Dynamic optimization of the radiator stamping process and die state evaluation method

Abstract

This invention relates to a dynamic optimization and mold condition assessment method for the stamping process of radiators in the field of information technology. The method includes: acquiring real-time pressure distribution data and real-time strain data of the sheet metal in key areas of the mold during the stamping process; predicting the material deformation trend using a preset deformation prediction model based on the real-time pressure distribution data and the real-time strain data; determining whether the material deformation trend exceeds a preset tolerance threshold; if the material deformation trend exceeds the tolerance threshold, generating a new combination of process parameters through reverse calculation; sending the new combination of process parameters to the stamping equipment control system for dynamic adjustment; acquiring actual size measurement data of the processed product; associating the actual size measurement data with the process parameter adjustment record and inputting it into a mold wear assessment model; and outputting the current mold wear condition assessment result through the mold wear assessment model.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for dynamic optimization of the radiator stamping process and mold condition evaluation. Background Technology

[0002] In the field of thermal management for modern electronic devices, the design and manufacturing of heat sinks are considered core components for ensuring equipment performance and stability, and their importance is self-evident. With the continuous increase in the power density of electronic products, heat sinks need to possess higher heat dissipation efficiency and more precise structures, which places extremely high demands on manufacturing processes. Stamping and forming of heat sinks, as a mainstream processing method, directly determines product performance and production costs, but it still faces many challenges in practical applications.

[0003] Currently, while radiator stamping methods meet basic requirements to some extent, their limitations cannot be ignored. Many processes are often unable to adapt to the demands of complex structures, especially when dealing with high-density fins or special shape designs, where processing accuracy and efficiency are often limited. This limitation is not only reflected in the quality of individual products but also in the coordination and stability of the entire production process, affecting the feasibility of large-scale production. A deeper analysis of this field reveals that the core technical challenges lie in two key factors. The first is the issue of material deformation control during the stamping process. When metal sheets are subjected to high-intensity extrusion, uneven deformation is prone to occur, leading to uneven fin thickness or distorted shapes, thus affecting heat dissipation performance.

[0004] Secondly, this deformation problem further exacerbates mold wear, especially when machining materials with high hardness. The mold surface wears down rapidly due to frequent friction, causing a continuous decline in machining accuracy. These two factors influence each other, creating a vicious cycle. For example, in the production of high-density heat sink fins, material deformation may cause the fin roots to thin or even break, while mold wear will lead to increasingly larger dimensional deviations in subsequently processed fins, ultimately preventing the product from achieving the designed heat dissipation performance.

[0005] Therefore, effectively controlling material deformation during the stamping process while extending the service life of the mold has become a key issue in improving the quality and efficiency of radiator manufacturing. Solving this problem not only affects the performance of individual products but also directly impacts the stability and economic benefits of the entire production line. Summary of the Invention

[0006] This invention provides a method for dynamic optimization of the heat sink stamping process and mold condition evaluation, mainly including:

[0007] The system acquires real-time pressure distribution data and sheet metal strain data in key areas of the die during the stamping process. Based on the real-time pressure distribution data and the real-time strain data, a preset deformation prediction model is used to predict the material deformation trend. The system determines whether the material deformation trend exceeds a preset tolerance threshold. If the material deformation trend exceeds the tolerance threshold, a new combination of process parameters is generated through reverse calculation. The new combination of process parameters is sent to the stamping equipment control system for dynamic adjustment. The system acquires the actual dimensional measurement data of the processed product. The actual dimensional measurement data is associated with the process parameter adjustment record and input into the die wear assessment model. The die wear assessment model outputs the current die wear status assessment result. If the wear status assessment result reaches a preset maintenance warning line, a die maintenance command is triggered and the relevant parameters of the deformation prediction model are updated. Furthermore, acquiring real-time pressure distribution data of the key areas of the die and real-time strain data of the sheet metal during the stamping process includes: collecting real-time pressure distribution signals from the key areas of the die using sensors; processing the real-time pressure distribution signals using a data acquisition system to determine the real-time pressure distribution data; based on the real-time pressure distribution data, collecting signals from strain sensors attached to the sheet metal surface and calculating the deformation amount to obtain sheet metal deformation monitoring results; if the sheet metal deformation monitoring results exceed a preset threshold, tracking deformation changes in real time using fiber Bragg grating sensors to obtain process parameter adjustment criteria; and based on the process parameter adjustment criteria, performing gridding mapping on the real-time pressure distribution data and comparing it with the sheet metal deformation monitoring results to obtain the final pressure distribution data and strain data during the stamping process. Furthermore, the step of predicting the material deformation trend using a preset deformation prediction model based on the real-time pressure distribution data and the real-time strain data includes: constructing the deformation prediction model by combining historical processing data with material mechanical properties, and determining the basic parameters of the model; inputting the real-time pressure distribution data according to the basic parameters of the model to obtain an initial deformation estimate; calibrating the initial deformation estimate using the real-time strain data to obtain an adjusted deformation value; verifying the adjusted deformation value against process parameter inputs, and determining the material deformation trend in a specific area. Furthermore, the step of determining whether the material deformation trend exceeds a preset tolerance threshold includes: obtaining preset thickness and shape tolerance thresholds through product design specifications; predicting the material deformation trend by combining material properties and historical processing data to obtain a predicted value; determining whether the deformation trend corresponding to the predicted value exceeds the thickness and shape tolerance thresholds, and obtaining a trend judgment result.Furthermore, if the material deformation trend exceeds the tolerance threshold, a new combination of process parameters is generated through reverse calculation, including: obtaining output data from the deformation prediction model; determining the initial process parameter adjustment direction based on the difference between the output data and the initial parameters through reverse calculation, and obtaining a parameter offset value; integrating material characteristic attributes according to the parameter offset value, generating a preliminary combination of process parameters using processing history reference data, and determining the preliminary load distribution; judging whether the equipment load balance meets the preset conditions based on the preliminary load distribution, and if not, adjusting relevant parameters to obtain a balanced parameter set; generating a new combination through the balanced parameter set for verification, obtaining trend judgment result feedback, and determining the final process parameter update. Furthermore, the step of sending the new combination of process parameters to the stamping equipment control system for dynamic adjustment includes: issuing a command to the stamping equipment control system through the new combination of process parameters to obtain parameter replacement results; obtaining equipment response data based on the parameter replacement results; if the equipment response data meets the preset threshold, determining that the dynamic adjustment is complete, and obtaining an optimized combination of process parameters. Furthermore, obtaining the actual size measurement data of the processed product includes: obtaining the size measurement data of the processed product through a precision measuring device; comparing the size measurement data with the standard size to obtain the actual deviation value; if the actual deviation value meets a preset threshold, then determining that the size measurement data meets the requirements, and obtaining the optimized combination of process parameters. Furthermore, associating the actual size measurement data with the process parameter adjustment record and inputting it into the mold wear assessment model includes: obtaining the actual size measurement data and the process parameter adjustment record, generating an associated dataset through data association processing; extracting cumulative usage cycle information from the associated dataset and inputting it into the mold wear assessment model; calculating the wear degree using a linear regression algorithm through the mold wear assessment model; using the cumulative usage cycle information as the independent variable and the size deviation as the dependent variable, fitting a regression equation to obtain a preliminary wear index; integrating production batch tracking information based on the preliminary wear index to determine the final wear assessment output. Furthermore, the step of outputting the current mold wear status assessment result through the mold wear assessment model includes: acquiring historical process parameters and product size deviation data, extracting equipment operation logs from them, and constructing a preliminary correlation dataset; processing the preliminary correlation dataset through statistical analysis methods to determine the correlation between process parameters and size deviations; inputting the correlation into the mold wear assessment model, and calculating the wear index through linear regression in combination with cumulative production cycle information; and outputting the current mold wear status assessment result from the wear index.Furthermore, if the wear condition assessment result reaches a preset maintenance warning line, a mold maintenance command is triggered and the relevant parameters of the deformation prediction model are updated. This includes: extracting cumulative production cycle information from the wear condition assessment result; determining whether the cumulative production cycle information has reached a preset maintenance warning line and obtaining a judgment conclusion; if the judgment conclusion is that the warning line has been reached, triggering a mold maintenance command; synchronously acquiring the latest wear assessment data through the mold maintenance command and determining the data adjustment range; updating the initial parameters related to the mold condition in the deformation prediction model according to the data adjustment range; extracting production cycle monitoring indicators from the initial parameters and judging the stability of the monitoring indicators; adjusting the model accuracy according to the stability of the monitoring indicators and outputting the updated mold condition parameters.

