Dynamic compensation control system and method for high-precision trimming die

Abstract

The invention discloses a dynamic compensation control system and method for a high-precision trimming die, and relates to the technical field of mechanical control, and the system comprises a trimming die monitoring module which is used for real-time monitoring and obtaining a trimming die monitoring data flow; the upper die compensation analysis module is used for performing health expectation deviation compensation analysis on an upper trimming die to generate an upper die compensation feature decision; the lower die compensation analysis module is used for carrying out compensation analysis on the trimming lower die and generating a lower die compensation characteristic decision; the guiding compensation analysis module is used for generating a guiding compensation characteristic decision; the fixed compensation analysis module is used for carrying out compensation control analysis on the trimming fixing assembly and generating a fixed compensation characteristic decision; and the trimming compensation execution module is used for performing compensation control on the trimming die. The technical problems that in the prior art, the production quality of a trimming die is unstable, and the requirement for product precision is difficult to meet are solved, and the technical effects that precise dynamic compensation is conducted on the trimming die, and therefore the working precision and stability of the trimming die are remarkably improved are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of mechanical control, and in particular to a dynamic compensation control system and method for a high-precision trimming die. Background Art

[0002] In modern manufacturing, trimming dies play a vital role in the production process of many products. In the fields of automobile manufacturing, electronic equipment production, aerospace parts processing, etc., the edge accuracy and quality requirements of parts are becoming increasingly stringent. Traditional trimming dies face many challenges during use. The trimming process is prone to dimensional deviation, irregular shape, poor surface quality and other problems.

[0003] The existing technology has the technical problem that the production quality of the trimming die is unstable and it is difficult to meet the product precision requirements. Summary of the invention

[0004] The present application provides a dynamic compensation control system and method for a high-precision trimming die, which is used to solve the technical problem in the prior art that the production quality of the trimming die is unstable and it is difficult to meet the product precision requirements.

[0005] In view of the above problems, the present application provides a dynamic compensation control system and method for a high-precision trimming die.

[0006] In a first aspect of the present application, a dynamic compensation control system for a high-precision trimming die is provided, the system comprising: A trimming die monitoring module, the trimming die monitoring module is used to monitor the trimming die in real time and obtain a trimming die monitoring data stream, wherein the trimming die includes an upper trimming die, a lower trimming die, a trimming guide component and a trimming fixing component; an upper die compensation analysis module, the upper die compensation analysis module performs a health expectation deviation compensation analysis on the trimming upper die based on an upper die health expectation prediction channel and in combination with the trimming die monitoring data stream, and generates an upper die compensation feature decision; a lower die compensation analysis module, the lower die compensation analysis module performs a compensation analysis on the trimming lower die based on a multi-feature lower die compensation factor and the trimming die monitoring data stream, and generates a lower die compensation feature decision; a guide compensation analysis module, the The guide compensation analysis module performs compensation analysis on the trimming guide component based on the guide compensation decision dual channels in combination with the trimming die monitoring data stream to generate a guide compensation feature decision; the fixed compensation analysis module performs compensation control analysis on the trimming fixed component based on the cutting workpiece fixing force perception data and the cutting workpiece fixing position perception data in combination with the trimming die monitoring data stream to generate a fixed compensation feature decision; the trimming compensation execution module is used to perform compensation control on the trimming die based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision.

[0007] A second aspect of the present application provides a dynamic compensation control method for a high-precision trimming die, the method comprising: Monitor the trimming die in real time to obtain a trimming die monitoring data stream, wherein the trimming die includes an upper trimming die, a lower trimming die, a trimming guide assembly and a trimming fixed assembly; based on the upper die health expectation prediction channel, combined with the trimming die monitoring data stream, perform a health expectation deviation compensation analysis on the upper trimming die to generate an upper die compensation feature decision; based on the multi-feature lower die compensation factor and the trimming die monitoring data stream, perform a compensation analysis on the trimming lower die to generate a lower die compensation feature decision; based on the guide compensation decision dual channel, combine the trimming die monitoring data stream to perform a compensation analysis on the trimming guide assembly to generate a guide compensation feature decision; based on the cutting workpiece fixing force perception data and the cutting workpiece fixing position perception data, combine the trimming die monitoring data stream to perform compensation control analysis on the trimming fixed assembly to generate a fixed compensation feature decision; based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision, perform compensation control on the trimming die.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: Monitor the trimming die in real time, obtain the trimming die monitoring data stream, generate the upper die compensation feature decision based on the upper die health expectation prediction channel; perform compensation analysis on the trimming lower die based on the multi-feature lower die compensation factor and the trimming die monitoring data stream, and generate the lower die compensation feature decision; generate the guide compensation feature decision based on the guide compensation decision dual channel; generate the fixed compensation feature decision based on the cutting workpiece fixed force perception data and the cutting workpiece fixed position perception data; perform compensation control on the trimming die based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision. The precise dynamic compensation of the trimming die is achieved, thereby significantly improving the working accuracy and stability of the trimming die. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of the structure of a dynamic compensation control system for a high-precision trimming die provided in an embodiment of the present application; Figure 2A schematic flow chart of a dynamic compensation control method for a high-precision trimming die provided in an embodiment of the present application.

[0011] Explanation of the reference numerals: trimming die monitoring module 10 , upper die compensation analysis module 20 , lower die compensation analysis module 30 , guide compensation analysis module 40 , fixed compensation analysis module 50 , trimming compensation execution module 60 . DETAILED DESCRIPTION

[0012] The present application provides a dynamic compensation control system and method for a high-precision trimming die, which is used to solve the technical problem in the prior art that the production quality of the trimming die is unstable and it is difficult to meet the product precision requirements.

[0013] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application. Embodiment 1

[0014] like Figure 1 As shown, the present application provides a dynamic compensation control system for a high-precision trimming die, the system comprising: The trimming die monitoring module 10 is used to monitor the trimming die in real time and obtain a trimming die monitoring data stream, wherein the trimming die includes an upper trimming die, a lower trimming die, a trimming guide assembly and a trimming fixing assembly.

[0015] Specifically, the trimming die monitoring module 10 integrates a series of advanced and highly sensitive monitoring instruments and technologies, including but not limited to high-precision displacement sensors, pressure sensors, temperature sensors, and image acquisition devices. These sensors and devices are arranged at key positions of the trimming die to ensure that the operating status of each part of the trimming die can be fully and accurately captured. For the trimming upper die, monitor its pressure distribution when in contact with the workpiece, the degree of surface wear, and the slight displacement changes during the working process. For the trimming lower die, pay close attention to its deformation when under pressure, temperature changes, and the accumulation of internal stress. In terms of the trimming guide component, accurately measure its guiding accuracy, smoothness of movement, and the degree of tightness with other parts of the mold. For the trimming fixing component, obtain in real time the size of its fixing force on the workpiece, the accuracy of the fixing position, and the stability in continuous work. Through these advanced monitoring methods, the trimming die monitoring module 10 can collect data at a very high frequency and quickly convert these data into trimming die monitoring data streams. These data streams contain rich and detailed information, such as the real-time working parameters of each component, the state change trend, and possible abnormal fluctuations. It provides a crucial, accurate and reliable data basis for subsequent analysis, compensation and control, and is one of the key links to ensure the high-precision and high-efficiency operation of the trimming die.

[0016] The upper die compensation analysis module 20 performs health expectation deviation compensation analysis on the trimming die based on the upper die health expectation prediction channel in combination with the trimming die monitoring data stream to generate an upper die compensation feature decision.

