constitutive equation, the parameters such as elastic modulus, yield strength and Poisson's ratio are input into the material attribute library of the simulation, and the elastic-plastic deformation constitutive relationship of the flat wire material is constructed; the parameters of the control equipment are taken as the load boundary conditions of the simulation, the parameters such as forming pressure, driving speed and heating temperature are converted into dynamic load and temperature field loading conditions in the simulation process; the geometric parameters of the hairpin mold are taken as the constraint boundary conditions of the simulation, and the parameters such as mold cavity size, clamping gap and positioning reference are converted into geometric constraint conditions in the simulation process. Based on the above three types of boundary conditions, a three-dimensional solid model consistent with the actual hairpin forming structure is constructed, which completely restores the physical structure of the straight section, curved section and end forming area of the hairpin, and then the elastic-plastic mechanical deformation simulation and thermal conduction thermodynamic simulation of the full process of hairpin forming are carried out. The stress distribution state, strain change value, temperature field distribution range and geometric deformation displacement of the hairpin forming at each time period are calculated in real time during the simulation process, the simulation covers all physical change processes of the hairpin from the initial state to the final forming state, and there is no simulation missing in any forming link. The hairpin deformation mechanism simulation data generated after the simulation is a structured time series data set, which contains stress values, strain values, temperature values, geometric deformation coordinate values and plastic deformation cumulative amounts of each part of the hairpin at different time nodes. All simulation data are quantitative values and completely consistent with the actual forming physical law, and clearly reveal the dynamic change law of each physical quantity in the hairpin forming process.

[0074] Step S13: The hairpin deformation mechanism simulation data is divided into stages, and hairpin deformation mechanism stage simulation data is generated.

[0075] In the embodiment of the application, the standardization stage division processing of the hairpin deformation mechanism simulation data is carried out, and the division process is realized through time sequence feature clustering analysis technology. The distance threshold of the clustering analysis is 0.02 mm, the time threshold of the stage division is 0.5 s, the deformation rate determination threshold is 0.01 mm / s, the stress value mutation threshold is 50 MPa, the temperature value stability threshold is ±2℃, and the clustering feature dimension is three core indexes of deformation rate, stress value and temperature value. When carrying out stage division, first, the time sequence features of the hairpin deformation mechanism simulation data are extracted, the overall deformation rate of the hairpin at different time nodes, the stress value of the core part and the temperature value of the forming area are extracted, then the double-index mutation features of the deformation rate and the stress value are taken as the core division basis, and the stable state of the temperature value is taken as the auxiliary division basis, and the continuous simulation data is cut without overlapping. Based on the above division rule, the whole process of the hairpin deformation is accurately divided into three independent and continuous deformation stages. The first stage is the initial positioning stage of the hairpin. The deformation rate interval of this stage is 0-0.01 mm / s, the stress value interval is 0-50 MPa, and the temperature value is maintained at room temperature 25℃±2℃. The core feature of this stage is that the hairpin only has rigid displacement without plastic deformation. The second stage is the bending forming stage of the hairpin. The deformation rate interval of this stage is 0.01 mm / s-0.05 mm / s, the stress value interval is 50 MPa-300 MPa, and the temperature value interval is 80℃-150℃. The core feature of this stage is that the hairpin has large-scale elastic-plastic bending deformation, and the stress value gradually approaches the material yield strength. The third stage is the shaping and sizing stage of the hairpin. The deformation rate interval of this stage is 0.001 mm / s-0.01 mm / s, the stress value interval is 300 MPa-350 MPa, and the temperature value is maintained at 120℃±5℃. The core feature of this stage is that the plastic deformation of the hairpin tends to be stable, and the stress value reaches the material yield strength and completes the sizing. After the stage division is completed, the hairpin deformation mechanism stage simulation data is generated. The data includes the start and end time nodes of each deformation stage, the deformation range of each stage, the stress distribution interval, the temperature field feature and the plastic deformation cumulative amount. The simulation data of each stage is an independent time sequence data set, and the stage boundary is a quantitative feature mutation node, realizing the stage-by-stage accurate description of the whole process of the hairpin deformation.

[0076] Step S14: According to the hairpin deformation mechanism stage simulation data, the unit architecture conversion processing of each hairpin deformation mechanism stage is carried out, and the hairpin deformation mechanism unit architecture data is generated.

[0077] In the embodiment of the application, based on the geometric structure characteristics and deformation law of each deformation stage in the stage simulation data of the hairpin deformation mechanism, the space deformation unit is independently divided according to the actual physical structure of the hairpin. In the initial positioning stage of the hairpin, the hairpin is divided into two basic deformation units of straight line segment unit and end positioning unit, and the unit division is based on the geometric shape of the hairpin and the positioning constraint condition. In the bending forming stage of the hairpin, the hairpin is divided into four core deformation units of inside bending unit, outside stretching unit, transition section deformation unit and end forming unit, and the unit division is based on the stress concentration area and the main plastic deformation position of the hairpin. In the shaping stage of the hairpin, all the deformation units in the bending forming stage are retained, and two auxiliary deformation units of shaping and pressing unit and shaping and pressure maintaining unit are newly added, and the unit division is based on the shaping stress position and the shaping constraint area of the hairpin. Each divided deformation unit is assigned a unique unit number, and the spatial three-dimensional coordinates, geometric dimensions, deformation stage, deformation correlation range and stress and strain constraint conditions of each unit are recorded, and the deformation priority and physical constraint boundary of each deformation unit in the corresponding stage are labeled. After the unit architecture conversion is completed, the hairpin deformation mechanism unit architecture data is generated. The data is a structured unit feature set, which includes the number, deformation stage, spatial structure parameters, deformation constraint conditions and physical correlation characteristics of all deformation units. Each deformation unit is an independent analysis subject, and the unit boundary is completely coincided with the actual forming structure of the hairpin, so as to realize the accurate disassembly of the local deformation characteristics of the hairpin in each stage.

[0078] Step S15: Perform hairpin process parameter attribute type analysis processing according to the flat wire motor hairpin process parameters, and generate hairpin process attribute type analysis data;

[0079] In the embodiment of the present application, based on the flat wire motor carding process parameters, according to the physical properties of the parameters and the actual action mechanism in the carding forming, the precise classification analysis without intersection is carried out. The first type is material property analysis data, which is the analysis result of the material physical property parameters of the carding material, including the quantitative numerical interval, attribute label and deformation response characteristics of the material elastic modulus, yield strength, Poisson's ratio and thermal conductivity coefficient, and the influence law and critical threshold of each material physical property parameter on the plastic deformation of the carding are clear, such as the elastic modulus determines the anti-deformation ability of the material, the yield strength determines the starting point of the plastic deformation of the material, all analysis indexes are quantitative values and without qualitative description; The second type is control equipment analysis data, which is the analysis result of the control equipment setting parameters, including the quantitative control dimension, numerical range and equipment driving characteristics of the forming pressure, driving speed, heating temperature and clamping gap, and the driving mode and action strength of each device parameter on the carding forming are clear, such as the forming pressure determines the force of deformation, the driving speed determines the rate of deformation, all analysis indexes are corresponding to the actual running parameters of the equipment and accurate; The third type is carding die restriction analysis data, which is the analysis result of the die geometric parameters and constraint conditions, including the quantitative restriction threshold, geometric constraint characteristics and forming boundary conditions of the die cavity size, positioning accuracy, geometric constraint angle and clamping gap, and the constraint range and restriction degree of each parameter of the die on the local deformation of the carding are clear, such as the die cavity size determines the geometric contour of the carding, the positioning accuracy determines the forming position deviation of the carding, all analysis indexes are quantitative geometric parameters and consistent with the actual structure of the die. The carding process attribute type analysis data generated after the analysis is a structured classification feature set, the three types of analysis data have no overlap and no omission, and all analysis results are quantitative numerical characteristics and physical law characteristics, which provide accurate attribute feature support for subsequent construction of the association relationship between the deformation unit and the process factor.