[0008] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0009] To address the complex business scenario of linking fin deformation control and die wear in stamping, this method acquires real-time die pressure and sheet metal strain data, uses a pre-set deformation prediction model to predict material deformation trends, and compares these trends with design tolerance thresholds. If the predicted trend deviates from the expected range, it performs reverse calculations and dynamically issues new combinations of stamping speed and blank holder force parameters to adjust the process in real time. After batch processing, the actual fin size data is associated with the adjustment records and input into a wear assessment model that incorporates the cumulative usage cycle, thereby outputting the die wear status. When the wear reaches the warning line, a maintenance command is triggered, and the initial parameters of the deformation prediction model are updated simultaneously. This achieves a closed-loop integration of dynamic optimization of process parameters and adaptive management of die status, ultimately ensuring both product forming accuracy and die health status within a single production cycle. Attached Figure Description

[0010] Figure 1 This is a flowchart of the dynamic optimization and mold condition evaluation method for the heat sink stamping process of the present invention;

[0011] Figure 2 This is a schematic diagram of the framework of step S102 in the dynamic optimization and mold condition evaluation method for the heat sink stamping process of the present invention.

[0012] Figure 3 This is a schematic diagram of the framework of step S104 in the dynamic optimization and mold condition evaluation method for the heat sink stamping process of the present invention.

[0013] Figure 4 This is a schematic diagram of step S108 in the dynamic optimization and mold condition evaluation method for the heat sink stamping process of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0015] like Figures 1-4 The dynamic optimization and mold condition evaluation method for the heat sink stamping process in this embodiment may specifically include:

[0016] S101. Obtain real-time pressure distribution data and real-time strain data of the sheet metal in the key areas of the die during the stamping process.

[0017] Real-time pressure distribution signals are acquired from key areas of the mold by means of sensor installation locations, and the real-time pressure distribution is determined using a data acquisition system. Based on the real-time pressure distribution, real-time monitoring technology is applied to the sheet metal strain data. This real-time monitoring technology includes acquiring signals from strain sensors attached to the sheet metal surface and calculating the deformation amount to obtain sheet metal deformation monitoring results. If the sheet metal deformation monitoring results exceed a preset threshold, an abnormal deformation alarm is triggered using a strain measurement method. This strain measurement method uses a fiber Bragg grating sensor to track deformation changes in real time, obtaining a basis for adjusting process parameters. Based on the process parameter adjustment basis, pressure distribution analysis is integrated with mold optimization design. This pressure distribution analysis includes meshing the real-time pressure distribution and comparing it with the sheet metal deformation monitoring results to obtain real-time pressure distribution data of key areas of the mold and real-time strain data of the sheet metal during the stamping process.

[0018] In one implementation, the stamping process uses an integrated sensor system to acquire real-time pressure distribution data in key areas of the mold.

[0019] Specifically, pressure sensor arrays are first arranged in key areas of the mold, such as the punch contact surface and the bottom of the mold cavity. These sensors can be piezoelectric or resistive, and are evenly distributed to cover the entire area. During the process, when the press starts, the sensors capture pressure changes in real time, convert them into digital signals through a data acquisition module, and further transmit them to the central processing unit for distribution analysis.

[0020] For example, in automotive parts stamping scenarios, this arrangement ensures spatial resolution of pressure data down to the centimeter level, thus reflecting the uniformity of the die load. Furthermore, real-time strain data acquisition for the sheet metal relies on the installation of strain gauges. Fiber optic strain sensors or strain gauges are fixed to the surface of the sheet metal or embedded in a location; these devices can monitor the tensile, compressive, and shear strains of the sheet metal during stamping deformation.

[0021] It should be noted that the working principle of the strain gauge is based on the change of material resistance or optical signal. When the plate is deformed by force, the sensor outputs a corresponding signal, which is amplified and filtered to generate a real-time data stream.

[0022] In one possible implementation, for sheet metal stamping, such as the forming of electronic device housings, the sensor sampling frequency is set to 100 times per second to capture rapid deformation dynamics.

[0023] Preferably, the synchronous acquisition of pressure distribution data and strain data is achieved through a timestamp mechanism.

[0024] Specifically, the system uses a unified clock source to ensure that the two types of data are correlated within the same time frame, which facilitates subsequent analysis of the interaction between the mold and the sheet metal.

[0025] For example, in a mass production environment, this synchronization helps identify the location of strain anomalies corresponding to pressure peaks, thus preventing defects such as cracks from occurring.

[0026] In one embodiment, the data processing module further performs gridded mapping on the collected pressure distribution.

