Manufacturing method and device of L-shaped connecting plate and electronic equipment

Abstract

This application discloses a manufacturing method, apparatus, and electronic device for an L-shaped connecting plate, comprising: acquiring multi-dimensional original working condition parameters for the stamping of an asymmetric L-shaped connecting plate and generating an original stamping dataset; constructing a thermal stratification prediction model, combining dual thresholds to divide thermal levels and matching weights, acquiring full-time thermal data, and outputting full-domain stratified time-series thermal data and primary cooling control parameters after inter-frame residual iterative correction; establishing a nonlinear correlation mapping based on the aforementioned data, parameters, and springback residual compensation algorithm, statistically analyzing residuals, calculating cumulative deviations, obtaining springback deviation trends, and outputting mold partition temperature control thresholds and high-precision cooling parameters; establishing a thermal asynchronous equilibrium model and calibrating the thermal parameters of the stamping slide block, acquiring thermal response time difference and thermal deformation misalignment; and matching cooling parameters with forming accuracy constraints to manufacture the asymmetric L-shaped connecting plate. Through the above settings, the quality of the manufactured asymmetric L-shaped connecting plate can be improved.

Description

Technical Field

[0001] This application relates to the field of stamping technology, and in particular to a method, apparatus and electronic device for manufacturing an L-shaped connecting plate. Background Technology

[0002] During the stamping process of asymmetric L-shaped connecting plates, the asymmetric, variable cross-section, and stress concentration of the sheet metal result in significant differences in the plastic deformation amplitude and interfacial frictional heat generation intensity at different locations. Traditional stamping temperature control processes employ a uniform control mode across the entire area, failing to address the differentiated heat generation characteristics of each region. This easily leads to uneven temperature distribution at different locations within the asymmetric L-shaped connecting plate. The root of the bend is prone to excessive heat accumulation due to large plastic deformation, forming localized hot spots and causing the sheet metal to overheat and soften. The folded sidewalls and sheet metal end faces exhibit excessively large temperature gradients due to temperature control lag, inducing quality defects such as workpiece springback, warping, necking, and inaccurate forming curvature. Simultaneously, the asynchronous thermal response of the mold, stamping slide, and other equipment with the asymmetric L-shaped connecting plate further exacerbates the temperature control imbalance, directly reducing the forming accuracy, dimensional consistency, and assembly reliability of the L-shaped connecting plate, failing to meet the quality requirements of high-precision mass production. Summary of the Invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of this application is to provide a method, apparatus and electronic device for manufacturing an L-shaped connecting plate, which can improve the quality of manufacturing L-shaped connecting plates.

[0004] To achieve the above objectives, this application adopts the following technical solution: A method for manufacturing an L-shaped connecting plate, comprising: Obtain multi-dimensional original working condition parameters during the stamping process of asymmetric L-shaped connecting plate, and generate stamping original dataset based on multi-dimensional original working condition parameters; A thermal stratification prediction model is constructed based on the original stamping dataset. The thermal stratification topology of the asymmetric L-shaped connecting plate is divided by combining the dual thresholds of plastic deformation work and interface friction power consumption to define the thermal stratification of the asymmetric L-shaped connecting plate. The deformation heat weight and friction heat weight are matched differently for the thermal stratification. The full-time thermal data of the thermal stratification of the asymmetric L-shaped connecting plate during stamping are obtained simultaneously. The thermal prediction parameters are corrected by combining the inter-frame residual iteration method. The full-domain stratified time-series thermal data of the thermal stratification and the primary cooling control parameters are output. Based on the global hierarchical time-series thermodynamic data, primary cooling control parameters, and springback residual compensation algorithm, a nonlinear correlation mapping relationship is established between the temperature fluctuation and temperature accumulation time of the thermodynamic level and the springback angle and forming curvature of the asymmetric L-shaped connecting plate. This allows for real-time statistical analysis of the temperature time-series residual of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculation. This enables the acquisition of the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold. A thermodynamic asynchronous equilibrium model is established based on the full-domain layered time-series thermodynamic data, zoned temperature control thresholds, and high-precision cooling parameters. The heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate are calibrated to obtain the thermodynamic response time difference and thermal deformation misalignment. By combining the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate, the cooling sequence and cooling power parameters of the mold, stamping slide, and asymmetric L-shaped connecting plate are differentiated and matched to generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, so as to realize the manufacturing of the asymmetric L-shaped connecting plate.