[0080] Step S16: Based on the carding process attribute type analysis data, the dependent mapping relationship of each deformation unit and process factor feature is designed, and a deformation unit-process factor feature dependent relationship model is generated;

[0081] In the embodiment of the present application, based on the material attribute analysis data, the material deformation response factor characteristic analysis of the hairpin deformation is carried out, the influence coefficient of the material elastic modulus change on the deformation stiffness of the deformation unit, the defining index of the yield strength change on the plastic deformation critical value of the deformation unit, and the correlation characteristics of the Poisson's ratio on the transverse deformation of the deformation unit are extracted, and the generated material deformation response factor characteristic data contains the material physical quantity index corresponding to each deformation unit, the deformation response law, and the critical deformation threshold, which clearly defines the deformation characteristic difference of different deformation units under the influence of material physical parameters; secondly, based on the control equipment analysis data, the control equipment driving factor characteristic analysis of the hairpin deformation is carried out, the influence coefficient of the forming pressure change on the deformation rate of the deformation unit, the correlation degree of the driving speed change on the forming precision of the deformation unit, and the regulation and control characteristics of the heating temperature change on the material plasticity of the deformation unit are extracted, and the generated control equipment driving factor characteristic data contains the equipment operation quantitative index corresponding to each deformation unit, the driving action law, and the forming rate correlation characteristics, which clearly defines the forming characteristic difference of different deformation units under the driving of equipment parameters; again, based on the hairpin die restriction analysis data, the geometric restriction factor characteristic analysis of each local area of the hairpin is carried out, the constraint range of the die cavity size on the geometric profile of the deformation unit, the influence threshold of the positioning accuracy on the forming position of the deformation unit, and the correlation characteristics of the die clamping gap on the surface quality of the deformation unit are extracted, and the generated hairpin geometric restriction factor characteristic data contains the die quantitative index corresponding to each deformation unit, the geometric constraint law, and the forming boundary restriction characteristics, which clearly defines the deformation boundary difference of different deformation units under the constraint of the die. After completing the characteristic analysis of the three types of factors, based on the physical mechanism of the hairpin forming, the corresponding correlation mapping rules of each deformation unit and the three types of process factor characteristics are constructed, which clearly defines the subordinate corresponding relationship between the deformation unit and the process factor characteristics, and is verified through the forming mechanism of the deformation unit, to ensure that each process factor characteristic is subordinate to and only subordinate to the corresponding deformation unit, without cross-subordination or omission, such as the clamping gap characteristics only subordinate to the end forming unit and the transition section deformation unit, not to other deformation units. After the design is completed, the deformation unit-process factor characteristic subordinate relationship model is generated, which is a structured subordinate relationship set containing the number of all deformation units, the corresponding material / equipment / die process factor characteristics, the subordinate relationship determination basis, the non-cross-subordination verification result and other core contents, which clearly defines the process factor characteristic range to which each deformation unit is exclusively subordinate, and realizes the accurate subordinate mapping of the deformation unit and the process factor characteristics.

[0082] Step S17: transmit the hairpin deformation mechanism unit architecture data to the deformation unit-process factor characteristic subordinate relationship model for digital process characteristic mapping processing of each deformation unit, and generate hairpin deformation digital unit data.

[0083] In the embodiment of the application, based on the data of the card forming mechanism unit architecture, the number, the deformation stage to which it belongs, the spatial structure parameter, and the deformation constraint condition of all deformation units in the data are matched with the corresponding material deformation response factor characteristic data, control equipment driving factor characteristic data, and card forming geometric restriction factor characteristic data in full-dimension accurate matching according to the preset association rule of the deformation unit-process factor characteristic affiliation model. In the matching process, each feature of each deformation unit is quantitatively assigned. The quantitative indexes of material properties, the quantitative parameters of equipment operation, and the quantitative thresholds of mold constraints are deeply fused with the spatial structure, deformation law, and the deformation stage to which the deformation unit belongs. For each deformation unit that completes the feature matching, standardized digital process feature construction is carried out. Each deformation unit is given a complete digital feature set containing the unit number, the deformation stage to which it belongs, the material deformation response quantitative feature, the control equipment driving quantitative feature, the mold geometric restriction quantitative feature, the deformation critical threshold, and the forming precision requirement. All indexes of the feature set are quantitative numerical data without qualitative description characteristics. The digital features can completely reflect the physical structure, deformation law, process constraints, and the corresponding digital unit affiliation of the deformation unit. After the mapping process is completed, card forming digital unit data is generated. The data is a structured digital feature set containing the complete digital process features of all deformation units. The digital features of each deformation unit completely correspond to the actual forming process factors and deformation laws. The coding rules of the digital features are uniform, the feature quantification is accurate, and the standardized, structured, and accurate core data support is provided for subsequent production behavior feature analysis, which directly serves the data analysis and digital optimization of the flat wire motor card forming parameters.

[0084] Further, the card forming process attribute type analysis data in step S15 includes material attribute analysis data, control equipment analysis data, and card forming mold restriction analysis data.

[0085] Further, step S16 includes the following steps:

[0086] Step S161: Perform material deformation response factor characteristic analysis of the card forming deformation according to the material attribute analysis data to generate material deformation response factor characteristic data;

[0087] In the embodiment of the application, the key quantitative indicators such as elastic modulus, yield strength, Poisson's ratio and thermal conductivity coefficient in the material properties are extracted, and the corresponding association links of each index and the hairpin deformation unit are established. According to the structural characteristics and deformation requirements of different deformation units, the influence coefficient of elastic modulus change on the deformation stiffness of each deformation unit is calculated, wherein the elastic modulus influence coefficient calculation benchmark of the inner bending unit is 205GPa, and each fluctuation of 1GPa corresponds to a deformation stiffness change coefficient of 0.005. The elastic modulus influence coefficient calculation benchmark of the outer stretching unit is the same, but the deformation stiffness change coefficient is 0.006. Based on the yield strength numerical interval 300MPa-350MPa, the plastic deformation critical threshold of each deformation unit is defined, the plastic deformation starting yield strength threshold of the inner bending unit is 310MPa, that of the outer stretching unit is 305MPa, and that of the transition section deformation unit is 315MPa. The transverse deformation association characteristics of each deformation unit are quantitatively analyzed by Poisson's ratio 0.27, and the correlation ratio of the transverse deformation and longitudinal deformation of the straight section unit is determined as 0.27:1. The bending section unit adjusts the correlation ratio to 0.27:1.2 due to the difference in stress distribution. After the analysis, the material deformation response factor characteristic data is generated, which is classified and integrated according to the deformation unit number and contains the core contents such as the material physical property quantitative indicators corresponding to each deformation unit, the elastic modulus-deformation stiffness influence coefficient, the yield strength-plastic deformation critical threshold, the Poisson's ratio-transverse deformation association ratio and the thermal conductivity coefficient-temperature deformation response coefficient. The influence law of material properties on the deformation process of each deformation unit is clearly defined.

[0088] Step S162: Control device driving factor characteristic analysis according to the control device analysis data, generating control device driving factor characteristic data;

[0089] In the embodiment of the application, the quantitative control dimension and numerical range of the equipment parameters such as forming pressure, servo driving speed, heating temperature, clamping gap and the like are extracted, the driving correlation mapping of the equipment parameters and the deformation units is established, the influence coefficient of the forming pressure change on the deformation rate of each unit is calculated for the deformation units in different deformation stages, the inner bending unit in the bending forming stage is increased by 0.008 mm / s in deformation rate for each 1 MPa increase in forming pressure in the interval of 2 MPa-8 MPa, and the outer stretching unit is increased by 0.009 mm / s in deformation rate for each 1 MPa increase, the correlation degree of the servo driving speed change and the forming precision of each deformation unit is analyzed, the driving speed correlation degree weight of the linear segment unit is 0.3, the driving speed correlation degree weight of the bending segment unit is 0.5, and the driving speed correlation degree weight of the end forming unit is 0.4, the regulation and control characteristics of the temperature change on the material plasticity of each deformation unit are analyzed based on the interval of 80℃-150℃ of the heating temperature, the material plasticity of the bending segment unit is increased by 0.02 for each 10℃ increase in temperature, and the plasticity of the shaping stage unit is increased by 0.015, the control equipment driving factor characteristic data is generated after the analysis, the data is classified and correlated according to the deformation units and the equipment parameters, contains the key information such as the quantitative range of the equipment parameters corresponding to each deformation unit, the forming pressure-deformation rate influence coefficient, the driving speed-forming precision correlation degree, the temperature-plasticity regulation and control coefficient, and the clamping gap-surface quality constraint coefficient, and accurately reveals the driving action law of the control equipment parameters on the forming process of each deformation unit.