[0027] For example, the mold area is divided into multiple grid cells, and the pressure value of each cell is calculated by averaging the values ​​from a sensor array, forming a real-time pressure heatmap. This mapping process involves the application of signal denoising and interpolation algorithms. Denoising uses median filtering to remove noise interference, while interpolation estimates blank areas based on data from adjacent points, thus generating a complete distribution map. In precision instrument stamping scenarios, this method improves the accuracy of data visualization, allowing operators to adjust stamping parameters in real time.

[0028] Understandably, the analysis of strain data involves the calculation of the strain tensor.

[0029] Specifically, the collected multi-point strain values ​​are input into the calculation framework, and the direction and magnitude of the principal strain are obtained through coordinate transformation.

[0030] For example, in the stamping of thick plates, such as in the forming of building components, principal strain calculations reveal the plastic deformation threshold of the plate. If this threshold is exceeded, the system can trigger an alarm to prevent overload. This analysis process does not rely on complex numerical simulations but rather on rapid estimation based on a pre-set material property table, ensuring real-time performance. Furthermore, in another implementation, a wireless transmission module is integrated to enable remote data monitoring. Sensor data is transmitted wirelessly to a cloud platform for storage and further processing.

[0031] For example, on a large-scale factory stamping line, this module supports data aggregation from multiple machines to generate overall process statistics, thereby optimizing the production process.

[0032] In one embodiment, the system further includes a feedback control mechanism that adjusts the hydraulic parameters of the press based on real-time pressure and strain data.

[0033] Specifically, if the pressure distribution is uneven, the control unit automatically increases local lubrication or slows down the punch movement to balance the load. In appliance casing stamping applications, this mechanism reduces sheet metal waste and improves forming accuracy.

[0034] It should be noted that these data acquisition methods are applicable to various stamping materials, such as steel plates or aluminum alloys, demonstrating the versatility of the technical solution. Through the above steps, comprehensive monitoring of the stamping process is achieved.

[0035] S102. A preset deformation prediction model is adopted. The model is constructed based on the material mechanical properties and historical processing data. According to the real-time pressure distribution data and the real-time strain data, the material deformation trend of the fin area under the current process parameters is predicted.

[0036] A deformation prediction model is constructed by combining historical processing data with material mechanical properties to obtain the model's basic parameters. Based on the input of real-time pressure distribution according to these basic parameters, the initial deformation estimate for the fin region is determined. The initial deformation estimate is calibrated using real-time strain data to obtain the adjusted deformation value. The adjusted deformation value is then verified against the input process parameters to determine the material deformation trend under fin thickness monitoring.

[0037] In one implementation, the deformation prediction model is constructed based on the material's mechanical properties and historical processing data.

[0038] Specifically, the material's mechanical properties, including elastic modulus, Poisson's ratio, and yield strength, are first collected. These properties are obtained through standard material testing and serve as the basic parameters of the model. Historical processing data is derived from records of previous stamping operations, such as pressure values, strain measurements, and final deformation results. This data forms a dataset used for parameter calibration of the model.

[0039] For example, in the automotive body panel stamping scenario, the model integrates the mechanical properties of aluminum alloys with hundreds of processing histories to generate a predictive framework. Furthermore, the model is constructed using a finite element analysis framework combined with empirical formulas.

[0040] It should be noted that the finite element analysis framework transforms the mechanical properties of materials into a meshed simulation, while historical data is used to correct simulation biases. Specifically, this process involves inputting historical processing data into a regression algorithm to calculate deformation coefficients, thereby adjusting the model's response to real-time inputs.

[0041] In one possible implementation, for sheet metal stamping such as mobile phone casing forming, the model first divides the fin area into mesh cells, and then trains a mapping relationship of deformation trends based on historical data to ensure that the prediction covers tensile and bending deformation.

[0042] Preferably, the prediction process is performed based on real-time pressure distribution data and real-time strain data.

[0043] For example, the system takes this data as input and imports it into a preset model, estimating stress concentration points in the fin region through matrix operations. Predicting deformation trends involves calculating the probability of potential thickness reduction or wrinkle formation, based on material mechanical properties such as stress-strain curves compared to thresholds in historical data.

[0044] For example, in the stamping application of home appliance panels, if the real-time pressure display peak exceeds the historical average, the model predicts that the fin area may experience a deformation deviation of more than 5%, thus prompting an adjustment of process parameters.

[0045] Understandably, the model's construction emphasizes data fusion mechanisms.

[0046] Specifically, material mechanics properties provide the theoretical foundation, such as using Hooke's Law to describe the elastic stage, while historical processing data optimizes model parameters through machine learning methods such as support vector machines. This fusion process includes data cleaning, feature extraction, and validation steps. First, historical data is cleaned to remove outliers; then, key features such as the maximum strain point are extracted; and finally, the model's accuracy in simulated stamping is validated. In the stamping scenario of precision mechanical parts, this mechanism ensures the model's adaptability to plates of different thicknesses, achieving a prediction accuracy of over 90%.

[0047] In one embodiment, the prediction of the material deformation trend in the fin region under the current process parameters is achieved through iterative calculation. Furthermore, the system updates the input data in real time, and the model outputs a deformation trend curve, displaying the deformation path from initial contact to final forming.

[0048] For example, in the stamping of steel sheets such as the forming of building components, the forecast reveals that if the hydraulic parameters remain constant, localized plastic flow may occur in the fin area, leading to dimensional deviations. This output helps the operator to intervene in a timely manner.

[0049] In one embodiment, the model supports multi-material compatibility.

[0050] Preferably, for alloy materials, the mechanical property parameters are adjusted and a specific subset of historical data is incorporated to achieve targeted prediction. In the stamping of electronic component brackets, this method is extended to high-temperature environments, and the predicted deformation trend takes into account the influence of thermal strain.

[0051] It should be noted that these construction and prediction steps demonstrate the versatility of the technical solution in the stamping field, and ensure the robustness of the model through application in different scenarios.

[0052] S103. Determine whether the predicted material deformation trend exceeds the preset fin thickness and shape tolerance threshold, wherein the threshold is determined according to the product design specifications.

[0053] Preset fin thickness and shape tolerance thresholds are obtained from product design specifications, and the predicted values ​​are determined by combining material properties and processing history. A trend judgment is obtained by determining whether the deformation trend corresponding to the predicted value exceeds the thickness and shape tolerance thresholds.

[0054] In one implementation, determining whether the predicted material deformation trend exceeds preset fin thickness and shape tolerance thresholds first requires clarifying the threshold determination process. These thresholds are determined based on product design specifications, which typically include the thickness range and allowable shape deviation values ​​for the fin area. For example, in a stamping process, the specifications might specify a thickness tolerance of ±0.05 mm and a shape tolerance of a bending deviation not exceeding 2 degrees. These thresholds are set by analyzing product functional requirements and material properties to ensure the final product meets mechanical strength and assembly requirements.