[0017] Specifically, the upper die compensation analysis module 20 is an important part for analyzing the expected deviation compensation of the trimming upper die. Through a large amount of experimental data, historical operation records and professional knowledge, a model is constructed that can simulate the expected health state of the trimming upper die under different working conditions. This model takes into account a variety of factors, such as the material properties of the trimming die, the working environment, the frequency of use, etc., to predict the ideal health state of the upper die as accurately as possible. The module receives information from the trimming die monitoring data stream in real time. These data streams contain many real-time parameters related to the trimming upper die, such as the temperature distribution of the upper die, the pressure changes, the vibration amplitude and frequency, etc. Then, the real-time parameters extracted from the monitoring data stream are input into the constructed upper die health expectation prediction model. The model will perform calculations and analyses based on these parameters and output the predicted expected health state of the trimming upper die under the current situation. In order to determine the deviation between the actual health state of the trimming upper die and the expected health state, the module will compare the predicted expected health state with the actual monitored upper die state in detail to accurately quantify the degree and specific direction of the deviation. Based on the analysis results of this healthy expectation deviation, the preset compensation strategy and algorithm in the module will be triggered to generate the upper die compensation feature decision. For example, if the analysis finds that the temperature of the upper die is too high and exceeds the healthy expectation range, the compensation feature decision is to instruct to adjust the working parameters of the upper die, such as reducing the stamping speed to reduce frictional heat; or prompt to check whether the cooling system is operating normally, whether maintenance or replacement of related parts is required, etc. For another example, if the vibration frequency of the upper die is monitored to be abnormal, the decision will require adjusting the fixing method of the upper die or checking the tightness of related connecting parts. In this way, the upper die compensation analysis module 20 can promptly detect the deviation of the healthy state of the trimming upper die and generate corresponding compensation decisions to ensure that the trimming upper die always maintains a high-precision and stable working state, prolongs its service life, and improves the quality and efficiency of trimming processing.

[0018] The lower die compensation analysis module 30 performs compensation analysis on the trimming lower die based on the multi-feature lower die compensation factor and the trimming die monitoring data stream to generate a lower die compensation feature decision.

[0019] Specifically, the die compensation analysis module works based on the multi-feature die compensation factors and trimming die monitoring data stream. The multi-feature die compensation factors specifically include die deformation, uneven die force, residual die material, die heat effect, and die displacement. The trimming die monitoring data stream contains various relevant data of the trimming die in the actual operation process. These monitoring data are obtained and then the multi-feature die compensation factors are comprehensively considered. For example, die deformation may lead to a decrease in processing accuracy, uneven die force may affect the life of the die and product quality, residual die material may interfere with subsequent processing, die heat effect may change the physical properties of the die, and die displacement may make the trimming position inaccurate. By establishing a specific model, the module can quantitatively analyze and process these factors, assign corresponding weights to each compensation factor, and combine the specific data in the monitoring data stream to calculate the specific characteristics and degree of compensation for the die. The generated die compensation feature decision includes specific measures such as adjustment, repair or replacement of the die. For example, if the analysis finds that the lower mold is deformed greatly, the decision is to reshape or replace the lower mold; if the force on the lower mold is uneven, the structure or processing parameters of the mold need to be adjusted to improve the force situation; for the residual material on the lower mold, it is necessary to add a cleaning process or improve the demoulding method; if it is a heat-affected problem, it is necessary to optimize the cooling system or adjust the processing technology to control the temperature; and for the displacement of the lower mold, it is necessary to calibrate the positioning device or strengthen the fixing method, etc. Through the work of the lower mold compensation analysis module 30, problems with the trimming lower mold can be discovered in time, and corresponding compensation decisions can be generated to improve the working performance of the trimming lower mold, ensure the processing quality and extend the service life of the mold.

[0020] The guide compensation analysis module 40 performs compensation analysis on the trimming guide component based on the guide compensation decision dual channel in combination with the trimming die monitoring data stream to generate a guide compensation feature decision.

[0021] Specifically, the dual channels of guide compensation decision-making include the upper and lower die guide cooperation compensation decision-making channel and the guide state abnormality compensation decision-making channel. The upper and lower die guide cooperation compensation decision-making channel focuses on the cooperation between the upper and lower dies through the guide components during the trimming process. The established model will consider factors such as the relative position, motion synchronization, and contact pressure distribution of the upper and lower dies. When the trimming die monitoring data stream shows that there is a deviation in the cooperation of the upper and lower dies, such as inaccurate relative position resulting in the trimming size not meeting the requirements, or poor motion synchronization affecting the processing efficiency and quality, the model of this channel will analyze the cause and degree of the deviation and calculate the compensation adjustments that need to be made, such as adjusting the installation position or angle of the guide component to improve the cooperation accuracy of the upper and lower dies. The guide state abnormality compensation decision-making channel focuses on monitoring whether the state of the guide component itself is abnormal. The model will consider the wear degree, deformation, temperature change, vibration and other parameters of the guide components. If the trimming die monitoring data stream indicates that the guide components are severely worn, excessively deformed, overheated or have abnormal vibrations, the model of this channel will evaluate the impact of the abnormal state on the trimming guide according to the set standards and strategies, and determine the corresponding compensation measures, which may include replacing worn parts, strengthening cooling or adding shock absorbers. Combining the analysis results of the two channels with the real-time information provided by the trimming die monitoring data stream, a comprehensive model is established to ultimately generate a guide compensation feature decision that can effectively improve the performance of the trimming guide components and improve the trimming processing accuracy and stability.

[0022] The fixed compensation analysis module 50 performs compensation control analysis on the trimming fixing component based on the cutting workpiece fixing force perception data and the cutting workpiece fixing position perception data in combination with the trimming die monitoring data stream to generate a fixed compensation feature decision.

[0023] Specifically, the fixed force sensing data and fixed position sensing data of the cutting workpiece are obtained. The fixed force sensing data can reflect the size and stability of the clamping force applied by the fixed component to the workpiece, while the fixed position sensing data clarifies whether the specific position of the workpiece in the fixed component is accurate and stable. At the same time, the module receives the trimming die monitoring data stream, which contains rich information such as the working status of the trimming fixed component and the operating parameters of related components. The module will perform comprehensive processing and analysis on these data. By comparing the actual sensing data of the fixing force with the preset ideal fixing force range, it is judged whether the current fixing force is sufficient and stable. If the fixing force is insufficient, the workpiece will be displaced during the trimming process, affecting the processing accuracy; if the fixing force is too large, it will cause damage to the workpiece. For the fixed position sensing data, it will be compared with the standard fixed position requirements to check whether the workpiece is in the correct position. Position deviation will lead to inaccurate trimming position, thus affecting product quality. Other relevant information in the trimming die monitoring data stream, such as the wear of the fixing component and temperature changes, is combined to further evaluate the overall working condition of the fixing component. Using the optimal control model, with the goal of minimizing the fixed error, improving system stability and reducing energy consumption, the optimal compensation control strategy is solved, the required compensation adjustment amount and adjustment method are calculated, and finally, the fixed compensation feature decision is generated. These decisions include adjusting the size of the fixing force, recalibrating the fixing position, replacing worn fixing parts, optimizing the structure or working parameters of the fixing component, etc., to ensure that the trimming fixing component can firmly and accurately fix the workpiece, provide a stable and reliable foundation for trimming processing, and ensure processing quality and production efficiency.

[0024] The trimming compensation execution module 60 is used to perform compensation control on the trimming die based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision.