[0005] Furthermore, the multi-dimensional original working parameters include sheet material parameters, real-time stamping load, stamping speed, friction coefficient between the die and sheet contact surface, sheet plastic deformation work, initial temperature of the die and sheet, and equipment thermal inertia parameters; multi-dimensional original working parameters are obtained during the stamping process of the asymmetric L-shaped connecting plate, and a stamping original dataset is generated based on these parameters, including: Lock in the complete stamping process of the asymmetric L-shaped connecting plate, screen out unstable operating periods such as equipment start-up and shutdown and idle standby, and simultaneously collect sheet material parameters, real-time stamping load, stamping speed, friction coefficient of the contact surface between the die and the sheet, plastic deformation work of the sheet, initial temperature of the die and the sheet, and thermal inertia parameters of the equipment during continuous and stable stamping operation. Among them, the material parameters of the sheet include the sheet yield strength, thermal conductivity and sheet thickness parameters, and the thermal inertia parameters of the equipment include the mold heat storage coefficient and the heat dissipation coefficient of the stamping slide. For all the original working condition parameters collected, abnormal data removal operations are carried out one by one to remove the jump data caused by the instantaneous impact of stamping. Then, the sampling noise of the acquisition equipment is removed by filtering and noise reduction. Finally, the working condition parameters of different dimensions and different numerical ranges are subjected to unified dimension normalization processing to obtain standardized effective parameter data. All valid working condition data that have completed preprocessing are classified and organized according to parameter type and stamping sequence, and uniformly packaged into a structured data format to construct the original stamping dataset for subsequent model training and calculation.

[0006] Furthermore, a thermo-stratification prediction model is constructed based on the original stamping dataset. A thermo-stratification topology of the asymmetric L-shaped connecting plate is performed using dual thresholds of plastic deformation work and interfacial frictional power consumption to define the thermo-stratification levels of the asymmetric L-shaped connecting plate. Deformation heat weights and friction heat weights are then applied differentially to the thermo-stratification levels, including: The entire original stamping dataset after regular packaging was used as training input data to complete the parameter training and construction of the thermal stratification prediction model and lock the basic operation parameters of the model. Based on the sheet metal specifications and stamping conditions of the asymmetric L-shaped connecting plate, fixed plastic deformation work threshold and interface friction power consumption threshold are pre-calibrated, and the two thresholds are used simultaneously as the criterion for classifying the thermal levels of the sheet metal. Based on the dual threshold judgment criteria, the bending area, side wall area, and outer edge area of ​​the asymmetric L-shaped connecting plate are subjected to point-by-point topological traversal detection. According to the relationship between the real-time plastic deformation work value, interface friction power consumption value and the preset threshold at each detection point, three independent thermal levels are defined: the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the bend, and the low-heat zone on the end face of the plate. The original stamping data corresponding to each thermal level is retrieved. Based on the differences in the proportion of heat generated by plastic deformation and heat generated by interfacial friction in each level, a unique deformation heat weight and friction heat weight are matched for each type of thermal level. All the differentiated weight coefficients are then embedded into the corresponding level calculation node of the thermal stratification prediction model.