[0055] Specifically, threshold determination involves referencing standards such as ISO specifications or internal design guidelines, converting them into quantifiable numerical thresholds for subsequent comparisons. Furthermore, the predicted material deformation trends are derived from the aforementioned model outputs, including thickness reduction rates and shape change curves. The judgment process is performed by comparing these trend values ​​with the thresholds.

[0056] For example, the predicted thickness reduction rate is compared numerically with a preset threshold. If the predicted value exceeds the threshold, it is determined to be excessive. This comparison can employ simple logical operations, such as threshold checks, to ensure real-time performance. In the home appliance panel stamping scenario, this judgment helps identify potential defects and maintain product consistency.

[0057] Preferably, the system can trigger an alarm mechanism when the threshold is exceeded.

[0058] Specifically, if the predicted deformation trend shows that the fin thickness reduction exceeds a threshold of 5%, the system records the deviation point and suggests adjusting the pressure parameters. This mechanism, based on the threshold specification, provides a basis for process optimization.

[0059] For example, in automotive body panel stamping, if the shape tolerance threshold is 1 mm and the predicted trend shows a deviation of 1.2 mm, it is judged as exceeding the tolerance threshold, thus guiding operator intervention.

[0060] It should be noted that the threshold is not fixed based on the product design specifications and can be adjusted according to different materials.

[0061] In one possible implementation, for aluminum alloy fins, the specification emphasizes corrosion resistance, resulting in a stricter thickness threshold, such as ±0.03 mm; while for steel sheet stamping, the specification focuses on strength, with the threshold relaxed to ±0.1 mm. This adjustment process involves reviewing specification documents, extracting key parameters, and inputting them into the system database to ensure the appropriateness of the judgment.

[0062] For example, the judgment step can be integrated into the overall stamping monitoring system. Based on real-time data, the system first obtains the predicted trend, then applies thresholds for point-by-point comparison, such as judging whether the thickness of multiple grid cells of the fin exceeds the specified threshold. This integration method is widely used in the stamping of electronic component brackets and can effectively prevent batch defects. Through this judgment, the system achieves a quantitative assessment of deformation trends, supporting quality control of continuous production lines.

[0063] Understandably, the innovation of this judgment lies in closely combining the predicted trend with the standardized threshold to form a closed-loop feedback.

[0064] Specifically, in the stamping of precision mechanical parts, the threshold determination process involves hierarchical analysis of specifications. First, the product design is decomposed into thickness and shape sub-specifications. Then, thresholds are quantified; for example, shape tolerances are set as specific vector deviation limits based on geometric tolerance theory. Judgment is then made by comparing the predicted curve with the threshold curve using an algorithm, calculating the deviation integral, and determining if the integral exceeds a preset value. This detailed comparison ensures the accuracy of the judgment and brings timely error correction benefits to operations, such as reducing scrap rates.

[0065] In one embodiment, for multiple batch stamping operations, a threshold is determined based on accumulated historical data. Furthermore, the system analyzes past cases that exceed specifications and fine-tunes the threshold to adapt to changes in specifications, always remaining within the product design framework.

[0066] For example, in the stamping of steel plates for building components, this optimization automatically logs when it detects a trend exceeding a threshold, supporting subsequent revisions to the specifications.

[0067] In one embodiment, the decision process supports visual output.

[0068] Preferably, the system generates a chart displaying the deformation trend versus the threshold curve, facilitating intuitive understanding by the operator. In the stamping process of mobile phone casing forming, this output reveals that if the trend exceeds the shape tolerance threshold, it will lead to assembly mismatch, thus prompting parameter adjustments. This visualization enhances the practicality of the judgment.

[0069] S104. If the predicted material deformation trend exceeds the preset threshold, a new combination of stamping speed and blank holder force process parameters is generated by reverse calculation based on the output of the deformation prediction model. The combination takes into account the equipment load balance.

[0070] If the predicted material deformation trend exceeds a preset threshold, output data is obtained from the deformation prediction model. The initial stamping speed and blank holder force adjustment direction are determined by reverse calculation based on the difference between the output data and the initial parameters, resulting in a parameter offset value. Based on the parameter offset value, material properties are integrated, and a preliminary combination of process parameters is generated using historical processing reference data to determine the initial load distribution. For the initial load distribution, it is determined whether the equipment load balance meets the preset balance conditions. If not, the stamping speed and blank holder force are adjusted to obtain a balanced parameter set. A new combination is generated using the balanced parameter set for verification, and trend judgment feedback is obtained to determine the final process parameter update. From the final process parameter update, a new combination of stamping speed and blank holder force process parameters is output.

[0071] In one implementation, if the predicted material deformation trend exceeds a preset threshold, a reverse calculation process is initiated to generate a new combination of stamping speed and blank holder force process parameters based on the output of the deformation prediction model. The deformation prediction model is typically based on finite element analysis or machine learning methods to simulate the stress distribution and deformation path of the material during the stamping process, and its output includes deformation vectors and strain rate data.

[0072] Specifically, the reverse calculation starts with these output data and uses a reverse optimization algorithm to solve for the adjustment value of the input parameters, ensuring that the new parameters can pull the deformation trend back within the threshold.

[0073] For example, in the stamping of home appliance panels, if the model output shows that the thickness reduction rate exceeds a threshold, reverse calculation can iteratively adjust the speed parameters, starting by reducing the initial value by 10% to test the simulation effect. Furthermore, the reverse calculation process involves using the model output as the reverse input to the objective function.

[0074] It should be noted that the deformation prediction model is treated as a black box system here. Its internal mechanism includes meshed material simulation, where each mesh cell records the deformation trend value. During calculation, key deformation indicators exceeding the threshold, such as local stress peaks, are first extracted. Then, the combined value of stamping speed and blank holder force is back-calculated using gradient descent or similar optimization strategies.

[0075] Specifically, the process can be broken down into initializing the parameter set, simulation iteration, and verification steps. For example, during initialization, the speed range is set to 50 to 100 mm / s, and the blank holder force is set to 200 to 500 Newtons. Then, iterative calculations are performed until the simulated deformation meets the threshold. This method, when applied in the stamping of automotive body panels, can effectively address the problem of complex surface deformation.

[0076] Preferably, the generated parameter combination takes into account the equipment load balance.

[0077] For example, load balancing refers to ensuring that the motor torque and hydraulic system pressure do not exceed 80% of the equipment's rated values ​​when the press is operating with new parameters, in order to avoid overload. Constraints are introduced into the calculation, such as coupling the load model with the deformation model, and monitoring total energy consumption.