[0025] Specifically, the trimming compensation execution module 60, as the key execution link of the entire trimming die compensation control system, undertakes the important task of converting various compensation feature decisions into actual compensation control actions. When receiving the upper die compensation feature decision, the trimming compensation execution module 60 will accurately control the relevant actuators according to the specific decision content, such as the adjustment requirements for the upper die working parameters, to achieve the adjustment of the pressure, movement speed, position and other aspects of the upper die. For the lower die compensation feature decision, the module will perform corresponding compensation operations on the temperature, force distribution, displacement and other aspects of the lower die according to the decision instructions, including adjusting the cooling system, changing the support structure or optimizing the processing parameters. When processing the guide compensation feature decision, the trimming compensation execution module 60 will control the adjustment device of the guide component to accurately control the position, angle, lubrication state and other aspects of the guide component to ensure the accuracy and stability of the guide. For the fixed compensation feature decision, the module will operate the control component of the fixed component to effectively execute the size of the fixed force, the calibration of the fixed position and the replacement or maintenance of the fixed component. In order to achieve these compensation control actions, the trimming compensation execution module 60 will work closely with a series of actuators, sensors and controllers. It will monitor the execution effect of the compensation action in real time and compare the feedback information with the expected compensation target. If necessary, it will make further fine-tuning or corrections to ensure that the trimming die can achieve the expected high-precision, high-stability and high-quality processing requirements after compensation control.

[0026] In a possible implementation, the upper die compensation analysis module 20 further includes: According to the trimming die monitoring data stream, the associated data of the trimming die upper die are integrated to construct an upper die monitoring state matrix; the cutting workpiece feature data and the workpiece cutting upper die control decision corresponding to the trimming die are retrieved; based on the upper die basic data set of the trimming die upper die, in combination with the cutting workpiece feature data and the workpiece cutting upper die control decision, the health expectation state of the trimming die upper die is predicted according to the upper die health expectation prediction channel to obtain an upper die health expectation matrix; an upper die health expectation deviation compensation model is constructed, and the upper die health expectation deviation compensation model is embedded into the upper die compensation analysis module; based on the upper die monitoring state matrix and the upper die health expectation matrix, the upper die compensation feature decision is obtained according to the upper die health expectation deviation compensation model.

[0027] Specifically, the system first obtains the trimming die monitoring data stream. This data stream contains a large amount of raw data related to the trimming die working process. Then, through data screening and extraction methods, data directly related to the trimming die are selected from this huge amount of data, including temperature changes, pressure size and distribution, displacement, vibration frequency and amplitude and other parameters of the trimming die. These selected related data are sorted and classified, and data of similar nature or association are grouped together. Finally, the grouped and sorted data are arranged and combined in the form of a matrix to construct a die monitoring state matrix that can comprehensively and accurately reflect the real-time state of the trimming die. Each row and column of this matrix represents a specific type of data, for example, the first row represents temperature data, the second row represents pressure data, etc., and each column corresponds to the data value of different time points or sampling points. By constructing such a die monitoring state matrix, the complex and scattered trimming die related data can be integrated into a structured, easy to analyze and process form, providing a clear and accurate data basis for subsequent health expectation state prediction, compensation analysis and other operations.

[0028] The system will obtain the feature data of the workpiece cut corresponding to the current trimming die from the stored database or related data set. The feature data of the workpiece cut includes information about the material, such as the type of material (such as steel, aluminum, plastic, etc.), physical performance parameters such as hardness, strength, toughness, and surface roughness, thermal conductivity, and other properties of the material. The structural feature data may cover the shape of the workpiece (such as flat plate, curved plate, complex three-dimensional structure, etc.), size (length, width, thickness, etc.), and complexity of the structure (whether there are holes, protrusions, grooves, etc.). At the same time, the system will also retrieve the control decision of the upper die for workpiece cutting, which is based on process requirements, production standards, and previous experience. The operating instructions for the upper die in the process of cutting the workpiece are formulated. For example, the control parameters such as the movement speed, stroke, applied pressure, and dwell time of the upper die. By retrieving these cutting workpiece feature data and the control decision of the upper die for workpiece cutting, the system can provide a more comprehensive and accurate reference basis for the subsequent upper die state prediction and compensation analysis to ensure the accuracy and quality of trimming processing and improve production efficiency and product qualification rate.

[0029] First, the upper die basic data set of the trimming upper die contains basic information such as the initial design parameters, material properties, and manufacturing process of the upper die. Then, the cutting workpiece feature data (including the material type, hardness, shape structure, etc. of the workpiece) and the workpiece cutting upper die control decision (such as the specific control parameters of the upper die such as cutting speed, pressure, and stroke) are combined with the upper die basic data set. The model established by the upper die health expectation prediction channel is a complex system that integrates mechanical principles, material science knowledge, and data analysis technology. The mechanical model built based on the finite element analysis method is used to simulate the stress distribution and deformation when the upper die interacts with the workpiece; the deep learning neural network in machine learning is used to capture the potential relationship between various input factors and the upper die health status through learning a large amount of historical data. When the model is operated, the above integrated data set is used as input. The model predicts the friction and impact force that the upper die will bear during the cutting process based on the characteristics of the workpiece material, and calculates the location and degree of stress concentration based on the structural characteristics. According to the speed and pressure parameters in the control decision, the dynamic response and energy consumption of the upper die are simulated. After comprehensive calculation and analysis of the model, the health expectation state of the upper die under specific working conditions is obtained. These conditions include the wear prediction values ​​of various parts of the upper die, the amplitude and direction of deformation, the distribution of internal stress, and the location and development trend of possible fatigue cracks. Organizing and presenting these rich prediction results in the form of a matrix forms the upper die health expectation matrix. The rows of the matrix can represent different health status indicators, such as wear, deformation, stress, fatigue, etc.; the columns can correspond to different areas or time nodes of the upper die. Each matrix element specifically quantifies the expected health status of the corresponding area at a specific time. Through such a detailed and comprehensive process, it is possible to provide accurate and forward-looking predictions for evaluating the future health status of the upper die, and provide solid data support for subsequent maintenance and optimization decisions.

[0030] In order to construct the upper die health expectation deviation compensation model, it is necessary to make a detailed comparison and in-depth analysis of the obtained upper die health expectation matrix and the actual monitored upper die real-time status data. The differences between the two are measured from multiple dimensions, comparing the predicted wear degree with the actual measured wear amount, comparing the expected deformation amplitude with the actual deformation, and evaluating the deviation between the predicted stress distribution and the actual monitored stress distribution. The mean, standard deviation and other indicators in statistics are used to describe the central tendency and dispersion of the deviation to find out the key factors causing the deviation. Based on the quantified deviation degree, a specific compensation strategy is formulated, which involves adjusting the mechanical structure of the upper die, such as replacing severely worn parts and repairing deformed parts; it also includes re-optimization of cutting process parameters, such as appropriately increasing or decreasing the cutting speed, increasing or decreasing the applied pressure, etc. When formulating the compensation strategy, various practical constraints are also taken into account, such as cost constraints, production schedule requirements, and equipment operability, to ensure the feasibility and effectiveness of the compensation strategy. The constructed upper die health expectation deviation compensation model will be implemented in the form of algorithms and codes through programming, and embedded in the upper die compensation analysis module. During the embedding process, it is necessary to ensure that the model is compatible with the interface of the upper die compensation analysis module, that it can accurately receive the relevant data input of the upper die, and that the calculated compensation decision can be output to the subsequent processing flow of the module. A series of tests and verifications will also be carried out to ensure that the embedded model can run stably and accurately, and work in conjunction with other parts of the upper die compensation analysis module to achieve effective compensation and optimization of the upper die health status. Through such a specific and rigorous construction and embedding process, the upper die health expectation deviation compensation model can become a core component of the upper die compensation analysis module, playing a key role in ensuring the normal operation of the trimming die and improving production quality.