[0090] Step S163: performing geometric restriction factor characteristic analysis of each local area of the card according to the card mold restriction analysis data, and generating card geometric restriction factor characteristic data;

[0091] In the embodiment of the application, the quantitative restriction indexes of the mold cavity size, positioning accuracy, geometric constraint angle, and mold gap are extracted, the constraint association link of the mold geometric parameters and the local deformation units of the hairpin is established; according to the geometric shape requirements of each deformation unit, the constraint range of the mold cavity size on the geometric profile of the deformation unit is analyzed, the cavity size constraint range of the end forming unit is length ±0.01 mm and width ±0.008 mm, and the cavity size constraint range of the transition segment deformation unit is profile degree ±0.015 mm; based on the mold positioning accuracy parameters, the forming position influence threshold of each deformation unit is defined, the positioning accuracy influence threshold of the straight line segment unit is ±0.005 mm, and the positioning accuracy influence threshold of the curved segment unit is ±0.003 mm, and the position deviation of the deformation unit will be accumulated if the threshold is exceeded; the association features of the mold gap on the surface quality of the deformation unit are analyzed, the mold gap is in the interval of 0.02 mm-0.08 mm, the surface roughness Ra value of the deformation unit is increased by 0.05 μm for every 0.01 mm increase, and the sensitivity of the outer stretching unit to the mold gap is higher than that of the inner curved unit. After the analysis, the geometric restriction factor feature data of the hairpin is generated, the data is integrated according to the deformation unit number, contains the core contents such as the mold quantitative indexes corresponding to each deformation unit, the cavity size constraint range, the positioning accuracy influence threshold, the mold gap-surface quality association, and the geometric constraint angle restriction condition, and the restriction law of the mold geometric parameters on the deformation boundary of each deformation unit is clear.

[0092] Step S164: The dependent mapping relationship design of each deformation unit and the process factor feature is performed through the material deformation response factor feature data, the control equipment driving factor feature data, and the hairpin geometric restriction factor feature data, and a deformation unit-process factor feature dependent relationship model is generated.

[0093] In the embodiment of the present application, the dependent mapping relationship of each deformation unit and process factor characteristic is designed by the material deformation response factor characteristic data, the control equipment driving factor characteristic data and the card sending geometric restriction factor characteristic data, the dependent corresponding relationship of the deformation unit and the process factor characteristic is clear, a unified deformation unit number correlation reference is established, three types of process factor characteristic data are classified and regularized according to the deformation unit number, it is ensured that all process factor characteristic data can be accurately anchored through the unit number and the corresponding deformation unit, and cross-unit feature confusion is avoided. Based on the physical mechanism of card sending forming, the types of process factors actually involved in each deformation unit in the forming process are sorted out, and the dependent process factor characteristic details corresponding to each deformation unit are clear: for the material deformation response factor characteristic, the three core characteristics of the inner bending unit are dependent on the elastic modulus, yield strength and Poisson's ratio, the three core characteristics of the outer stretching unit are dependent on the elastic modulus, yield strength and thermal conductivity, and the two core characteristics of the straight line segment unit are dependent on the elastic modulus and Poisson's ratio, and the exclusive dependent relationship of the material characteristic to each deformation unit is clear; for the control equipment driving factor characteristic, the three characteristics of the inner bending unit are dependent on the forming pressure, driving speed and heating temperature, the two characteristics of the end forming unit are dependent on the forming pressure and the die gap, and the two characteristics of the shaping and pressing unit are dependent on the heating temperature and the driving speed, and the exclusive driving dependent relationship of the equipment parameters to each deformation unit is defined; for the mold geometric restriction factor characteristic, the three characteristics of the end forming unit are dependent on the cavity size, positioning accuracy and geometric constraint angle, the two characteristics of the transition segment deformation unit are dependent on the cavity size and the die gap, and the positioning accuracy of the straight line segment unit is dependent on the positioning accuracy, and the exclusive constraint dependent relationship of the mold parameters to each deformation unit is analyzed. According to the sorted dependent relationship, the forming mechanism of the deformation unit is verified to ensure that each process factor characteristic is dependent on and only dependent on the corresponding deformation unit, and there is no cross-dependent or missing dependent situation, such as the die gap characteristic only belongs to the end forming unit and the transition segment deformation unit, and does not belong to other deformation units. After the design is completed, the deformation unit-process factor characteristic dependent relationship model is generated, which is a structured dependent relationship set, including the number of all deformation units, the corresponding material / equipment / mold process factor characteristic details, the dependent relationship judgment basis, the cross-dependent verification result and other core contents, clearly defining the exclusive dependent process factor characteristic range of each deformation unit, and realizing the accurate dependent mapping of the deformation unit and the process factor characteristic.

[0094] Further, step S3 comprises the following steps:

[0095] Step S31: control equipment running state characteristic analysis is performed on the control equipment running data to generate control equipment running state characteristic data;

[0096] In the embodiment of the present application,

[0097] The control equipment operation state feature analysis of the control equipment operation data is carried out, and the equipment operation state feature quantization analysis is adopted, and the analysis dimensions cover three core dimensions of operation stability, parameter fluctuation characteristics and driving consistency. Taking the control equipment operation data as the core input, the data contains dynamic parameters such as driving torque, output power, displacement, running speed, forming area temperature; first, the time series data of various operation parameters is preprocessed, and the abnormal noise data is removed, and the data is regularized according to 10ms time interval; then the core feature indexes of various parameters are extracted, the driving torque fluctuation value (the difference between the maximum value and the minimum value in unit time), the power stability coefficient (the ratio of the standard deviation to the average value), the displacement deviation (the difference between the actual displacement and the reference displacement), the speed fluctuation rate (the ratio of the fluctuation amplitude to the reference speed), the temperature uniformity (the temperature difference of each monitoring point) are calculated; the feature judgment standard is set according to the operation parameters of different equipment components, the torque fluctuation value of the driving component is ≤0.5N·m, the power stability coefficient is ≤0.02, the displacement deviation of the transmission component is ≤0.01mm, the speed fluctuation rate is ≤0.01, the temperature uniformity of the heating component is ≤2℃; the extracted feature indexes are compared and analyzed with the judgment standard, and the control equipment operation state feature data classified according to the equipment components and arranged in time sequence are integrated, which contains the equipment operation core feature indexes, the feature compliance condition and the parameter fluctuation trend of each time node, and the dynamic change law of the equipment operation state is clearly quantized.

[0098] Step S32: control target local area analysis according to control equipment operation state feature data, and generate control target local area data;

[0099] In the embodiment of the present application, the control target local area is analyzed according to the control equipment running state feature data, and abnormal running feature data exceeding the judgment standard is screened out, including torque fluctuation exceeding the standard, displacement deviation being too large, temperature uniformity being insufficient and the like; based on the association logic of the equipment driving and the variable unit of the card forming, the abnormal running feature data is mapped to the corresponding card forming area, and the association link of the equipment running abnormality and the card variable unit is established, such as the driving torque fluctuation exceeding the standard corresponding to the inside bending unit and the outside stretching unit in the bending forming stage, and the displacement deviation being too large corresponding to the straight section unit and the end positioning unit; for the associated variable unit area, the influence degree on the overall forming quality is analyzed, the importance of the variable unit is determined in combination with the division of the variable unit in S14, and the control target local area is determined, such as the inside bending unit and the outside stretching unit being listed as a first control target area, and the straight section unit being listed as a second control target area; the spatial three-dimensional coordinates of each control target local area, the corresponding equipment running parameter type, the abnormal feature performance and the monitoring priority are recorded, and control target local area data is generated, which realizes the accurate association of the equipment running state and the card forming area, and provides clear target guidance for subsequent targeted geometric change analysis.

[0100] Step S33: card local geometric change feature analysis is performed on the card change data through the control target local area data, and card local geometric change feature data is generated;

[0101] In the embodiment of the present application, the card local geometric change feature analysis is performed on the card change data through the control target local area data, the morphological change image data and the size change data of the corresponding area are extracted from the card change data according to the spatial coordinates of the control target local area; the image contour extraction and fitting method is used to process the morphological change image of the target area, the contour curves at different time nodes are obtained, and the morphological deviation of the contour curves and the reference contour is calculated, such as the profile deviation of the inside bending unit and the angle deviation of the end forming unit; based on the size change data, the deformation amount of the actual size and the reference size of the target area is calculated, including the length deformation amount, the width deformation amount and the thickness deformation amount; the deformation amount in unit time is calculated according to the time interval, and the deformation rate of each target area is obtained; the time sequence change law of the deformation amount and the deformation rate is analyzed, and the deformation stable stage and the abnormal mutation stage are identified; the extracted local deformation amount, deformation rate, morphological deviation and size stability are classified and integrated according to the control target local area, and card local geometric change feature data is generated, which accurately describes the geometric change dynamic characteristics of the control target area, and is accurately associated with the corresponding equipment running state feature data through the time stamp.