[0078] In one possible implementation, a reverse computation algorithm is embedded in a load assessment module. This module first generates candidate parameter combinations, then simulates the equipment response, such as checking the hydraulic pump load when the blank holder force increases; if the load exceeds a balance threshold, the combination is discarded. This integration demonstrates versatility in electronic component bracket stamping and supports adjustments for continuous production lines.

[0079] In one embodiment, for multiple batch stamping operations, reverse calculation can accumulate historical data to optimize parameter generation. Furthermore, the system records past model outputs and adjustment results when thresholds were exceeded, forming an experience database to accelerate calculations.

[0080] For example, in the stamping of precision mechanical parts, if the predicted trend indicates that the shape deviation is excessive, the calculation process quickly generates a combination of speed of 80 mm / s and blank holder force of 300 N from the reference database, while verifying load balance to ensure stable equipment operation. This cumulative approach enhances the adaptability of parameter generation.

[0081] Understandably, the innovation of reverse computation lies in forming a closed-loop adjustment between the model output and process parameters.

[0082] Specifically, in the stamping of steel plates for building components, the calculation process is detailed as follows: First, the deformation curve output by the model is analyzed to identify peak points; then, the inverse equations for velocity and blank holder force are solved using numerical inversion methods. For example, assuming a linear relationship, the velocity adjustment is inversely proportional to the deformation rate, and the solution is iteratively solved until convergence; finally, a load balance check is incorporated to simulate the equipment response curve, ensuring that the parameter combination is within a safe range. This detailed process provides a basis for real-time optimization in business operations, such as reducing material waste and maintaining production efficiency.

[0083] In one embodiment, the system supports visual output of new parameter combinations.

[0084] Preferably, the generated chart displays a comparison of deformation trends before and after the adjustment, facilitating operator confirmation. In the stamping process of mobile phone casing forming, this output reveals how parameter changes balance the load, preventing excessive equipment vibration and thus achieving stable process control.

[0085] S105. The new combination of process parameters is sent to the stamping equipment control system to replace the currently executed process parameters, thereby achieving dynamic adjustment.

[0086] By combining the parameters, a command is sent to the stamping equipment control system to obtain parameter replacement results. Based on the parameter replacement results, equipment response data is acquired. If the response data meets a preset threshold, the dynamic adjustment is considered complete, and an optimized combination of process parameters is obtained.

[0087] In one implementation, the new combination of process parameters is sent to the stamping equipment control system, and parameter transmission is achieved through a dedicated communication interface.

[0088] Specifically, the communication interface is based on the Industrial Ethernet protocol, ensuring real-time data transmission and reliability. Upon receiving a new combination, the system first verifies the validity of the parameters, such as checking if the stamping speed and blank holder force are within the equipment's allowable range. Subsequently, the control system loads the new parameters into the execution queue, preparing to replace the current parameters. This distribution process, applied in home appliance panel stamping production lines, enables rapid response to deformation prediction results, avoiding material waste.

[0089] Preferably, a security lockout mechanism is incorporated into the distribution process to prevent equipment malfunctions during parameter replacement.

[0090] It should be noted that safety lock means pausing the current stamping cycle before replacement and resuming operation only after confirming that the new parameters are correct.

[0091] For example, in the stamping scenario of automotive body panels, the system first simulates a virtual cycle under the new parameters, monitors potential load fluctuations, and executes a replacement if everything is normal. This mechanism ensures that dynamic adjustments do not affect the overall production rhythm. Furthermore, when replacing the currently executed process parameters, the control system adopts a gradual update strategy.

[0092] Specifically, this strategy gradually adjusts parameter values, linearly transitioning from the current value to the new value, avoiding sudden changes that could cause equipment vibration or uneven deformation. In the stamping of precision mechanical parts, for example, when adjusting the speed from an initial 80 mm / s to a new value of 70 mm / s, the system decreases the speed in multiple small steps, with each step spaced 2 seconds apart, continuously monitoring material stress during this period. In this way, smooth parameter replacement is achieved, supporting the stable operation of continuous production lines.

[0093] For example, in the stamping operation of electronic component brackets, after a new parameter combination is issued, the control system records a replacement log, including a timestamp and a comparison of the parameters before and after. This log is used for subsequent auditing and optimization, helping to analyze the effect of dynamic adjustments.

[0094] Specifically, if the deformation trend returns to within the threshold after replacement, the system marks it as a successful case, forming a closed-loop feedback. This recording mechanism enhances the traceability of process parameter management and demonstrates practical value in multi-batch production.

[0095] Understandably, the core of achieving dynamic adjustment lies in the immediate effect of parameter replacement.

[0096] In one embodiment, the control system is equipped with a real-time monitoring module that immediately collects equipment sensor data after replacement, such as the deviation between the actual and set values ​​of the blank holder force. If the deviation exceeds a preset threshold, a rollback mechanism is triggered to restore the original parameters. In the stamping of steel plates for building components, this module monitors the entire stamping cycle through an integrated sensor network to ensure that the material deformation after adjustment conforms to the expected path, thereby maintaining production efficiency.

[0097] In one possible implementation, the distribution process supports remote cloud control, allowing operators to confirm new parameter combinations through an interface.

[0098] Specifically, the cloud system generates a visual report that shows a comparison of the device status before and after the replacement, such as changes in load balancing metrics.

[0099] Preferably, this remote method is applied in the stamping and forming of mobile phone casings, facilitating the adjustment of management parameters in a distributed factory and avoiding delays caused by on-site intervention. Furthermore, dynamic adjustment can be extended to multi-device collaborative scenarios.

[0100] For example, in a large stamping workshop, new parameter combinations are simultaneously distributed to multiple machines, and the control system coordinates and updates synchronously to ensure overall consistency of the production line. This collaboration demonstrates its advantages in stamping complex curved parts, reducing batch-to-batch deformation differences by uniformly adjusting speed and blank holder force.

[0101] It should be noted that device compatibility checks are taken into account during the distribution and replacement process.

[0102] Specifically, the system pre-queries the device model and firmware version; if they do not match, it generates a warning and suggests manual intervention.

[0103] In one embodiment, for older stamping presses, this check prevents parameter conflicts and enables safe dynamic adjustments.

[0104] S106. After completing a batch of radiator stamping using the new process parameters, obtain the actual fin size measurement data of the batch of products. The data is obtained through precision measuring equipment.

[0105] Fin size measurement data of the batch of products is obtained using precision equipment and compared with standard dimensions to obtain the actual deviation value. If the actual deviation value meets a preset threshold, the response data is determined to meet the requirements, and an optimized combination of process parameters is obtained.

[0106] In one implementation, after a batch of radiator stamping is completed using new process parameters, the system obtains actual fin size measurement data through precision measuring equipment.