[0031] Based on the upper die monitoring state matrix and upper die health expectation matrix obtained above, these two matrices are input into the upper die health expectation deviation compensation model. The model accurately quantifies the degree of deviation between the actual state of the upper die and the expected state. Based on the quantitative results of these deviations, these deviations are converted into specific compensation requirements, which involve multiple aspects, such as fine-tuning the structure of the upper die, changing the conditions of the working environment (such as temperature, humidity, etc.), and adjusting the matching accuracy between the upper die and other components. Then, these compensation requirements are screened and sorted according to the pre-set priority and feasibility criteria. For example, if the wear of the upper die has seriously affected the product accuracy, the decision to replace the worn parts will be given priority; if it is only a slight temperature deviation, it will be solved by optimizing the cooling system first. Finally, all the analysis and screening results are combined to generate the upper die compensation feature decision, which will clearly point out the specific compensation actions to be taken, including specific operation steps, required resources and time, and expected results. To ensure that the upper die can be restored to an ideal healthy state as soon as possible, and to ensure the normal operation of the trimming die and the quality of the product.

[0032] In a possible implementation, the upper die compensation analysis module 20 further includes: Deviation identification is performed based on the upper die monitoring state matrix and the upper die health expectation matrix to obtain the upper die monitoring expected deviation matrix; cutting impact evaluation is performed based on the upper die monitoring expected deviation matrix to generate the upper die deviation cutting impact degree; it is determined whether the upper die deviation cutting impact degree is greater than / equal to the predetermined upper die deviation cutting impact degree; if the upper die deviation cutting impact degree is greater than / equal to the predetermined upper die deviation cutting impact degree, the upper die health expectation deviation compensation model is activated; the upper die monitoring expected deviation matrix is ​​input into the upper die health expectation deviation compensation model to generate the upper die compensation feature decision.

[0033] Specifically, the acquired upper mold monitoring state matrix and upper mold health expectation matrix are compared and calculated element by element to identify the deviation between them. Through corresponding mathematical operations, difference calculations, and ratio calculations, a new matrix, the upper mold monitoring expectation deviation matrix, is obtained. Each element in this matrix quantitatively represents the degree of deviation between the actual monitoring state of the upper mold and the health expectation state in the corresponding position or parameter.

[0034] When evaluating the cutting impact, each element in the upper die monitoring expected deviation matrix will be carefully analyzed first. The deviations represented by these elements may include the position deviation, temperature deviation, pressure deviation, deformation deviation, etc. of the upper die. Different weights will be assigned to each deviation element according to its corresponding physical parameters and its importance in the cutting process. For example, the deviation of the upper die position may have a greater impact on the cutting accuracy, so it is given a higher weight; while a small deviation in temperature may have a relatively low weight. Taking all the deviation elements and their weights into consideration, the weighted summation method is used to calculate the upper die deviation cutting impact. This indicator will reflect the overall impact of the upper die deviation on the cutting process, and finally generate a quantitative upper die deviation cutting impact, which can intuitively reflect the impact of the upper die deviation on the cutting results in terms of accuracy, quality, efficiency, etc.

[0035] The specific value of the upper die deviation cutting influence is determined. This predetermined value is usually a threshold value set based on past experience, industry standards, product quality requirements, equipment performance and other factors. Then, the calculated current upper die deviation cutting influence is compared with this predetermined value. If the upper die deviation cutting influence is greater than or equal to the predetermined upper die deviation cutting influence, it means that the upper die deviation has reached a relatively serious level, which will have a significant adverse effect on the quality, efficiency, cost and other aspects of cutting.

[0036] The system immediately triggers the instruction to activate the upper die healthy expected deviation compensation model. After the model is awakened, it starts initialization and data preparation. Then, the upper die monitoring expected deviation matrix is ​​accurately input into the activated compensation model. After receiving the deviation matrix, the model will quickly start its internal complex calculation and analysis mechanism, and finally generate an accurate and feasible upper die compensation feature decision. This decision will specify key information such as the compensation method, amplitude, time and sequence in detail, providing clear guidance for subsequent upper die adjustment and repair work, so as to restore the normal working state of the upper die as soon as possible and ensure cutting quality and production efficiency.

[0037] In a possible implementation, the lower die compensation analysis module 30 further includes: The multi-feature lower die compensation factors include lower die deformation, uneven force on the lower die, residual lower die material, lower die thermal impact and lower die displacement; the associated data of the trimming die is integrated according to the trimming die monitoring data stream to generate a lower die monitoring block; based on the multi-feature lower die compensation factors, a lower die compensation decision is made according to the lower die monitoring block to generate a multi-feature lower die compensation decision; the multi-feature lower die compensation decisions are integrated to obtain the lower die compensation feature decision.

[0038] Specifically, the multi-feature die compensation factor is a key factor used to evaluate and adjust the state of the trimming die. Die deformation refers to the change in shape of the die due to factors such as pressure and temperature during use, which may lead to a decrease in cutting accuracy. Uneven force on the die means that the forces borne by various parts of the die are inconsistent, which will affect the service life and cutting effect of the die, and easily cause local wear or breakage. Die material residue means that after the cutting is completed, some raw materials are still attached to the die, which will interfere with subsequent cutting operations and affect product quality. The heat impact of the die refers to the heat generated during the cutting process that causes the temperature of the die to change, resulting in changes in its physical properties, such as hardness, strength, etc., which in turn affects the working performance of the die. Die displacement means that the die has moved or offset during work, which will destroy the originally set cutting position and accuracy, causing the product to not meet the requirements.

[0039] According to the trimming die monitoring data stream, the data directly related to the trimming die is screened out, and the data is distinguished and extracted according to specific identifiers, tags or data fields. The extracted trimming die-related data is classified and sorted, and grouped according to the data type, such as deformation-related data, force data, heat-affected data, etc. After that, the classified and sorted groups of data are integrated and associated. For example, the deformation data and force data of the die at the same time point or in the same working cycle are correlated to form a complete data set that can reflect the state of the die under specific working conditions. Finally, this integrated and associated data set is stored and managed as an independent die monitoring block. This block can provide centralized, orderly and targeted data support for subsequent analysis and decision-making, so as to more accurately evaluate the working status of the die and formulate corresponding compensation strategies.

[0040] Analyze the multi-feature die compensation factors, that is, conduct in-depth research on various factors such as die deformation, uneven die force, die material residue, die heat effect and die displacement. Clarify the normal range, abnormal threshold and specific impact of each factor on the die working performance and cutting quality. Match the data in the die monitoring block with these compensation factors one by one. For die deformation, compare the monitored deformation data with the normal deformation range; for uneven force, analyze the difference between the force distribution and the ideal uniform force; for material residue, determine the amount and position of the residue; for thermal effect, evaluate the effect of temperature change on the die; for die displacement, measure the value and direction of displacement. Based on these comparison and analysis results, make preliminary compensation decisions for each compensation factor. Combine the preliminary decisions of each compensation factor to form a multi-feature die compensation decision.

[0041] Sort out and analyze each multi-feature die compensation decision one by one. Understand the specific problems, proposed compensation measures, and expected effects of each decision. Then, evaluate the relationship between these compensation decisions. Some decisions promote each other and can work together to more effectively solve the problems of the die; while some decisions may have certain conflicts or constraints and need to be coordinated and balanced. For decisions that promote each other, integrate them together to form a unified compensation plan. For example, if one decision is to increase cooling measures to reduce the impact of heat, and another decision is to adjust processing parameters to reduce uneven force, these two decisions can be implemented together to achieve better compensation effects. For conflicting decisions, it is necessary to weigh them according to the actual situation. For example, one decision requires a lot of time and resources to fully repair the mold, while another decision recommends taking temporary adjustment measures to resume production as soon as possible. In this case, it may be necessary to comprehensively consider factors such as production schedule, cost, and repair effect to select the best solution. In the fusion process, the overall performance and working environment of the die also need to be considered. Ensure that the compensation feature decision can not only solve the specific problems currently found, but also will not have an adverse impact on other aspects of the die, and can adapt to the actual operating conditions on the production line. Through comprehensive analysis, coordination and optimization, the various multi-feature die compensation decisions are integrated into a complete, coordinated and feasible die compensation feature decision. This provides clear and comprehensive guidance for subsequent die compensation operations to maximize die performance and production efficiency.