[0007] Furthermore, the full-time thermodynamic data of the thermodynamic hierarchy during the stamping of the asymmetric L-shaped connecting plate is acquired synchronously, and the thermodynamic prediction parameters are corrected by combining the inter-frame residual iteration method. The full-domain layered time-series thermodynamic data of the thermodynamic hierarchy and the primary cooling control parameters are output, including: The defined high-heat zone at the root of the bend, the medium-heat zone on the side wall of the bend, and the low-heat zone on the end face of the sheet are used as independent temperature measurement and detection units. The complete stamping stroke of a single pressing, holding, and lifting is used as an independent time sequence cycle. The real-time temperature data of each sheet thermal unit and the mold contact surface at the corresponding bonding position are collected synchronously using a millisecond-level sampling frequency. The data are sorted and integrated according to the time sequence to construct exclusive full-time thermal data for each thermal level. The full-time thermal data corresponding to each thermal level are input into the thermal stratification prediction model one by one, and the initial thermal prediction results corresponding to each region are calculated. The thermal prediction results output by the model are compared with the actual measured temperature values ​​on site under the same time series, and the inter-frame residual between adjacent stamping time series cycles is accurately calculated. The inter-frame residuals obtained in real time are used as the only iterative correction variable, and the thermal prediction parameters inside the thermal stratification prediction model are replaced and updated one by one through the inter-frame residual iterative method. The calculation continues iteratively until the overall prediction error of the thermal stratification prediction model converges to within the preset accuracy threshold. The output is full-domain stratified time-series thermal data that fits the actual heat generation state of the board. At the same time, based on the temperature amplitude and time-series change patterns of different thermal levels, the corresponding primary cooling control parameters for each region are generated.

[0008] Furthermore, based on the global layered time-series thermodynamic data, primary cooling control parameters, and springback residual compensation algorithm, a nonlinear correlation mapping relationship is established between the temperature fluctuation, temperature accumulation time, and springback angle and forming curvature of the asymmetric L-shaped connecting plate at different thermodynamic levels. This allows for real-time statistical analysis of the temperature time-series residual of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculation. This yields the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters for the mold, including: The global stratified time-series thermal data and the primary cooling control parameters corresponding to each region are uniformly summarized from the iterative output of the thermal stratification prediction model and used as the basic input variables for the algorithm operation, and the rebound residual compensation algorithm is called. For the three independent thermal levels of high heat zone at the root of the bend, medium heat zone on the side wall of the bend, and low heat zone on the end face of the plate, independent calculation channels are built for each level, and nonlinear correlation mapping relationship between temperature fluctuation, temperature accumulation time and springback angle and forming curvature of the asymmetric L-shaped connecting plate is established one by one. Based on the established nonlinear correlation mapping relationship, the standard forming temperature parameters of the sheet metal and the real-time forming temperature parameters are compared sequentially to obtain the temperature time-series residuals corresponding to each thermal level. At the same time, the deformation deviation generated by multiple consecutive stamping processes is superimposed and accumulated. The system records the increase and decrease of the cumulative deviation in real time, and deduces the evolution trend of springback deviation corresponding to different thermal levels in subsequent continuous stamping processes. With the goal of offsetting the springback deviation of sheet metal delamination, the system continuously iterates and corrects the temperature control boundary values ​​of each zone of the mold, and finally outputs the zone temperature control threshold and high-precision cooling parameters of the mold that are adapted to the thermal levels of each sheet metal.

[0009] Furthermore, based on the global layered time-series thermal data, zoned temperature control thresholds, and high-precision cooling parameters, a thermal asynchronous equilibrium model is established. The heat generation response delay, heat dissipation attenuation coefficient, and critical thermal deformation threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate are calibrated to obtain the thermal response time difference and thermal deformation misalignment, including: The model's core input variables are summarized from the overall hierarchical time-series thermal data, mold zone temperature control thresholds, and high-precision cooling parameters. A thermodynamic asynchronous equilibrium model is built based on the heterogeneous thermal characteristics of multi-stamping sliders. The model parameter calibration program is started to perform parameter calibration for three types of structures: mold, stamping slider, and asymmetric L-shaped connecting plate. Among them, the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the overall structure of the mold are calibrated, the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the reciprocating stamping slider are calibrated, and the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the three thermal levels of the asymmetric L-shaped connecting plate are calibrated in layers. After completing the parameter calibration of all stamping slides and all sheet metal areas, the thermal response rate and thermal deformation degree of different stamping slides and different sheet metal thermal regions are compared laterally, and the thermal response time difference and thermal deformation misalignment between stamping slides and sheet metal partitions are quantitatively calculated.