[0102] Step S34: Design the card deformation unit timing window according to the card deformation digitalization unit data, and perform window division processing on the control equipment operation state characteristic data and the corresponding card local geometric change characteristic data through the card deformation unit timing window, to respectively generate window control equipment operation state characteristic data and corresponding window card local geometric change characteristic data.

[0103] In the embodiment of the application, the card deformation unit timing window is designed according to the card deformation digitalization unit data, the forming cycle data of each deformation unit is extracted, and the corresponding card deformation unit timing window is designed according to the forming cycle. The window duration strictly matches the forming cycle of each deformation unit. For example, if the forming cycle of the inner bending unit is 0.8s, the corresponding timing window duration is set to 0.8s, and if the forming cycle of the outer stretching unit is 0.7s, the timing window duration is set to 0.7s, ensuring that each timing window completely covers the forming process of the corresponding deformation unit. A one-to-one correspondence between the timing window and the deformation unit is established according to the deformation unit number. Each timing window is assigned a unique window number, and the window number rule is deformation unit number + timing sequence number. Based on the designed timing window, the control equipment operation state characteristic data and the card local geometric change characteristic data are synchronously windowed. The division process strictly matches the timestamp for accurate matching, ensuring that the equipment operation data and geometric change data in the same window correspond to the same forming period of the same deformation unit. After the division is completed, the correlation of the window data is checked, and abnormal data with unmatched timestamps are removed. Finally, window control equipment operation state characteristic data and window card local geometric change characteristic data are generated respectively. Both types of data are classified and integrated according to the window number, realizing the accurate aggregation of equipment operation data and geometric change data according to the deformation unit timing.

[0104] Step S35: Map the window control equipment operation state characteristic data and the corresponding window card local geometric change characteristic data to the card deformation digitalization unit data for deformation unit production behavior characteristic analysis processing, to generate deformation unit production behavior characteristic data.

[0105] In the embodiment of the present application, the window control device running state feature data and the corresponding window card issuing local geometric change feature data are mapped to the card issuing deformation digital unit data to perform production behavior feature analysis and processing of the deformation unit, and the analysis dimensions cover four core dimensions of time sequence running behavior, dynamic deformation behavior, device-deformation correlation behavior and quality trend behavior. According to the corresponding relationship between the window number and the deformation unit number, the two types of window data are accurately mapped to the corresponding deformation unit in the card issuing deformation digital unit data; the window device running feature and the window geometric change feature of the same deformation unit are fused, and the correlation coefficient of the device running parameter and the geometric deformation amount is calculated, such as the correlation coefficient of the forming pressure and the bending deformation amount, the correlation coefficient of the driving speed and the size precision; the running behavior change law of the same deformation unit in different time sequence windows is analyzed, including the time sequence fluctuation law of the device running parameter, the time sequence evolution law of the geometric deformation amount, and the time sequence change law of the correlation coefficient; the stable feature and the abnormal feature in the production behavior of the deformation unit are identified, such as the stable behavior of stable device running and geometric deformation meeting the benchmark, and the abnormal behavior of geometric deformation exceeding the standard caused by device parameter fluctuation; the fused and analyzed features are integrated according to the deformation unit number to generate deformation unit production behavior feature data, which contains time sequence running feature indexes, dynamic deformation feature indexes, device-deformation correlation coefficients and behavior stability judgment results of each deformation unit, and completely describes the dynamic behavior law of each deformation unit in the actual production process, providing core data support for the subsequent correlation analysis of process parameter change and deviation response.

[0106] Further, step S4 comprises the following steps:

[0107] Step S41: performing production behavior trend state analysis of the deformation unit on the deformation unit production behavior feature data to generate deformation unit production behavior trend state data;

[0108] In the embodiment of the present application, the production behavior trend state analysis of the deformation unit is carried out, and the time sequence and the coupling analysis technology of the correlation characteristics are used to realize the process. The time sequence sampling interval of the technology is 10 ms, the correlation coefficient calculation accuracy is ±0.001, the trend determination threshold is 0.8 (strong correlation) and 0.5 (moderate correlation), the influence degree quantization accuracy is ±0.005, and the analysis dimensions cover four core dimensions of deformation autocorrelation trend, device parameter correlation trend, deformation development direction and stable state determination. The analysis operation takes the deformation unit production behavior characteristic data generated in step S35 as the core input, first extracts the time sequence running characteristic index and the dynamic deformation characteristic index of each deformation unit, focuses on the deformation amount data of the same deformation unit in adjacent time periods, calculates the Pearson correlation coefficient of the deformation amount of the adjacent 10 time nodes, the correlation coefficient determination criterion of the inner bending unit is 0.85, and the correlation coefficient determination criterion of the outer stretching unit is 0.83. When the correlation coefficient is greater than or equal to 0.8, it is determined that the deformation trend continuity is strong; based on the correlation coefficient distribution law, the time sequence change trend of the deformation amount is mined, and the increasing trend, the decreasing trend and the stable trend are distinguished. If the deformation amount correlation coefficient of the inner bending unit is 0.86 and the deformation amount is gradually increased by 0.002 mm for 5 windows, it is determined that the deformation trend is increasing. The parameter change data such as driving torque, speed and temperature in the control device running state characteristics are synchronously extracted, the correlation degree between the device parameter change and the deformation trend is calculated, the correlation degree weight of the driving torque change and the deformation trend of the inner bending unit is set to 0.45, the correlation degree weight of the speed change is 0.35, and the correlation degree weight of the temperature change is 0.2. The influence degree of the device parameter on the deformation trend is determined through the correlation degree quantization analysis, for example, the torque fluctuation of 0.3 N·m corresponds to the increase of 0.001 mm / window in the increasing amplitude of the deformation trend. Combined with the deformation trend and the device parameter influence law, the deformation development direction, the stable state duration and the change rate of each deformation unit are determined, and the deformation unit production behavior trend state data is generated. The data is classified according to the deformation unit number and contains the time sequence correlation coefficient distribution, the deformation trend type, the trend continuation time length, the device parameter influence weight and the stable state identifier.

[0109] Step S42: According to the deformation unit production behavior trend state data, the production behavior quality deviation analysis is carried out, and the production behavior quality deviation data is generated.

[0110] In the embodiment of the application, production behavior quality deviation analysis is carried out, and the process is realized based on trend state-target threshold comparison analysis technology. The deviation calculation precision is set to ±0.001 mm, the shape accuracy tolerance threshold is ±0.015 mm, the size deviation threshold is ±0.01 mm, the angle deviation threshold is ±0.1°, and the analysis dimensions cover four core dimensions of deviation identification, deviation quantification, deviation classification and deviation time period positioning. The analysis operation takes the trend state data of the deformation unit production behavior and the forming target data as input, establishes a comparison link of the trend state data and the forming target data, and compares the dynamic deformation characteristic indexes (profile, size, angle, etc.) of each deformation unit with the corresponding forming target threshold at each time node. Production behavior data that exceeds the shape accuracy tolerance, size deviation threshold and angle deviation threshold is screened out, such as the profile deviation of the inner side bending unit of 0.018 mm and the angle deviation of the end forming unit of 0.15°. The screened deviation data is quantitatively analyzed, the deviation amount of the actual value and the target value is calculated, and the deviation amount is divided into three categories according to the deviation amount, i.e. slight deviation (deviation amount ≤0.005 mm / 0.05°), moderate deviation (0.005 mm<deviation amount ≤0.01 mm / 0.05°<deviation amount ≤0.1°) and severe deviation (deviation amount >0.01 mm / deviation amount >0.1°). The specific time period of the deviation occurrence is traced through the time stamp, the equipment operation parameter state of the corresponding time period is associated, and the equipment parameter value at the time of the deviation occurrence is determined, such as the forming pressure of 7.5 MPa and the driving speed of 3.8 mm / s at the time of the severe deviation. The deviation type, deviation degree, occurrence time period and corresponding equipment parameter state are integrated to generate production behavior quality deviation data, which realizes accurate quantification and traceability of the quality deviation and provides clear basis for subsequent deviation attribution analysis.