[0107] Specifically, the process first ensures that the stamping cycle is complete and the product has cooled to room temperature to avoid thermal deformation affecting measurement accuracy. Subsequently, measuring equipment such as laser scanners or coordinate measuring machines scan the critical dimensions of the fins, such as height, width, and spacing.

[0108] It should be noted that this acquisition method automates the radiator production line, with equipment directly connected to the control system for real-time data transmission. Furthermore, the precision measuring equipment includes a system combining optical sensors and contact probes to capture the three-dimensional geometric data of the fins.

[0109] For example, in the stamping process of heat sinks for electronic devices, the equipment first calibrates the reference points, then scans the surface of each fin to generate point cloud data, which is subsequently converted into dimensional values. This method ensures data accuracy and supports rapid inspection of batch products.

[0110] Preferably, the data acquisition process incorporates an error correction mechanism. Specifically, the system compares the measured values ​​with the standard tolerance; if the deviation exceeds a threshold, an abnormal sample is marked. In one possible implementation, for high-density finned heat sinks, the measuring device employs multi-angle scanning to cover all sides of the fins, avoiding omissions caused by shading, thereby obtaining a comprehensive dimensional dataset.

[0111] Understandably, this data is uploaded to the analysis module via a dedicated interface.

[0112] For example, in industrial radiator stamping, the data is immediately stored in a structured format after acquisition, facilitating subsequent comparison of the effects of new parameters. This implementation enhances the reliability of process verification.

[0113] For example, after stamping is completed, the operator can manually assist in measuring complex fin areas.

[0114] Specifically, portable measuring instruments are used to fill blind spots in automated equipment and ensure data integrity.

[0115] In one embodiment, this process is applied to the production of aluminum alloy heat sinks, and the measurement data is used to assess fin uniformity and maintain product quality consistency.

[0116] S107. The actual fin size measurement data is associated with the process parameter adjustment record and input into the mold wear assessment model, which integrates the cumulative usage cycle information.

[0117] The actual fin size measurement data and process parameter adjustment records are acquired, and a correlated dataset is generated through data association processing. Cumulative usage cycle information is extracted from this correlated dataset and input into a mold wear assessment model. The model calculates the wear degree using a linear regression algorithm, which fits the data points using the least squares method. Using the cumulative usage cycle as the independent variable and the fin size deviation as the dependent variable, a regression equation is fitted to obtain a preliminary wear index. The equation is in the form y = a multiplied by x plus b, where y is the deviation, x is the cycle, and a and b are coefficients. Based on the preliminary wear index, production batch tracking information is integrated to determine the final mold wear assessment output.

[0118] In one implementation, the specific process of associating actual fin size measurement data with process parameter adjustment records first involves establishing a corresponding relationship through a data management system.

[0119] Specifically, the system matches the measured values ​​such as fin height and width with the stamping pressure and speed parameters in the adjustment records, forming a correlated dataset. This correlation ensures that the impact of parameter changes on dimensions can be traced during subsequent analysis. On the radiator stamping production line, this operation is automated through a software interface, avoiding manual errors. Furthermore, the correlated data is input into the mold wear assessment model.

[0120] It should be noted that the mold wear assessment model is a predictive framework based on historical data, used to assess the degree of wear on molds during use. The core of this model lies in integrating multi-source information and using algorithms to analyze the relationship between dimensional deviations and parameters, thereby inferring wear trends.

[0121] For example, in the aluminum alloy radiator processing scenario, the model first receives the associated dataset, then calculates the difference between the deviation value and the standard value, which serves as the input for the wear index.

[0122] Preferably, the model integrates cumulative usage period information.

[0123] Specifically, the cumulative usage period information refers to the total number of stamping operations or the cumulative time since the die was first used. The fusion process involves incorporating this information as a weighting factor into the model calculation.

[0124] For example, combining periodic data with dimensional measurement deviations can generate a comprehensive wear score.

[0125] In one possible implementation, for high-frequency stamping radiator production lines, the model uses a weighted average method to process this information, ensuring that the assessment accurately reflects the long-term usage impact. This fusion enhances the model's predictive ability for die life.

[0126] For example, after data association is completed, the steps for inputting the model include a preprocessing phase.

[0127] Specifically, the system cleans the associated data, removes outliers, and then converts the cumulative cycle information into a standardized format, such as converting cycle counts into percentages, to facilitate model calculations. In the stamping of heat sinks for electronic devices, this preprocessing helps the model respond quickly and output a wear assessment report.

[0128] Understandably, the internal mechanism of the mold wear assessment model is based on the principle of statistical regression. To elaborate further, this model establishes the relationship between dimensional data, parameter records, and cycle information through regression analysis.

[0129] For example, assuming that the wear rate accelerates with increasing cycles, the model will simulate this trend.

[0130] In one embodiment, for densely finned heat sinks, the model additionally considers the cumulative effect of temperature factors, but the focus remains limited to wear assessment. This interpretation clarifies how the model extracts insights from the data to support production decisions. In another embodiment, the model, which incorporates cumulative usage cycle information, can be extended to batch assessments.

[0131] Specifically, the system processes multiple batches of data at once, associates the measurements and parameters of each batch, and accumulates the periodic information to generate an overall wear curve.

[0132] For example, in industrial radiator production, this method allows operators to monitor the synchronized status of multiple molds, ensuring timely maintenance. Furthermore, the process outputs include wear warnings.

[0133] Specifically, if the wear score exceeds the threshold after model calculation, an alarm will be triggered.

[0134] In one embodiment, for precision heat sink stamping, this output is directly connected to the control system to automatically adjust subsequent parameters. This mechanism optimizes die life and reduces unexpected failures in business operations.

[0135] It should be noted that the connection between the association and input steps is achieved through a real-time database.

[0136] For example, data is immediately correlated after being transmitted from the measuring device and then seamlessly input into the model, avoiding delays. In the field of heat sink manufacturing, this logical flow ensures the continuity of evaluation.

[0137] S108. Using the mold wear assessment model, based on the correlation between historical process parameters and product size deviations, the relationship is constructed through statistical analysis, and the current mold wear status assessment result is output.

[0138] Historical process parameters and product dimensional deviation data are acquired, and equipment operation logs are extracted from this data to construct a preliminary correlation dataset. This preliminary correlation dataset is processed using statistical analysis methods to determine the correlation between process parameters and dimensional deviations. This correlation is then input into a mold wear assessment model, which is pre-trained using historical data. The model's inputs include the correlation and cumulative production cycle information extracted from the equipment operation logs. A wear index is calculated using linear regression. The current mold wear status assessment result is output from the wear index.

[0139] In one implementation, the mold wear assessment model is used to conduct an assessment based on the correlation between historical process parameters and product size deviations.