[0042] In a possible implementation, the lower die compensation analysis module 30 further includes: According to the multi-feature lower die compensation factor, feature recognition is performed on the lower die monitoring block to obtain lower die deformation monitoring data; according to the multi-feature lower die compensation factor, a lower die deformation cutting impact evaluation model and a lower die deformation compensation decision model are constructed; based on the lower die deformation monitoring data, according to the lower die deformation cutting impact evaluation model, a lower die deformation cutting impact evaluation coefficient is obtained; it is determined whether the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold; if the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold, the lower die deformation monitoring data is input into the lower die deformation compensation decision model to obtain a lower die deformation compensation decision; the lower die deformation compensation decision is added to the multi-feature lower die compensation decision, and the lower die compensation decision is continued to be performed according to the multi-feature lower die compensation factor and the lower die monitoring block to obtain the multi-feature lower die compensation decision.

[0043] Specifically, the specific factors included in the multi-feature die compensation factor are clarified, such as die deformation, uneven force, material residue, thermal impact and displacement. The die monitoring block is monitored comprehensively and carefully using relevant monitoring technologies and equipment. The monitoring methods and means may involve a variety of sensors, measuring instruments or technical methods to accurately obtain data related to the multi-feature die compensation factor. During the monitoring process, special attention will be paid to those features and indicators that can reflect the deformation of the die. For example, by measuring the displacement changes of specific parts of the die, the degree of deformation of the die can be understood; the stress distribution of the die can be analyzed to determine whether the force is uniform; the residual material can be detected, and the temperature changes of the die during operation and its impact on deformation can be monitored. The various data obtained from the monitoring are integrated and analyzed to extract the key information that can accurately describe the deformation of the die, that is, the die deformation monitoring data.

[0044] The construction of the evaluation model for the impact of die deformation cutting determines the key factors affecting die deformation cutting. In addition to the multi-feature die compensation factors mentioned above, such as die deformation, uneven force, material residue, thermal impact and displacement, it is also necessary to further analyze which specific factors have the most significant impact on die deformation cutting. Through monitoring and measurement in actual production, a large amount of die deformation cutting data is obtained, including die deformation, force conditions, temperature changes, material properties and product quality after cutting under different working conditions. The collected data is analyzed to extract features related to the impact of die deformation cutting. Select a suitable model structure, use the decision tree algorithm to analyze the multi-feature die compensation factor, use it as the input feature of the decision tree, and automatically determine the conditions for dividing nodes according to the characteristics and distribution of the data. For example, if the die deformation is an important feature, the decision tree may first divide the nodes according to the size of the deformation and divide the data into different subsets. Automatically determine the conditions for dividing nodes according to the characteristics and distribution of the data. For example, if the die deformation is an important feature, the decision tree may first divide the nodes according to the size of the deformation and divide the data into different subsets. In this way, a decision tree is constructed that can predict the degree of influence of die deformation cutting based on multi-feature die compensation factors. The selected model is trained using the collected data. By adjusting the model parameters, the model can predict the degree of influence of die deformation cutting as accurately as possible based on the input features. The performance of the model is evaluated using cross-validation and other techniques, and necessary optimization is performed. The trained model is verified using an independent test data set to evaluate the accuracy, reliability and generalization of the model. The construction of the die deformation compensation decision model determines the specific goals of die deformation compensation, such as controlling the deformation within a certain range, improving the dimensional accuracy of the product, etc., and considering various constraints in actual production, such as cost constraints and equipment capabilities.

[0045] The decision tree algorithm is also used to build a lower die deformation compensation decision model, and a large amount of sample data including lower die deformation features (such as deformation amount, deformation position, deformation speed, etc.), multi-feature lower die compensation factors (such as force conditions, material properties, temperature, etc.) and corresponding compensation measures (such as adjusting mold structure, changing processing parameters, etc.) are collected. The data is preprocessed, including data cleaning, missing value processing, and feature engineering (such as feature selection, feature scaling, etc.) so that the algorithm can better process the data. The key features related to the lower die deformation compensation decision are selected from the collected data, and these features will be used as input nodes of the decision tree. Starting from the root node, a feature that can best split the data set into different subsets is selected as the split feature. According to the different values ​​of the selected feature, the data set is split into multiple child nodes. On each child node, the above steps are repeated to select a new split feature until the stopping condition is met, such as reaching the predetermined tree depth, the number of samples in the child node is too small, or the purity of the node is high enough. Use a part of the data (test set) that does not participate in the construction of the decision tree to evaluate the model. Common evaluation indicators include accuracy, recall rate, F1 value, etc. Through repeated experiments and optimization, combined with domain knowledge and experience, the lower die deformation cutting impact evaluation model and the lower die deformation compensation decision model are continuously improved so that they can accurately predict the deformation cutting impact based on the multi-feature lower die compensation factor and provide effective compensation decisions.

[0046] The monitoring data of the lower die deformation is obtained through relevant monitoring equipment and technology. These data include deformation amount, deformation position, deformation speed and other parameters. The collected original monitoring data is cleaned, screened and sorted, and outliers and erroneous data are removed to ensure the accuracy and reliability of the data. The pre-processed lower die deformation monitoring data is input into the lower die deformation cutting impact evaluation model to obtain the corresponding results.

[0047] The calculated lower die deformation cutting impact evaluation coefficient is compared with the pre-set lower die deformation cutting impact evaluation threshold. If the evaluation coefficient is greater than or equal to the threshold, it means that the lower die deformation has a significant impact on cutting and compensation measures need to be taken. The acquired lower die deformation monitoring data is input into the lower die deformation compensation decision model, and the input data is analyzed and calculated to finally give a compensation decision for the lower die deformation, and this lower die deformation compensation decision is added to the previous multi-feature lower die compensation decision. Based on the multi-feature lower die compensation factor and the lower die monitoring block, the lower die compensation decision is analyzed and judged again, and the final multi-feature lower die compensation decision is determined by comprehensively considering the compensation decision obtained before and the newly acquired monitoring data and compensation factors. The whole process is a dynamic and step-by-step optimization process, which aims to achieve accurate compensation for the lower die through continuous monitoring, analysis and decision-making, and improve the performance and accuracy of the lower die in the cutting process.

[0048] In a possible implementation, the guidance compensation analysis module 40 further includes: According to the trimming die monitoring data stream, the associated data of the trimming guide component is integrated to generate a guide component angle feature point cloud, a guide component position feature point cloud and a guide component state feature information; according to the trimming die monitoring data stream, an upper die position feature point cloud and a lower die position feature point cloud are obtained; according to the upper die position feature point cloud and the lower die position feature point cloud, a matching offset detection is performed to generate an upper and lower die matching offset point cloud; based on the guide component angle feature point cloud, the guide component position feature point cloud and the upper and lower die matching offset point cloud, according to the upper and lower die guide matching compensation decision channel of the guide compensation dual channel, an upper and lower die guide matching compensation decision is obtained; based on the guide component state feature information, according to the guide state abnormality compensation decision channel of the guide compensation dual channel, a guide state abnormality compensation decision is obtained; the upper and lower die guide matching compensation decision and the guide state abnormality compensation decision are integrated to obtain the guide compensation feature decision.