[0010] Furthermore, combining the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece, the cooling sequence and cooling power parameters of the differentially matched mold, stamping slide, and asymmetric L-shaped connecting plate are determined, including: The thermal response time difference and thermal deformation misalignment obtained by quantitative calculation are used as the basis for judging the thermal imbalance of the stamping slider. Combined with the forming accuracy constraint of the asymmetric L-shaped connecting plate workpiece, the allowable forming tolerance threshold of each thermal level of the asymmetric L-shaped connecting plate is locked. Based on the thermal characteristics of the mold and the stamping slide, as well as the forming characteristics of the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the fold, and the low-heat zone on the end face of the sheet, the degree of thermal imbalance and the level of deformation risk of different stamping slides and different sheet areas are distinguished. According to the risk level, the cooling start time and cooling duration are matched with the cooling sequence parameters, and the corresponding cooling power parameters are matched with the corresponding heat dissipation power to complete the differential matching of cooling parameters for all stamping slides and sheet metal sections.

[0011] Furthermore, a method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate is generated to achieve the manufacturing of the asymmetric L-shaped connecting plate, including: The cooling timing and cooling power parameters of the different thermal zones of the mold, stamping slide, and asymmetric L-shaped connecting plate are retrieved and matched. All cooling parameters are unified as the process execution benchmark and integrated to form an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate. During the continuous stamping process, the system simultaneously calls up the pre-completed thermal hierarchy topology division results, the springback deviation evolution trend of each region, and the calibration data of the thermal parameters of multiple stamping slides to monitor the stamping conditions and changes in the thermal state of the sheet metal in real time. Based on real-time operating conditions, the cooling sequence and cooling power parameters corresponding to each stamping slide and each thermal zone of the sheet metal are dynamically fine-tuned to continuously calibrate the stamping temperature of the sheet metal and complete the batch stamping manufacturing of asymmetric L-shaped connecting plates.

[0012] To achieve the above objectives, this application adopts the following technical solution: A method for manufacturing an asymmetric L-shaped connecting plate: An apparatus for manufacturing an asymmetric L-shaped connecting plate, comprising: The acquisition unit is used to acquire multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and generate the original stamping dataset based on the multi-dimensional original working condition parameters. The processing unit is used to construct a thermal stratification prediction model based on the original stamping dataset. It performs thermal stratification topology partitioning of the asymmetric L-shaped connecting plate by combining plastic deformation work and interface friction power consumption as dual thresholds to define the thermal stratification levels of the asymmetric L-shaped connecting plate. It differentiates the deformation heat weight and friction heat weight for each thermal stratification level, simultaneously acquires full-time-series thermal data of the asymmetric L-shaped connecting plate during stamping, and corrects the thermal prediction parameters using an inter-frame residual iteration method. It outputs full-domain stratified time-series thermal data and primary cooling control parameters. Based on the full-domain stratified time-series thermal data, primary cooling control parameters, and springback residual compensation algorithm, it establishes a thermal stratification... The nonlinear correlation mapping between temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate is established to statistically analyze the temperature time-series residual of the asymmetric L-shaped connecting plate in real time and complete the deviation accumulation calculation. This allows for the acquisition of the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold. Based on the global layered time-series thermodynamic data, partitioned temperature control threshold, and high-precision cooling parameters, a thermodynamic asynchronous equilibrium model is established to calibrate the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate, respectively, to obtain the thermodynamic response time difference and thermal deformation misalignment. The control unit combines the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece. It differentiates the cooling sequence and cooling power parameters of the mold, stamping slide, and asymmetric L-shaped connecting plate to generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, thereby realizing the manufacturing of the asymmetric L-shaped connecting plate.