[0140] It should be noted that the aforementioned relationships are constructed through statistical analysis, and the specific process involves collecting historical data and applying correlation analysis methods.

[0141] For example, on a radiator stamping production line, the system extracts process parameters such as stamping pressure and speed, along with corresponding product dimensional deviation data, from past production records. It then calculates the correlation coefficients between these parameters and the deviations, forming a quantitative association. This construction ensures the model accurately captures the impact of parameter changes on dimensions, supporting subsequent wear assessment. Furthermore, the statistical analysis process includes data screening and model fitting steps.

[0142] Specifically, the system first filters historical data, removing outlier records, and then uses linear regression or other statistical tools to establish the mathematical relationship between parameters and deviations. This is relevant to the aluminum alloy radiator manufacturing process.

[0143] For example, assuming historical data shows that increased stamping pressure leads to greater fin height deviation, the analysis would quantify this trend and generate a correlation coefficient matrix. This method helps identify key influencing factors in a production environment, avoiding blind adjustments.

[0144] Preferably, the model receives the constructed correlation as input and outputs the wear status assessment result by combining it with the current production data.

[0145] Understandably, the output includes wear levels or scores, based on cumulative bias analysis.

[0146] For example, in high-precision heat sink stamping, the model calculates the matching degree between the current dimensional deviation and historical correlations. If the deviation exceeds a preset threshold, a moderate wear condition is output. This output is presented through a software interface, facilitating operator decision-making.

[0147] In one possible implementation, multivariate analysis is incorporated when statistical analysis constructs associations.

[0148] Specifically, the system considers the interaction of multiple process parameters, such as the combined effect of pressure and temperature, and constructs a comprehensive relationship through covariance calculation. On an electronic device heat sink production line, this analysis reveals implicit correlations between parameters, improving the model's predictive accuracy. Further explanation reveals that the core of the construction process lies in data standardization, such as converting parameter values ​​to a unified unit, and then applying Pearson correlation analysis to quantify the strength. This detailed construction ensures the reliability of the correlations.

[0149] For example, when processing a densely finned heat sink, the model outputs a wear status report that includes a deviation trend graph.

[0150] Specifically, based on statistically constructed correlations, the model simulates the wear process, outputting results such as "wear rate reaches 30%" and suggesting maintenance timing. In mass production of industrial radiators, this output is connected to the control system for automated response. Furthermore, in another implementation, the statistical analysis of correlations is extended to time-series data.

[0151] Specifically, the system analyzes the temporal changes of historical parameters and deviations, and uses an autoregressive model to construct dynamic relationships.

[0152] For example, in a continuous stamping production line, this method captures the gradual process of wear, and the output reflects an immediate assessment of the current state. This extension enhances the model's sensitivity to long-term trends.

[0153] It should be noted that the process of outputting the current mold wear condition assessment results emphasizes objective calculation.

[0154] Specifically, after integrating the correlations, the model generates the final score through weighted summation.

[0155] In one embodiment, for precision heat sink molds, the output includes a detailed log recording the sources of deviations to help optimize the process. This mechanism enables continuity in mold management and reduces production interruptions.

[0156] In one embodiment, the model's output further includes an early warning function.

[0157] Specifically, if the correlation analysis based on statistical data shows a continuously increasing deviation, the system outputs a high-risk status and triggers an alarm. In the field of radiator manufacturing, this function ensures timely intervention to maintain production efficiency.

[0158] S109. If the wear condition assessment result reaches the preset mold maintenance early warning line, a mold maintenance or replacement command is triggered, and the initial parameters related to the mold condition in the deformation prediction model are updated simultaneously. The update is based on the latest wear assessment data to adjust the model accuracy.

[0159] The wear status assessment results are obtained, and cumulative production cycle information is extracted from the results. Based on this information, it is determined whether a preset mold maintenance early warning line has been reached, and a judgment conclusion is obtained. If the judgment conclusion is that the early warning line has been reached, a mold maintenance command is triggered. The latest wear assessment data is simultaneously obtained through this command, and the data adjustment range is determined. The initial parameters related to the mold status in the deformation prediction model are updated according to the data adjustment range. Production cycle monitoring indicators are extracted from these parameters, and the stability of the monitoring indicators is judged. The model accuracy is adjusted based on the stability of the monitoring indicators, and the updated mold status parameters are output.

[0160] In one implementation, when the wear condition assessment result reaches the preset mold maintenance warning line, the system automatically triggers a mold maintenance or replacement command.

[0161] Specifically, the early warning line is set based on historical deviation thresholds. For example, on an aluminum alloy radiator stamping production line, if the assessment results show that the wear level exceeds the medium level, the system sends a maintenance command to the operator terminal. This triggering ensures timely intervention and maintains the stable operation of the production line.

[0162] It should be noted that the instructions include specific steps, such as pausing the current batch of production and scheduling the maintenance team. Furthermore, upon triggering the instructions, the system simultaneously updates the initial parameters related to the mold state in the deformation prediction model.

[0163] Understandably, the deformation prediction model is used to estimate the dimensional changes of a product during the stamping process, and these initial parameters include the die hardness coefficient and friction factor. The update process is adjusted based on the latest wear assessment data, for example, by mapping deviations in the assessment results to the parameter space to improve model accuracy. In the scenario of heat sink fin processing, this update helps the model more accurately simulate the potential impact of wear on product deformation.

[0164] Preferably, the update mechanism involves a data integration step.

[0165] Specifically, the system collects the latest wear assessment data, such as cumulative dimensional deviations, and then applies a weighted average method to adjust the initial parameters.

[0166] For example, in the production of high-precision heat sinks, if evaluation data indicates that wear is causing uneven pressure distribution, the system will reduce the hardness coefficient by a corresponding percentage. This adjustment process ensures the model's sensitivity to the current mold condition and improves the reliability of predictions.

[0167] In one possible implementation, the initial parameter update of the deformation prediction model includes a verification step.

[0168] Specifically, after the update, the system runs simulation tests, comparing the predicted deviation with the actual measured value. If the difference narrows, the adjustment is confirmed to be effective. On the stamping line for heat sinks in electronic devices, this verification supports continuous production and avoids quality problems caused by model errors.

[0169] For example, in the batch processing of dense finned heat sinks, after a maintenance command is triggered, the update process generates a log to record parameter changes.

[0170] Specifically, the logs show that initial parameters were adjusted from default values ​​to optimized values ​​based on wear data, such as the friction factor changing from 0.15 to 0.12. This logging facilitates subsequent auditing and optimization. Furthermore, the preset warning lines take into account production environment factors.

[0171] Specifically, in a continuous stamping production line, the early warning line can be dynamically adjusted, and a threshold can be set based on historical data trends.