[0049] Specifically, the trimming die monitoring data stream contains a lot of complex and large amounts of data about the trimming guide assembly. These data come from various sensors installed on the trimming guide assembly, such as angle sensors, position sensors, and sensors for monitoring wear, vibration, impact, and environmental conditions. The system will preliminarily screen and classify these data streams and extract the data related to the angle of the trimming guide assembly. Through further processing and analysis of these angle data, the angle measurement values ​​at different time points and under different working conditions are converted into a series of coordinate points using principal component analysis. These coordinate points together constitute the guide assembly angle feature point cloud, which can intuitively reflect the change law and distribution of the guide assembly angle. Similarly, data related to the guide assembly position are selected from the data stream. After processing and calculation, the position measurement value is converted into a position feature point cloud, which accurately shows the position movement and change trajectory of the guide assembly in space. For the state feature information of the guide assembly, the system will process the data from the wear sensor, vibration sensor, impact sensor, and environmental monitoring sensor respectively. For wear monitoring, the data such as material loss and surface roughness change obtained by the sensor are analyzed to evaluate the degree of wear. The data from the vibration sensor is used to determine the frequency, amplitude and pattern of vibration to understand the vibration characteristics of the guide assembly. The impact monitoring data helps determine the intensity and time of impact. Environmental monitoring data includes information such as temperature and humidity, which is used to evaluate the impact of the environment on the guide assembly. Combining these data, comprehensive status characteristic information about the trimming guide assembly is formed, including wear, vibration level, whether it is subjected to impact and the environmental conditions. Through such detailed and comprehensive integration and analysis of the trimming die monitoring data stream, the guide assembly angle feature point cloud, position feature point cloud and status feature information are generated.

[0050] The monitoring data stream is preprocessed, including removing noise, erroneous data, and outliers. Then, the position data of the upper and lower molds at different times are extracted in chronological order and at a specific sampling frequency. For the upper mold, the position coordinates (x, y, and z coordinates in three-dimensional space) obtained at each sampling time are combined to form a series of points. These points constitute the upper mold position feature point cloud in three-dimensional space, reflecting the position change trajectory and distribution of the upper mold during the working process. Similarly, for the lower mold, the position data is processed in the same way to obtain the lower mold position feature point cloud.

[0051] First, the angle feature point cloud of the guide component is deeply analyzed. Through mathematical statistical methods, such as calculating the mean, variance and distribution of the angle, the change law and stability of the guide component angle are understood. At the same time, pattern recognition technology is used to identify possible abnormal angle patterns. For the position feature point cloud of the guide component, spatial geometric analysis and topological methods are used to determine the position range, movement trajectory and deviation from the ideal position of the guide component. The position characteristics can also be intuitively displayed by building a three-dimensional model of the position. For the upper and lower mold matching offset point cloud, deviation analysis and error evaluation algorithms are used to accurately calculate the size, direction and distribution of the offset. In addition, historical data or standard data are compared to determine whether the offset is within an acceptable range. The information of angle, position and offset obtained from the above analysis is input into the upper and lower mold guide matching compensation decision channel of the guide compensation dual channel. The upper and lower mold guide matching compensation decision channel is constructed based on a neural network and is learned and optimized through a large amount of training data. These training data contain the angle characteristics, position characteristics, upper and lower mold matching offset characteristics of the guide component under various different conditions and the corresponding optimal compensation decisions. By learning from these data, the neural network can automatically extract the complex patterns and relationships hidden in the data. When new angle feature point clouds, position feature point clouds, and upper and lower mold matching offset point cloud data are input into the channel, the neural network will quickly perform calculations and reasoning, evaluate and judge the current situation based on the knowledge and patterns it has learned, and output corresponding compensation decision recommendations.

[0052] Similarly, based on the state characteristic information of the guidance component, a compensation decision is obtained by using the guidance state abnormality compensation decision channel constructed according to the neural network model. By comprehensively analyzing the various state characteristic information of the guidance component and based on the preset algorithm in the channel, a specific compensation decision for the guidance state abnormality is finally generated.

[0053] Evaluate the upper and lower die guide coordination compensation decisions and guide state abnormality compensation decisions to find out the correlation and conflict. Then weigh and coordinate according to production needs, equipment performance and cost, and determine the priority if there is a conflict. Then integrate and optimize the compensation measures to form a comprehensive and unified guide compensation feature decision, and finally review and verify the decision to ensure that it is feasible, reasonable and effective.

[0054] In a possible implementation, the fixed compensation analysis module 50 further includes: Obtain an expected fixing force for cutting the workpiece and an expected fixing position for cutting the workpiece; perform deviation detection on the fixing force sensing data for cutting the workpiece according to the expected fixing force for cutting the workpiece, and obtain a workpiece fixing force deviation detection result; perform deviation detection on the fixing position sensing data for cutting the workpiece according to the expected fixing position for cutting the workpiece, and obtain a workpiece fixing position deviation detection result; perform associated data integration of the trimming fixing component according to the trimming die monitoring data stream, and obtain fixing component monitoring data; activate a pre-built fixing compensation correction model; input the workpiece fixing force deviation detection result, the workpiece fixing position deviation detection result, and the fixing component monitoring data into the fixing compensation correction model, and obtain the fixing compensation feature decision.

[0055] Specifically, the expected fixing force and fixing position of the workpiece are obtained. This step takes into account the material, shape, size, cutting process requirements, previous practical experience, theoretical calculations and other factors of the workpiece. Through in-depth analysis of these factors, the expected values ​​of the fixing force and fixing position that can ensure cutting quality and accuracy under ideal conditions are obtained.

[0056] According to the expected fixing force of the cutting workpiece, the deviation detection of the fixing force sensing data of the cutting workpiece is carried out, and the fixing force sensing data collected in real time is accurately compared with the preset fixing force expectation. Through comparison, it can be clearly determined whether the sensing data is higher or lower than the expected fixing force, as well as the specific value and degree of deviation, so as to obtain the deviation detection result of the workpiece fixing force.

[0057] The fixed position sensing data of the cutting workpiece is detected for deviation according to the fixed position expectation of the cutting workpiece. The fixed position sensing data and the fixed position expectation are compared in the same rigorous way to determine whether there is a position deviation and the specific situation of the deviation, and obtain the workpiece fixed position deviation detection result.

[0058] The associated data integration of the trimming fixture component is performed according to the trimming die monitoring data stream. The key information related to the trimming fixture component, such as the pressure, temperature, deformation, etc. of the component, is extracted from a large amount of monitoring data streams, and these data are screened, sorted and comprehensively analyzed to obtain the fixture component monitoring data that fully reflects the working status of the trimming fixture component. In the whole process, after the collection, processing and deviation detection of relevant data are completed, and the workpiece fixing force deviation detection results, workpiece fixing position deviation detection results and fixing component monitoring data are obtained, it is time to activate the pre-built fixing compensation correction model. This fixing compensation correction model is based on the neural network architecture in previous work, and consists of many interconnected neurons that form a complex hierarchical structure. When constructing this neural network model, a large amount of historical data related to the fixing of cutting workpieces was collected, including the actual and expected values ​​of the fixing force and fixing position of different types of workpieces, as well as the corresponding compensation correction measures and the final effect data. The neural network is then trained using this data, during which it continuously adjusts the connection weights between neurons to learn patterns and regularities in the data, so that it can accurately predict effective compensation correction strategies based on newly input fixed force deviation detection results, fixed position deviation detection results, and fixed component monitoring data. Once the model is successfully activated, it is ready to receive subsequent inputs of various detection results and monitoring data for further processing and analysis to generate the final fixed compensation feature decision.