[0111] Step S43: Perform quality deviation attribution relationship analysis of the deformation unit according to the card deformation digital unit data and the production behavior quality deviation data, and generate deformation unit quality deviation attribution relationship data;

[0112] In the embodiment of the application, the quality deviation attribution relationship analysis of the deformation unit is performed according to the digital data of the card-forming deformation unit and the production behavior quality deviation data, and the key information such as the deviation occurrence area space coordinates, the deviation type, and the deviation time period in the production behavior quality deviation data is extracted; the space coordinate matching link of the deviation area and the deformation unit is established based on the deformation unit number, the space three-dimensional coordinates, and the deformation stage to which the information in the digital data of the card-forming deformation unit belongs; the deviation area coordinates and the deformation unit coordinates are compared one by one according to the deformation unit number, and when the coincidence degree is greater than or equal to 95%, it is determined that the deviation is attributed to the corresponding deformation unit, for example, the angle deviation (coordinates X: 50-60 mm, Y: 15-25 mm, Z: 5-10 mm) of the end forming area is completely matched with the end forming unit coordinates, and it is determined that the attribution is the end forming unit; the matched attribution relationship is verified to confirm the consistency of the deviation occurrence time period and the deformation unit forming cycle, for example, the forming cycle of the inner bending unit is 0.8 s, and if the deviation occurrence time period is within the cycle, the verification is passed; the deformation unit number, the attribution verification result, the deviation type, and the degree corresponding to each quality deviation are recorded, the deformation unit quality deviation attribution relationship data is generated, the data realizes one-to-one correspondence between the quality deviation and the deformation unit, and the deviation occurrence of each deformation unit is clear.

[0113] Step S44: The process parameter change and deviation response correlation analysis processing of the deformation unit is performed through the deformation unit quality deviation attribution relationship data, and the deformation unit parameter change-deviation response correlation data is generated.

[0114] In the embodiment of the present application, the process parameter change of the change unit and the correlation analysis processing of the deviation response are carried out by changing the quality deviation attribution relationship data of the change unit, the digital change unit data of the coil forming is grouped according to the process parameter value change condition, the interval of 2-8 MPa of the forming pressure is divided into 12 groups at an interval of 0.5 MPa, the interval of 0.5-4 mm / s of the driving speed is divided into 12 groups at an interval of 0.3 mm / s, and the interval of 80-150 ℃ of the heating temperature is divided into 8 groups at an interval of 10 ℃; the process parameter change direction and amplitude of each group of data are identified, such as the forming pressure is identified as “positive change, amplitude 0.5 MPa” when the forming pressure is increased from 3 MPa to 3.5 MPa, and the forming pressure is identified as “reverse change, amplitude 0.5 MPa” when the forming pressure is decreased from 4 MPa to 3.5 MPa; based on the quality deviation attribution relationship data of the change unit, the deviation occurrence times, the deviation degree and the deviation type of each change unit under each parameter condition are counted, such as the mild deviation of the outer stretching unit occurs 3 times and the moderate deviation occurs once in the forming pressure group of 3.5-4.0 MPa; the quantitative correlation coefficient of the parameter change amount and the deviation degree of each group is calculated, the profile deviation amount of the inner bending unit is increased by 0.003 mm when the forming pressure is positively changed by 0.5 MPa, and the correlation coefficient is 0.6; based on the statistical results and the quantitative coefficient, the corresponding relationship between the process parameter change amount (change direction and amplitude) and the deviation degree and the deviation type is established, the deviation response law under different parameter change conditions is determined, the change unit parameter change-deviation response correlation data is generated, and the data provides a core quantitative basis for subsequent process parameter optimization and equipment adjustment.

[0115] Further, the step S41 comprises the following steps:

[0116] Step S411: Perform local deformation self-correlation trend feature analysis on the window coil local geometry change feature data to generate local deformation self-correlation trend feature data of the coil;

[0117] In the embodiment of the application, the local deformation self-correlation trend feature analysis of the card is carried out, and a time series self-correlation quantification analysis technology is used to realize the process. The technology configures a time series analysis window length of 0.1s, a self-correlation coefficient calculation accuracy of ±0.001, a trend continuation judgment threshold of a correlation coefficient of 5 consecutive windows ≥0.8, a deformation amount change amplitude threshold of 0.001mm / window, and an analysis dimension covering three core dimensions of adjacent time series correlation, deformation trend continuity, and deformation amount gradient change. The window card local geometric change feature data is taken as the core input, which contains the local deformation amount, deformation rate, and profile deviation of each deformation unit at different time series windows. First, the window data is classified and regularized according to the deformation unit number, the time series deformation amount data of each deformation unit is extracted, and the data is sliced according to the 0.1s time series window. Then, the Pearson self-correlation coefficient of the deformation amount of the adjacent two time series windows is calculated, the inner bending unit self-correlation coefficient judgment criterion is 0.85, the outer stretching unit is 0.83, and the straight line segment unit is 0.82. The time series correlation characteristics of the deformation amount are identified through the distribution law of the continuous window self-correlation coefficient. Based on the self-correlation coefficient result, the deformation trend type is divided. When the correlation coefficient of 5 consecutive windows is ≥0.8 and the deformation amount is gradually increased by 0.001mm-0.003mm, it is determined as an increasing trend. When the correlation coefficient of 5 consecutive windows is ≥0.8 and the deformation amount is gradually reduced by 0.001mm-0.003mm, it is determined as a decreasing trend. When the correlation coefficient of 5 consecutive windows is ≥0.8 and the deformation amount fluctuation amplitude is ≤0.0005mm, it is determined as a stable trend. The deformation amount gradient change value (deformation amount change amplitude per unit time) under each trend is calculated, the deformation trend change rate is converted combined with the window length, the card local deformation self-correlation trend feature data classified according to the deformation unit number and arranged in time series is generated, and the data contains the self-correlation coefficient, trend type preliminary judgment result, deformation amount gradient change value, and trend continuation window number of each time node.

[0118] Step S412: According to the window control equipment running state feature data, the influence feature analysis of the card local deformation trend is carried out, and the card local deformation influence feature data is generated.

[0119] In the embodiment of the application, the influence feature analysis of the local deformation trend of the card issuing bureau is performed according to the window control equipment running state feature data, the core running parameter change data of the control equipment in each time window is extracted, including the forming pressure fluctuation, the servo drive speed change, and the forming area temperature change, wherein the forming pressure fluctuation is extracted with an accuracy of 0.01 MPa, the drive speed change is extracted with an accuracy of 0.01 mm / s, and the temperature change is extracted with an accuracy of 0.1 ℃; then, the time sequence association link of the equipment parameter change data and the deformation trend data is established according to the deformation unit number, the correlation coefficient of each type of parameter change and the deformation amount change amplitude of the corresponding deformation unit is calculated, the correlation coefficient calculation standard of the forming pressure change of the inner bending unit and the deformation amount is 0.6, the correlation coefficient standard of the drive speed change and the deformation amount is 0.4, and the correlation coefficient standard of the temperature change and the deformation amount is 0.2; the influence weight of each type of parameter is allocated based on the correlation coefficient size, the forming pressure change influence weight accounts for 50%, the drive speed change accounts for 35%, and the temperature change accounts for 15%, the comprehensive influence degree of each equipment parameter on the deformation trend is obtained through weighted calculation; the parameter change amplitude, the correlation coefficient, the influence weight, and the comprehensive influence degree are integrated to generate the card issuing bureau local deformation influence feature data classified according to the deformation unit number and the time window.

[0120] Step S413: Perform the production behavior trend state analysis of the deformation unit production behavior feature data through the card issuing bureau local deformation self-correlation trend feature data and the card issuing bureau local deformation influence feature data, and generate the deformation unit production behavior trend state data.