[0172] For example, if past assessments show a gradual increase in deviation, the system lowers the warning threshold to trigger it earlier. This flexibility enhances the system's adaptability in industrial radiator production.

[0173] In one embodiment, synchronous updates are also incorporated into a feedback loop.

[0174] Specifically, the updated model is applied to the next evaluation cycle, forming a closed-loop optimization. In precision radiator mold management, this cycle reduces human intervention and ensures the continuity of parameter adjustments.

[0175] It should be noted that the execution of the triggering command includes a notification mechanism.

[0176] For example, the system displays maintenance details through an interface and records changes in model accuracy before and after updates. In the field of radiator manufacturing, this mechanism enables efficient mold status management.

[0177] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for dynamic optimization and mold condition evaluation of the radiator stamping process, characterized in that, include: Acquire real-time pressure distribution data and real-time strain data of sheet metal in key areas of the die during the stamping process; Based on the real-time pressure distribution data and the real-time strain data, a preset deformation prediction model is used to predict the material deformation trend. Determine whether the deformation trend of the material exceeds a preset tolerance threshold; If the material deformation trend exceeds the tolerance threshold, a new combination of process parameters is generated through reverse calculation. The new combination of process parameters is sent to the stamping equipment control system for dynamic adjustment; the actual dimensional measurement data of the processed product is obtained; The actual size measurement data is associated with the process parameter adjustment record and then input into the mold wear assessment model; the mold wear assessment model outputs the current mold wear status assessment result; If the wear condition assessment result reaches the preset maintenance warning line, a mold maintenance command is triggered and the relevant parameters of the deformation prediction model are updated.

2. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The acquisition of real-time pressure distribution data and real-time strain data of the sheet metal in the key areas of the die during the stamping process includes: acquiring real-time pressure distribution signals from the key areas of the die using sensors; processing the real-time pressure distribution signals using a data acquisition system to determine the real-time pressure distribution data; based on the real-time pressure distribution data, acquiring signals from strain sensors attached to the sheet metal surface and calculating the deformation amount to obtain sheet metal deformation monitoring results; if the sheet metal deformation monitoring results exceed a preset threshold, tracking deformation changes in real time using fiber Bragg grating sensors to obtain a basis for adjusting process parameters; and based on the process parameter adjustment basis, performing gridding mapping on the real-time pressure distribution data and comparing it with the sheet metal deformation monitoring results to obtain the final pressure distribution data and strain data during the stamping process.

3. The dynamic optimization and mold condition evaluation method for the radiator stamping process as described in claim 1, characterized in that, The step of predicting material deformation trends using a preset deformation prediction model based on the real-time pressure distribution data and the real-time strain data includes: constructing the deformation prediction model by combining historical processing data with material mechanical properties, and determining the basic parameters of the model; inputting the real-time pressure distribution data according to the basic parameters of the model to obtain an initial deformation estimate; calibrating the initial deformation estimate using the real-time strain data to obtain an adjusted deformation value; and verifying the adjusted deformation value against the input process parameters to determine the material deformation trend in a specific region.

4. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The step of determining whether the material deformation trend exceeds a preset tolerance threshold includes: obtaining a preset thickness threshold and shape tolerance threshold through product design specifications; predicting the material deformation trend by combining material characteristics and processing history data to obtain a predicted value; and determining whether the deformation trend corresponding to the predicted value exceeds the thickness threshold and the shape tolerance threshold to obtain a trend judgment result.

5. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, If the material deformation trend exceeds the tolerance threshold, a new combination of process parameters is generated through reverse calculation, including: obtaining output data from the deformation prediction model; determining the adjustment direction of the initial process parameters based on the difference between the output data and the initial parameters through reverse calculation, and obtaining a parameter offset value; integrating material property attributes based on the parameter offset value, generating a preliminary combination of process parameters using historical processing reference data, and determining the preliminary load distribution; judging whether the equipment load balance meets the preset conditions based on the preliminary load distribution, and if not, adjusting the relevant parameters to obtain a balanced parameter set; verifying the new combination generated by the balanced parameter set, obtaining trend judgment feedback, and determining the final process parameter update.

6. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The step of sending the new combination of process parameters to the stamping equipment control system for dynamic adjustment includes: sending an instruction to the stamping equipment control system through the new combination of process parameters to obtain parameter replacement results; obtaining equipment response data based on the parameter replacement results; and determining that the dynamic adjustment is complete and obtaining the optimized combination of process parameters if the equipment response data meets a preset threshold.

7. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The process of obtaining actual size measurement data of the processed product includes: obtaining size measurement data of the processed product through precision measuring equipment; comparing the size measurement data with standard dimensions to obtain actual deviation values; and if the actual deviation values ​​meet a preset threshold, determining that the size measurement data meets the requirements and obtaining an optimized combination of process parameters.

8. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The step of associating the actual size measurement data with the process parameter adjustment records and inputting them into the mold wear assessment model includes: acquiring the actual size measurement data and the process parameter adjustment records, and generating an associated dataset through data association processing; extracting cumulative usage cycle information from the associated dataset and inputting it into the mold wear assessment model; calculating the wear degree using a linear regression algorithm through the mold wear assessment model; fitting a regression equation using the cumulative usage cycle information as the independent variable and the size deviation as the dependent variable to obtain a preliminary wear index; and integrating production batch tracking information based on the preliminary wear index to determine the final wear assessment output.

9. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, The step of outputting the current mold wear status assessment result through the mold wear assessment model includes: acquiring historical process parameters and product size deviation data, extracting equipment operation logs from them, and constructing a preliminary correlation dataset; processing the preliminary correlation dataset using statistical analysis methods to determine the correlation between process parameters and size deviations; inputting the correlation into the mold wear assessment model, combining it with cumulative production cycle information, and calculating wear indicators through linear regression; and outputting the current mold wear status assessment result from the wear indicators.

10. The method for dynamic optimization and mold condition evaluation of the radiator stamping process as described in claim 1, characterized in that, If the wear condition assessment result reaches a preset maintenance warning line, a mold maintenance command is triggered and the relevant parameters of the deformation prediction model are updated, including: extracting cumulative production cycle information from the wear condition assessment result; determining whether the cumulative production cycle information has reached a preset maintenance warning line and obtaining a judgment conclusion; if the judgment conclusion is that the warning line has been reached, triggering a mold maintenance command; synchronously acquiring the latest wear assessment data through the mold maintenance command and determining the data adjustment range; updating the initial parameters related to the mold condition in the deformation prediction model according to the data adjustment range; extracting production cycle monitoring indicators from the initial parameters and judging the stability of the monitoring indicators; adjusting the model accuracy according to the stability of the monitoring indicators and outputting the updated mold condition parameters.

Images