[0059] The workpiece fixing force deviation detection results, workpiece fixing position deviation detection results and fixing component monitoring data obtained through preliminary processing and detection are prepared in a specific data format and order. Then, these sorted data are input into the pre-built and trained fixing compensation correction model (neural network model), and finally the fixing compensation feature decision is output. This decision comprehensively considers various information such as the workpiece fixing force, the deviation of the fixing position and the status of the fixing component. The output fixing compensation feature decision includes specific adjustment parameters, such as detailed operation instructions such as adjusting the size of the fixing force, correcting the offset of the fixing position, replacing or repairing the fixing component. These decisions are aimed at correcting the current fixing deviation, improving the fixing effect of the cut workpiece, and ensuring the accuracy and quality of the cutting process. Embodiment 2

[0060] Based on the same inventive concept as the dynamic compensation control system of a high-precision trimming die in the above-mentioned embodiment, Figure 2 As shown, the present application provides a dynamic compensation control method for a high-precision trimming die, and the method in the embodiment of the present application and the system embodiment are based on the same inventive concept. The method includes: Step S100: monitor the trimming die in real time to obtain a trimming die monitoring data stream, wherein the trimming die includes an upper trimming die, a lower trimming die, a trimming guide assembly and a trimming fixing assembly.

[0061] Step S200: Based on the upper die health expectation prediction channel, the trimming die monitoring data stream is combined to perform health expectation deviation compensation analysis on the trimming die, and generate an upper die compensation feature decision.

[0062] Step S300: performing compensation analysis on the trimming die based on the multi-feature die compensation factor and the trimming die monitoring data stream to generate a die compensation feature decision.

[0063] Step S400: Based on the dual channels of guide compensation decision, the trimming guide component is subjected to compensation analysis in combination with the trimming die monitoring data stream to generate a guide compensation feature decision.

[0064] Step S500: Based on the cutting workpiece fixing force sensing data and the cutting workpiece fixing position sensing data, combined with the trimming die monitoring data stream, compensation control analysis is performed on the trimming fixing component to generate a fixing compensation feature decision.

[0065] Step S600: performing compensation control on the trimming die based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision.

[0066] Furthermore, step S200 further includes: According to the trimming die monitoring data stream, the associated data of the trimming die upper die are integrated to construct an upper die monitoring state matrix; the cutting workpiece feature data and the workpiece cutting upper die control decision corresponding to the trimming die are retrieved; based on the upper die basic data set of the trimming die upper die, in combination with the cutting workpiece feature data and the workpiece cutting upper die control decision, the health expectation state of the trimming die upper die is predicted according to the upper die health expectation prediction channel to obtain an upper die health expectation matrix; an upper die health expectation deviation compensation model is constructed, and the upper die health expectation deviation compensation model is embedded into the upper die compensation analysis module; based on the upper die monitoring state matrix and the upper die health expectation matrix, the upper die compensation feature decision is obtained according to the upper die health expectation deviation compensation model.

[0067] Furthermore, step S200 further includes: Deviation identification is performed based on the upper die monitoring state matrix and the upper die health expectation matrix to obtain the upper die monitoring expected deviation matrix; cutting impact evaluation is performed based on the upper die monitoring expected deviation matrix to generate the upper die deviation cutting impact degree; it is determined whether the upper die deviation cutting impact degree is greater than / equal to the predetermined upper die deviation cutting impact degree; if the upper die deviation cutting impact degree is greater than / equal to the predetermined upper die deviation cutting impact degree, the upper die health expectation deviation compensation model is activated; the upper die monitoring expected deviation matrix is ​​input into the upper die health expectation deviation compensation model to generate the upper die compensation feature decision.

[0068] Furthermore, step S300 also includes: The multi-feature lower die compensation factors include lower die deformation, uneven force on the lower die, residual lower die material, lower die thermal impact and lower die displacement; the associated data of the trimming die is integrated according to the trimming die monitoring data stream to generate a lower die monitoring block; based on the multi-feature lower die compensation factors, a lower die compensation decision is made according to the lower die monitoring block to generate a multi-feature lower die compensation decision; the multi-feature lower die compensation decisions are integrated to obtain the lower die compensation feature decision.

[0069] Furthermore, step S300 also includes: According to the multi-feature lower die compensation factor, feature recognition is performed on the lower die monitoring block to obtain lower die deformation monitoring data; according to the multi-feature lower die compensation factor, a lower die deformation cutting impact evaluation model and a lower die deformation compensation decision model are constructed; based on the lower die deformation monitoring data, according to the lower die deformation cutting impact evaluation model, a lower die deformation cutting impact evaluation coefficient is obtained; it is determined whether the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold; if the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold, the lower die deformation monitoring data is input into the lower die deformation compensation decision model to obtain a lower die deformation compensation decision; the lower die deformation compensation decision is added to the multi-feature lower die compensation decision, and the lower die compensation decision is continued to be performed according to the multi-feature lower die compensation factor and the lower die monitoring block to obtain the multi-feature lower die compensation decision.

[0070] Furthermore, step S400 further includes: According to the trimming die monitoring data stream, the associated data of the trimming guide component is integrated to generate a guide component angle feature point cloud, a guide component position feature point cloud and a guide component state feature information; according to the trimming die monitoring data stream, an upper die position feature point cloud and a lower die position feature point cloud are obtained; according to the upper die position feature point cloud and the lower die position feature point cloud, a matching offset detection is performed to generate an upper and lower die matching offset point cloud; based on the guide component angle feature point cloud, the guide component position feature point cloud and the upper and lower die matching offset point cloud, according to the upper and lower die guide matching compensation decision channel of the guide compensation dual channel, an upper and lower die guide matching compensation decision is obtained; based on the guide component state feature information, according to the guide state abnormality compensation decision channel of the guide compensation dual channel, a guide state abnormality compensation decision is obtained; the upper and lower die guide matching compensation decision and the guide state abnormality compensation decision are integrated to obtain the guide compensation feature decision.

[0071] Furthermore, step S500 further includes: Obtain an expected fixing force for cutting the workpiece and an expected fixing position for cutting the workpiece; perform deviation detection on the fixing force sensing data for cutting the workpiece according to the expected fixing force for cutting the workpiece, and obtain a workpiece fixing force deviation detection result; perform deviation detection on the fixing position sensing data for cutting the workpiece according to the expected fixing position for cutting the workpiece, and obtain a workpiece fixing position deviation detection result; perform associated data integration of the trimming fixing component according to the trimming die monitoring data stream, and obtain fixing component monitoring data; activate a pre-built fixing compensation correction model; input the workpiece fixing force deviation detection result, the workpiece fixing position deviation detection result, and the fixing component monitoring data into the fixing compensation correction model, and obtain the fixing compensation feature decision.

[0072] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0074] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A dynamic compensation control system for a high-precision trimming die, characterized in that: The system comprises: A trimming die monitoring module, the trimming die monitoring module is used to monitor the trimming die in real time and obtain a trimming die monitoring data stream, wherein the trimming die includes a trimming upper die, a trimming lower die, a trimming guide assembly and a trimming fixing assembly; An upper die compensation analysis module, which performs health expectation deviation compensation analysis on the trimming die based on an upper die health expectation prediction channel and in combination with the trimming die monitoring data stream, and generates an upper die compensation feature decision; A lower die compensation analysis module, which performs compensation analysis on the trimming die based on a multi-feature lower die compensation factor and the trimming die monitoring data stream to generate a lower die compensation feature decision; A guide compensation analysis module, which performs compensation analysis on the trimming guide component based on a dual-channel guide compensation decision and combines the trimming die monitoring data stream to generate a guide compensation feature decision; A fixed compensation analysis module, which performs compensation control analysis on the trimming fixing component based on the cutting workpiece fixing force sensing data and the cutting workpiece fixing position sensing data in combination with the trimming die monitoring data stream to generate a fixed compensation feature decision; A trimming compensation execution module is used to perform compensation control on the trimming die based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision.