[0121] In the embodiment of the present application, the production behavior trend state of the deformation unit is analyzed by using the local deformation self-correlation trend characteristic data and the local deformation influence characteristic data of the card, and a multi-feature fusion trend determination technology is used to realize the process, the trend consistency verification threshold is set as the matching degree of the equipment influence characteristic and the self-correlation trend ≥ 90%, the stable state determination standard is that the deformation amount fluctuation of 10 continuous windows ≤ 0.0005 mm, the change rate calculation accuracy is ± 0.0001 mm / s, the precise alignment link of the three types of data is established according to the deformation unit number and the time sequence window number, and the self-correlation trend, the equipment influence characteristic and the production behavior characteristic of the same deformation unit and the same time sequence window are fully matched; then the self-correlation trend characteristic and the equipment influence characteristic are fused, the self-correlation trend type is compared and verified with the theoretical deformation trend under the influence of the equipment parameters, if the matching degree ≥ 90%, the trend effectiveness is confirmed, and if the matching degree < 90%, the trend determination result is corrected by re-tracing the adjacent window data; based on the verified effective trend, the deformation development direction (increasing, decreasing, stable) of the deformation unit, the trend continuation time length (the window number of the continuous effective trend is converted into the time length), the deformation change rate (the ratio of the total deformation amount to the trend continuation time length) are calculated, and the stability index in the production behavior characteristic data is combined to determine the stable state duration of each deformation unit and the trigger condition of the unstable state; all the characteristics after the fusion analysis are integrated according to the deformation unit number to generate the deformation unit production behavior trend state data, which contains the time sequence self-correlation coefficient distribution, the final trend type, the trend continuation time length, the deformation change rate, the equipment parameter influence weight, the stable state identification and the duration of each deformation unit, and the time sequence trend law of the production behavior of each deformation unit is clearly quantified.

[0122] Further, step S44 includes the following steps:

[0123] Step S441: Group processing of the card deformation digital unit data under the process parameter value change condition to generate grouped card deformation digital unit data;

[0124] In the embodiment of the application, the card forming deformation digital unit data is grouped according to the process parameter value change condition, the card forming deformation digital unit data includes process parameter quantitative indicators and digital features corresponding to each deformation unit, such as forming pressure, driving speed and heating temperature. Firstly, the value range of all process parameters in the data is extracted, and the interval is divided according to the preset grouping interval. The forming pressure is divided into 12 continuous intervals (2.0-2.5 MPa, 2.5-3.0 MPa……7.5-8.0 MPa) with 0.5 MPa as the interval according to the interval of 2 MPa-8 MPa. The driving speed is divided into 12 continuous intervals (0.5-0.8 mm / s, 0.8-1.1 mm / s……3.7-4.0 mm / s) with 0.3 mm / s as the interval according to the interval of 0.5 mm / s-4.0 mm / s. The heating temperature is divided into 8 continuous intervals (80-90℃, 90-100℃……140-150℃) with 10℃ as the interval according to the interval of 80℃-150℃. Then, each parameter interval is uniquely identified, and the identification rule is parameter type abbreviation+interval serial number (such as pressure interval identification Y-01 to Y-12, speed interval identification S-01 to S-12, and temperature interval identification T-01 to T-08). The process parameter value in the card forming deformation digital unit data is accurately matched with each interval range, and the digital features (unit number, spatial parameter, deformation constraint, etc.) of the corresponding deformation unit are classified into the matched parameter interval, and the abnormal data with parameter value beyond the range is eliminated. Finally, the data is classified and integrated according to the parameter interval to generate grouped card forming deformation digital unit data, which includes core contents such as each group identification, corresponding process parameter interval range, complete digital features of the deformation unit, and the number of data in the group.

[0125] Step S442: Process parameter value change direction identification is performed on the grouped card forming deformation digital unit data to generate identification grouped card forming deformation digital unit data.

[0126] In the embodiment of the present application, the process parameter value change guide mark is performed on the grouped card forming and deforming digital unit data, and meanwhile the process parameter reference value (molding pressure reference value 3 MPa, driving speed reference value 2 mm / s, heating temperature reference value 110°C) determined in step S11 is associated; first, the difference value between the interval median value of each group of process parameters and the corresponding reference parameter is calculated, the difference value between the molding pressure interval median value and 3 MPa, the difference value between the driving speed interval median value and 2 mm / s, and the difference value between the heating temperature interval median value and 110°C are respectively taken as the parameter change judgment basis; the change direction judgment rule is set, when the difference value is greater than 0, it is judged as positive change, when the difference value is less than 0, it is judged as negative change, and when the difference value is equal to 0, it is judged as no change; the change amplitude of each group of parameters relative to the reference parameter is calculated, that is, the absolute value of the difference value, for example, the molding pressure Y-05 group interval is 4.5-5.0 MPa, the median value is 4.75 MPa, and the difference value with the reference 3 MPa is 1.75 MPa, which is judged as positive change, and the change amplitude is 1.75 MPa; each group is given a standardized mark containing parameter type, change direction and change amplitude, and the mark format is "parameter type-change direction-change amplitude" (such as Y-positive-1.75 MPa, S-negative-0.5 mm / s); the mark information is accurately associated with the grouped card forming and deforming digital unit data, to ensure that all data of each group carry complete mark information, and after integrity verification, the marked grouped card forming and deforming digital unit data is generated, which contains all the contents of the original grouped data and the newly added process parameter change guide mark information.

[0127] Step S443: According to the change unit quality deviation attribution relationship data and the marked grouped card forming and deforming digital unit data, the process parameter change and deviation response of the change unit are associated and analyzed, and the change unit parameter change-deviation response association data is generated.

[0128] In the embodiment of the present application, according to the quality deviation attribution relationship data of the deformation unit and the deformation digital unit data of the identification grouping card, the process parameter change and deviation response of the deformation unit are analyzed, the precise association link of the two groups of data is established according to the deformation unit number and the process parameter grouping identification, and the identification data and the deviation data of the same deformation unit and the same process parameter grouping are completely matched; based on the association link, the quality deviation characteristics of the corresponding deformation unit under each process parameter grouping are counted, including the number of deviation occurrences, the proportion of each type of deviation (mild, moderate, severe), the maximum and minimum deviation amount, the time sequence distribution of the deviation, such as the grouping identified as Y-positive-1.75MPa, the mild deviation of the inner bending unit occurs 4 times, the moderate deviation occurs 2 times, the maximum profile deviation is 0.012mm, and the deviation occurs in the molding period corresponding to the grouping; the quantitative correlation coefficient of the change direction and amplitude of each process parameter and the deviation degree is calculated, the Pearson correlation coefficient calculation method is used, the correlation coefficient of the positive change amplitude of the molding pressure and the profile deviation amount of the inner bending unit is 0.68, and the correlation coefficient of the reverse change amplitude of the driving speed and the size deviation amount of the straight line segment unit is 0.57; based on the correlation coefficient, the strong correlation (correlation degree≥0.7), the moderate correlation (0.5≤correlation degree<0.7), and the weak correlation (correlation degree<0.5) of the parameter change-deviation response combination are screened out, and the one-to-one corresponding association rule of the process parameter change amount (direction, amplitude) and the deviation type and the deviation degree is constructed; the association link information, the deviation statistical result, the quantitative correlation coefficient, and the association rule are integrated to generate the deformation unit parameter change-deviation response association data, which clearly defines the deviation response rule of each deformation unit under different process parameter change conditions, and provides a direct quantitative correlation basis for the subsequent digital optimization of the flat wire motor card forming parameters.

[0129] Further, step S5 comprises the following steps:

[0130] Step S51: according to the change unit parameter change-deviation response association data, the parameter influence characteristic analysis of the forming deviation is performed, and the forming deviation parameter influence characteristic data is generated;

[0131] In the embodiment of the present application, the parameter influence characteristic analysis of forming deviation is performed according to the variation-unit parameter variation-deviation response correlation data, the parameter influence significance, the deviation degree correlation strength and the multi-parameter interaction influence characteristics are analyzed. The strong correlation and medium correlation combinations with a correlation degree greater than or equal to 0.5 in the variation-unit parameter variation-deviation response correlation data are extracted, and the variation data of three types of core process parameters, i.e., forming pressure, driving speed and heating temperature, and the deviation data of the corresponding deformation units are screened out. According to the parameter variation gradient, the intervals are divided, the forming pressure is divided according to the 0.05 MPa gradient, the driving speed is divided according to the 0.03 mm / s gradient, and the heating temperature is divided according to the 1 ℃ gradient. The influence coefficient of the parameter variation on the deviation in each gradient interval is calculated. The forming pressure of the inner bending unit is increased by 0.05 MPa, the profile deviation is increased by 0.0003 mm, the influence coefficient is 0.68, the driving speed of the outer stretching unit is reduced by 0.03 mm / s, the size deviation is increased by 0.0002 mm, and the influence coefficient is 0.57. The comprehensive influence of the multi-parameter interaction on the deviation is analyzed, and the comprehensive influence coefficient when the forming pressure and the heating temperature change simultaneously is calculated. For example, when the forming pressure is positively changed by 0.5 MPa and the heating temperature is positively changed by 10 ℃, the comprehensive influence coefficient of the inner bending unit deviation is 0.72. According to the deformation unit number, the deviation type classification, the parameter influence coefficient, the variation gradient interval and the comprehensive influence characteristics, the forming deviation parameter influence characteristic data is generated, which contains the core influence parameters, the quantitative influence coefficient, the parameter variation gradient threshold and the multi-parameter interaction influence law corresponding to each deformation unit and different deviation types.

[0132] Step S52: The forming deviation parameter influence characteristic data is processed by a preset decision tree algorithm to design the deformation unit tree relationship node of the forming deviation parameter and the control equipment adjustment, so as to establish the forming deviation parameter-control equipment adjustment relationship tree model.

[0133] In the embodiment of the present application, the relationship node design of the forming deviation parameter and the control equipment adjustment variable unit tree is carried out, and the preset decision tree algorithm is adopted to realize the process, the algorithm is configured with node splitting threshold accuracy of ±0.001 mm (deviation amount), ±0.01 MPa (pressure), and ±0.01 mm / s (speed), the tree depth is set to be ≤8 layers, the minimum sample number of the leaf node is set to be 5 groups, and the node purity judgment threshold is ≥0.9. The design operation takes the forming deviation parameter influence characteristic data as the core input, determines the core process parameter change type (forming pressure change, driving speed change, heating temperature change) as the root node of the decision tree, and the root node splitting is based on the parameter influence coefficient. The parameter with an influence coefficient ≥0.6 is preferentially selected as the initial splitting branch of the root node; the intermediate nodes are hierarchically divided according to the variable unit number and the deviation degree (mild, moderate, and severe), the first layer intermediate node is the variable unit number, the second layer intermediate node is the deviation degree, the node splitting threshold is set according to the deviation amount threshold in the forming deviation parameter influence characteristic data, the mild deviation node splitting threshold is ≤0.005 mm / 0.05°, the moderate deviation is 0.005 mm<deviation amount≤0.01 mm / 0.05°<deviation amount≤0.1°, and the severe deviation is >0.01 mm / >0.1°; the leaf node is the control equipment adjustment direction, and the adjustment direction is set based on the correlation between the parameter change and the deviation. For example, when the forming pressure positively changes to cause severe deviation, the leaf node adjustment direction is “forming pressure reverse adjustment”, and when the driving speed reversely changes to cause moderate deviation, the adjustment direction is “driving speed positively adjustment”; the tree structure is constructed according to the above node hierarchical relationship, the parent-child association logic and the splitting rule of each node are determined, the forming deviation parameter-control equipment adjustment relationship tree model is generated, the model contains complete node hierarchy, splitting threshold, and adjustment direction mapping relationship, and the accurate link mapping from the forming deviation parameter to the control equipment adjustment direction is realized.

[0134] Step S53: Perform forming deviation parameter constraint characteristic analysis through the forming deviation parameter influence characteristic data, generate forming deviation parameter constraint characteristic data, and perform control equipment adjustment constraint processing on the tree nodes of the forming deviation parameter-control equipment adjustment relationship tree model by using the forming deviation parameter constraint characteristic data, to obtain a forming deviation parameter-control equipment adjustment constraint tree model.

[0135] In the embodiment of the present application, the effective variation range of each process parameter in the forming deviation parameter influence characteristic data is extracted, and the parameter constraint threshold is defined in combination with the reference value range. The upper limit of the forming pressure constraint is 8 MPa, and the lower limit is 2 MPa. The upper limit of the driving speed constraint is 4 mm / s, and the lower limit is 0.5 mm / s. The upper limit of the heating temperature constraint is 150 DEG C, and the lower limit is 80 DEG C. The correlation between the constraint threshold and the deviation is analyzed, and the deterioration degree of the deviation when exceeding the constraint threshold is verified. For example, when the forming pressure exceeds 8 MPa, the incidence of severe deviation of the inner side bending unit increases by 60%, and when the forming pressure is lower than 2 MPa, the incidence of size deviation of the straight line segment unit increases by 50%. The constraint threshold, the constraint correlation verification result, and the deviation deterioration law are integrated to generate forming deviation parameter constraint characteristic data. Based on the data, the leaf nodes of the forming deviation parameter-control equipment adjustment relationship tree model are constrained and processed, the adjustment direction exceeding the constraint threshold is eliminated, the adjustment parameter range is corrected, for example, the "forming pressure reverse adjustment" is corrected to "forming pressure reverse adjustment to the interval of 3.5 MPa-4.0 MPa", and it is ensured that the adjustment parameters of all tree nodes are within the effective constraint range, so that the forming deviation parameter-control equipment adjustment constraint tree model is obtained.

[0136] Step S54: Analyze the parameter sensitivity weight of the forming deviation by the forming deviation parameter influence characteristic data, and perform tree node weight optimization processing on the forming deviation parameter-control equipment adjustment constraint tree model by using the parameter sensitivity weight, so as to obtain a forming deviation parameter-control equipment adjustment optimization tree model.

[0137] In the embodiment of the application, the sensitivity weight analysis of the forming deviation parameter and the optimization of the tree node weight are carried out, the parameter sensitivity quantification analysis technology is adopted to realize the process, the sensitivity threshold is ≥0.6 (high sensitivity), 0.3≤sensitivity<0.6 (medium sensitivity), and <0.3 (low sensitivity), and the analysis dimensions cover three core dimensions of parameter sensitivity quantification, weight priority sorting, and node weight adaptation. Taking the forming deviation parameter influence characteristic data as the core input, the sensitivity analysis method is used to calculate the sensitivity coefficient of each process parameter change to the forming deviation, the sensitivity coefficient of the forming pressure change to the inner bending unit profile deviation is 0.72, the sensitivity coefficient of the driving speed change to the linear segment unit size deviation is 0.58, and the sensitivity coefficient of the heating temperature change to the end forming unit angle deviation is 0.35; based on the sensitivity coefficient, the weight level is divided, the high sensitivity parameter weight distribution accounts for 60%, the medium sensitivity accounts for 30%, and the low sensitivity accounts for 10%, for example, the forming pressure is a high sensitivity parameter, the corresponding tree node weight is assigned to 0.6, the driving speed is a medium sensitivity parameter, the weight is assigned to 0.3, and the heating temperature is a low sensitivity parameter, the weight is assigned to 0.1; the weight value is assigned to the corresponding node of the forming deviation parameter-control equipment adjustment constraint tree model, the root node is sorted according to the core parameter sensitivity weight, the intermediate node is adapted according to the deformation unit deviation sensitivity weight, and the leaf node is optimized according to the parameter sensitivity weight corresponding to the adjustment direction; through the weight optimization, the model preferentially responds to the deviation adjustment demand corresponding to the high sensitivity parameter, obtains the forming deviation parameter-control equipment adjustment optimization tree model, the model includes the weight assignment and priority sorting of each node, and the response priority of the high-impact parameter adjustment is improved.

[0138] Step S55: The forming deviation parameter-control equipment adjustment optimization tree model is used to analyze the control equipment optimization adjustment parameter of the deformation unit production behavior characteristic data for the card forming, and the control equipment optimization adjustment parameter is generated.

[0139] In the embodiment of the present application, the control equipment optimization adjustment parameter analysis of the forming deviation parameter-control equipment adjustment optimization tree model is used to control the forming of the variable unit production behavior characteristic data, the variable unit production behavior characteristic data is accurately matched with the node level of the optimization tree model according to the variable unit number and the deviation type, such as the severe profile deviation data of the inner bending unit matching the node link corresponding to “forming pressure change-inner bending unit-severe deviation” in the model; based on the weight, adjustment direction and constraint range of the matched node, the specific optimization adjustment parameter is calculated, the influence coefficient in the influence characteristic data of the forming deviation parameter is combined, and the required parameter adjustment amount is reversely deduced, such as 0.8 MPa of reverse adjustment is required for the inner bending unit due to 1.75 MPa of positive change of the forming pressure to produce 0.012 mm of profile deviation, 0.68 of influence coefficient, and the pressure is stabilized at 3.95 MPa after adjustment; the optimization adjustment parameter calculated is verified, the corresponding deviation amount change after simulation adjustment is simulated, and it is ensured that the deviation amount after adjustment is less than or equal to 0.005 mm (light deviation threshold); the optimization adjustment parameter is integrated according to the variable unit number and the equipment parameter type, the adjustment direction, adjustment amount and stable value after adjustment of each parameter are clear, the control equipment optimization adjustment parameter is generated, and the data is directly used for intelligent iterative optimization of the flat wire motor forming control equipment parameter, and the forming quality is accurately improved.