2. The system according to claim 1, characterized in that The upper die compensation analysis module performs health expectation deviation compensation analysis on the trimming upper die based on the upper die health expectation prediction channel and in combination with the trimming die monitoring data stream, and generates an upper die compensation feature decision, including: Integrate the associated data of the trimming upper die according to the trimming die monitoring data stream to construct an upper die monitoring state matrix; Retrieving the cutting workpiece feature data corresponding to the trimming die and the workpiece cutting upper die control decision; Based on the upper die basic data set of the trimming upper die, combined with the cutting workpiece feature data and the workpiece cutting upper die control decision, the health expectation state of the trimming upper die is predicted according to the upper die health expectation prediction channel to obtain an upper die health expectation matrix; Constructing an upper die health expectation deviation compensation model, and embedding the upper die health expectation deviation compensation model into the upper die compensation analysis module; Based on the upper die monitoring state matrix and the upper die health expectation matrix, and according to the upper die health expectation deviation compensation model, the upper die compensation feature decision is obtained.

3. The system according to claim 2, characterized in that Based on the upper die monitoring state matrix and the upper die health expectation matrix, and according to the upper die health expectation deviation compensation model, the upper die compensation feature decision is obtained, including: Perform deviation identification based on the upper mold monitoring state matrix and the upper mold health expectation matrix to obtain an upper mold monitoring expectation deviation matrix; Perform cutting impact evaluation according to the upper die monitoring expected deviation matrix to generate the upper die deviation cutting impact degree; Determining whether the upper die deviation cutting influence degree is greater than / equal to a predetermined upper die deviation cutting influence degree; If the upper die deviation cutting influence is greater than / equal to the predetermined upper die deviation cutting influence, activating the upper die health expectation deviation compensation model; The upper die monitoring expected deviation matrix is ​​input into the upper die health expected deviation compensation model to generate the upper die compensation feature decision.

4. The system according to claim 1, characterized in that The lower die compensation analysis module performs compensation analysis on the trimming lower die based on the multi-feature lower die compensation factor and the trimming die monitoring data stream to generate a lower die compensation feature decision, including: The multi-feature lower die compensation factors include lower die deformation, lower die force unevenness, lower die material residue, lower die heat effect and lower die displacement; Performing data integration of the trimming lower die according to the trimming die monitoring data stream to generate a lower die monitoring block; Based on the multi-feature lower die compensation factor, a lower die compensation decision is made according to the lower die monitoring block to generate a multi-feature lower die compensation decision; The multi-feature lower die compensation decisions are integrated to obtain the lower die compensation feature decision.

5. The system according to claim 4, characterized in that Based on the multi-feature die compensation factor, a die compensation decision is made according to the die monitoring block to generate a multi-feature die compensation decision, including: Performing feature recognition on the lower die monitoring block according to the multi-feature lower die compensation factor to obtain lower die deformation monitoring data; According to the multi-feature lower die compensation factors, a lower die deformation cutting impact evaluation model and a lower die deformation compensation decision model are constructed; Based on the lower die deformation monitoring data and the lower die deformation cutting impact evaluation model, a lower die deformation cutting impact evaluation coefficient is obtained; Determine whether the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold; If the lower die deformation cutting impact evaluation coefficient is greater than / equal to the lower die deformation cutting impact evaluation threshold, the lower die deformation monitoring data is input into the lower die deformation compensation decision model to obtain a lower die deformation compensation decision; The lower die deformation compensation decision is added to the multi-feature lower die compensation decision, and the lower die compensation decision is continued according to the multi-feature lower die compensation factor and the lower die monitoring block to obtain the multi-feature lower die compensation decision.

6. The system according to claim 1, wherein: The guide compensation analysis module performs compensation analysis on the trimming guide component based on the guide compensation decision dual channel and in combination with the trimming die monitoring data stream to generate a guide compensation feature decision, including: Performing correlation data integration of the trimming guide component according to the trimming die monitoring data stream to generate a guide component angle feature point cloud, a guide component position feature point cloud and a guide component state feature information; According to the trimming die monitoring data stream, a feature point cloud of an upper die position and a feature point cloud of a lower die position are obtained; Performing matching offset detection according to the upper die position feature point cloud and the lower die position feature point cloud to generate upper and lower die matching offset point clouds; The dual guide compensation decision channels include an upper and lower mold guide matching compensation decision channel and a guide state abnormality compensation decision channel; Based on the guide component angle feature point cloud, the guide component position feature point cloud and the upper and lower mold matching offset point cloud, and according to the upper and lower mold guide matching compensation decision channels in the guide compensation decision dual channels, an upper and lower mold guide matching compensation decision is obtained; Based on the state characteristic information of the guide component, and according to the guide state abnormality compensation decision channel in the guide compensation decision dual channel, a guide state abnormality compensation decision is obtained; The guide compensation feature decision is obtained by integrating the upper and lower mold guide matching compensation decision and the guide state abnormality compensation decision.

7. The system according to claim 1, characterized in that The fixed compensation analysis module performs compensation control analysis on the trimming fixing component based on the cutting workpiece fixing force sensing data and the cutting workpiece fixing position sensing data in combination with the trimming die monitoring data stream, and generates a fixed compensation feature decision, including: Obtaining an expected cutting workpiece fixing force and an expected cutting workpiece fixing position; Performing deviation detection on the cutting workpiece fixing force sensing data according to the cutting workpiece fixing force expectation to obtain a workpiece fixing force deviation detection result; Performing deviation detection on the sensing data of the fixed position of the cutting workpiece according to the desired fixed position of the cutting workpiece, and obtaining a detection result of the fixed position deviation of the workpiece; Performing data integration of the associated trimming fixed component according to the trimming die monitoring data stream to obtain fixed component monitoring data; Activate pre-built fixed compensation correction models; The workpiece fixing force deviation detection result, the workpiece fixing position deviation detection result and the fixing component monitoring data are input into the fixing compensation correction model to obtain the fixing compensation feature decision.

8. A dynamic compensation control method for a high-precision trimming die, characterized in that: The method is applied to the system according to any one of claims 1 to 7, and the method comprises: Real-time monitoring of the trimming die to obtain a trimming die monitoring data stream, wherein the trimming die includes a trimming upper die, a trimming lower die, a trimming guide assembly, and a trimming fixing assembly; Based on the upper die health expectation prediction channel, the upper die health expectation deviation compensation analysis is performed on the trimming die in combination with the trimming die monitoring data stream to generate an upper die compensation feature decision; Perform compensation analysis on the trimming die based on the multi-feature die compensation factor and the trimming die monitoring data stream to generate a die compensation feature decision; Based on the dual channels of guide compensation decision, the trimming guide component is compensated and analyzed in combination with the trimming die monitoring data stream to generate a guide compensation feature decision; Based on the sensing data of the fixing force of the cutting workpiece and the sensing data of the fixing position of the cutting workpiece, the compensation control analysis of the trimming fixing component is performed in combination with the trimming die monitoring data stream to generate a fixing compensation feature decision; The trimming die is compensated and controlled based on the upper die compensation feature decision, the lower die compensation feature decision, the guide compensation feature decision and the fixed compensation feature decision.

Images