[0140] Therefore, from any viewpoint, the embodiments should be considered as exemplary and non-limiting, the scope of the present application being defined by the appended claims and not by the above description, therefore all variations falling within the meaning and scope of the equivalent elements of the application file are intended to be included in the present application.

[0141] The above description is merely one specific implementation of the present application, which enables those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data analysis-based digital optimization method for forming parameters of a flat wire motor, characterized in that, The method comprises the following steps: Step S1: obtaining the flat wire motor hairpin forming process parameters to be formed; Based on the flat wire motor hairpin forming process parameters, the digital unit conversion processing of the hairpin deformation stage process parameters is performed to generate hairpin deformation digital unit data; Wherein, the flat wire motor hairpin process parameters include hairpin material physical parameters, control equipment setting parameters and hairpin forming target data, step S1 includes the following steps: Step S11: obtaining the flat wire motor hairpin forming process parameters to be formed; Step S12: according to the flat wire motor hairpin process parameters, the hairpin deformation mechanism simulation data is generated by performing hairpin deformation mechanism simulation processing; Step S13: the hairpin deformation mechanism simulation data is divided into stages to generate hairpin deformation mechanism stage simulation data; Step S14: according to the hairpin deformation mechanism stage simulation data, the unit architecture conversion processing of each hairpin deformation mechanism stage is performed to generate the hairpin deformation mechanism unit architecture data, wherein step S14 includes unit architecture conversion according to the initial positioning stage, bending deformation stage and shaping stage of the three continuous deformation stages corresponding to the hairpin deformation mechanism stage simulation data, each deformation stage is divided into several independent deformation units according to the hairpin structure part to construct the corresponding deformation unit architecture of each stage, and the hairpin deformation mechanism unit architecture data is generated; Step S15: according to the flat wire motor hairpin process parameters, the hairpin process attribute type analysis data is generated by performing hairpin process attribute type analysis processing, wherein the hairpin process attribute type analysis data includes material attribute analysis data, control equipment analysis data and hairpin die restriction analysis data; Step S16: based on the hairpin process attribute type analysis data, the dependent mapping relationship design of each deformation unit and process factor characteristic is performed to generate the deformation unit-process factor characteristic dependent relationship model; Wherein, step S16 includes: Step S161: according to the material attribute analysis data, the material deformation response factor characteristic analysis of the hairpin deformation is performed to generate the material deformation response factor characteristic data; Step S162: according to the control equipment analysis data, the control equipment driving factor characteristic analysis of the hairpin deformation is performed to generate the control equipment driving factor characteristic data; Step S163: according to the hairpin die restriction analysis data, the geometric restriction factor characteristic analysis of each local area of the hairpin is performed to generate the hairpin geometric restriction factor characteristic data; Step S164: through the material deformation response factor characteristic data, the control equipment driving factor characteristic data and the hairpin geometric restriction factor characteristic data, the dependent mapping relationship design of each deformation unit and process factor characteristic is performed to generate the deformation unit-process factor characteristic dependent relationship model; Step S17: the hairpin deformation mechanism unit architecture data is transmitted to the deformation unit-process factor characteristic dependent relationship model to perform digital process characteristic mapping processing of each deformation unit to generate hairpin deformation digital unit data; Step S2: using the multi-source sensor built-in in the flat wire motor hairpin forming control equipment to perform control equipment running data and corresponding hairpin change data monitoring processing of forming production, respectively obtaining control equipment running data and corresponding hairpin change data; Step S3: Based on the control device operation data and the corresponding card change data, the card deformation digital unit data is processed to analyze the production behavior characteristics of the deformation unit, and deformation unit production behavior characteristic data is generated; Wherein, step S3 includes the following steps: Step S31: Control device operation state characteristic analysis is performed on the control device operation data to generate control device operation state characteristic data; Step S32: Control target local area analysis is performed according to the control device operation state characteristic data to generate control target local area data; Step S33: Card local geometric change characteristic analysis is performed on the card change data through the control target local area data to generate card local geometric change characteristic data; Step S34: The card deformation unit time window is designed according to the card deformation digital unit data, and the control device operation state characteristic data and the corresponding card local geometric change characteristic data are window divided through the card deformation unit time window to generate window control device operation state characteristic data and corresponding window card local geometric change characteristic data respectively; Step S35: The window control device operation state characteristic data and the corresponding window card local geometric change characteristic data are mapped to the card deformation digital unit data to analyze the production behavior characteristics of the deformation unit, and deformation unit production behavior characteristic data is generated; Step S4: The process parameter change and deviation response correlation analysis of the deformation unit is performed through the deformation unit production behavior characteristic data to generate deformation unit parameter change-deviation response correlation data; Step S5: Based on the preset decision tree algorithm and the deformation unit parameter change-deviation response correlation data, a molding deviation parameter-control device adjustment optimization tree model is established; The control device optimization adjustment parameter analysis of the card forming of the deformation unit production behavior characteristic data is performed by using the molding deviation parameter-control device adjustment optimization tree model to generate the control device optimization adjustment parameter; Wherein, step S5 includes the following steps: Step S51: According to the deformation unit parameter change-deviation response correlation data, the parameter influence characteristic analysis of the molding deviation is performed to generate molding deviation parameter influence characteristic data; Step S52: The molding deviation parameter and the deformation unit tree relationship node design of the control device adjustment are performed on the molding deviation parameter influence characteristic data through the preset decision tree algorithm to establish a molding deviation parameter-control device adjustment relationship tree model; Step S53: The molding deviation parameter constraint characteristic analysis is performed through the molding deviation parameter influence characteristic data to generate molding deviation parameter constraint characteristic data, and the molding deviation parameter-control device adjustment relationship tree model is processed by the control device adjustment constraint of the tree node to obtain the molding deviation parameter-control device adjustment constraint tree model; Step S54: The parameter sensitivity weight of the molding deviation is analyzed through the molding deviation parameter influence characteristic data, and the parameter sensitivity weight is used to process the tree node weight optimization of the molding deviation parameter-control device adjustment constraint tree model to obtain the molding deviation parameter-control device adjustment optimization tree model; Step S55: generating control equipment optimization adjustment parameters by using the forming deviation parameter-control equipment adjustment optimization tree model to analyze the control equipment optimization adjustment parameters of the forming behavior characteristic data of the forming unit, and generating the control equipment optimization adjustment parameters.

2. The method for digital optimization of the flat wire motor lamination forming parameters based on data analysis according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: generating forming unit production behavior trend state data by analyzing the production behavior trend state of the forming unit production behavior characteristic data; Step S42: generating production behavior quality deviation data by analyzing the production behavior quality deviation according to the forming unit production behavior trend state data; Step S43: generating forming unit quality deviation attribution relationship data by analyzing the quality deviation attribution relationship of the forming unit according to the forming card deformation digital unit data and the production behavior quality deviation data; Step S44: generating forming unit parameter change-deviation response correlation data by analyzing the correlation between process parameter changes and deviation responses of the forming unit through the forming unit quality deviation attribution relationship data.

3. The method of claim 2, wherein the method is characterized by, Step S41 includes the following steps: Step S411: generating forming card local deformation self-correlation trend characteristic data by analyzing the forming card local deformation self-correlation trend characteristics according to the window forming card local geometric change characteristic data; Step S412: generating forming card local deformation influence characteristic data by analyzing the influence characteristics of the forming card local deformation trend according to the window control equipment running state characteristic data; Step S413: generating forming unit production behavior trend state data by analyzing the production behavior trend state of the forming unit production behavior characteristic data through the forming card local deformation self-correlation trend characteristic data and the forming card local deformation influence characteristic data.

4. The method for digital optimization of flat wire motor lamination forming parameters based on data analysis according to claim 2, characterized in that, Step S44 includes the following steps: Step S441: generating grouped forming card deformation digital unit data by grouping the process parameter value change conditions of the forming card deformation digital unit data; Step S442: generating identified grouped forming card deformation digital unit data by identifying the process parameter value change direction of the grouped forming card deformation digital unit data; Step S443: generating forming unit parameter change-deviation response correlation data by analyzing the correlation between process parameter changes and deviation responses of the forming unit according to the forming unit quality deviation attribution relationship data and the identified grouped forming card deformation digital unit data.

Images