[0013] To achieve the above objectives, this application adopts the following technical solution: An electronic device, comprising: Memory, which stores program instructions; A processor, a method for manufacturing an asymmetric L-shaped connecting plate when the processor executes program instructions stored in memory.

[0014] The aforementioned manufacturing method for L-shaped connecting plates addresses the root cause of uneven temperature control at different locations in asymmetrical L-shaped connecting plates through closed-loop control of multi-dimensional operating condition data acquisition, thermal stratification prediction, springback residual compensation, and asynchronous thermal balancing with differentiated cooling. This method precisely adapts to the differentiated heat generation characteristics of different regions of the asymmetrical L-shaped connecting plate, achieving precise temperature control in zones, eliminating localized hot spots at the bending root and excessive temperature differences between regions, and effectively suppressing forming defects such as workpiece springback dispersion, warping, and necking. Simultaneously, it quantitatively calibrates the asynchronous thermal deviation between the equipment and the workpiece, avoiding stamping center misalignment and die thermal fatigue wear, fully preserving the results of previous precision optimization. This method significantly improves the forming accuracy, batch production consistency, and assembly quality of asymmetrical L-shaped connecting plates, stabilizes the stamping process operation, extends equipment lifespan, and perfectly meets the high-precision, high-consistency industrial mass production requirements of asymmetrical L-shaped connecting plates. Attached Figure Description

[0015] Figure 1 This is a structural block diagram of an electronic device according to an embodiment of this application.

[0016] Figure 2 This is a flowchart illustrating a method for manufacturing an L-shaped connecting plate according to an embodiment of this application.

[0017] Figure 3 This is a structural diagram of the manufacturing apparatus for the L-shaped connecting plate according to an embodiment of this application.

[0018] Reference numerals 100, electronic device; 11, processor; 12, memory; 200, manufacturing apparatus for asymmetric L-shaped connecting plate; 21, acquisition unit; 22, processing unit; 23, control unit. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in specific embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0020] It should be noted that the terms "first," "second," and similar terms used in this application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. "A plurality of" indicates at least two. "Comprising" and similar terms mean that the elements or objects preceding "comprising" cover the elements or objects listed after "comprising" and their equivalents, and do not exclude other elements or objects. "Connection" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0021] The singular forms “a” and “the” used in this application specification and appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0022] The manufacturing method of the L-shaped connecting plate provided in this embodiment can be performed in electronic device 100 or similar device. Figure 1 This is a hardware structure block diagram of an electronic device 100 that implements an embodiment of this application. For example... Figure 1 As shown, the electronic device 100 may include one or more ( Figure 1 (Only one is shown) Memory 12 and processor 11. The electronic device 100 is a control terminal for processing asymmetric L-shaped connecting plates. It is used to control the temperature during the fabrication of asymmetric L-shaped connecting plates in order to improve the quality of the fabricated asymmetric L-shaped connecting plates.

[0023] The memory 12 stores program instructions, such as application software programs and modules, like a computer program for acquiring data from a mechanical water meter in this embodiment. The processor 11 executes the program instructions stored in the memory 12. By running the computer program stored in the memory 12, it can perform various functional applications and data processing, such as achieving constant temperature control when manufacturing the asymmetric L-shaped connecting plate.

[0024] The processor 11 may include, but is not limited to, a microprocessor (MCU) or a programmable gate array (FPGA).

[0025] Those skilled in the art will understand that Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device 100 described above. For example, the electronic device 100 may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown are illustrated.

[0026] In some embodiments, a portion of the method for manufacturing the asymmetric L-shaped connecting plate described below is performed in the electronic device 100, while another portion of the method for manufacturing the asymmetric L-shaped connecting plate can be performed on the related apparatus for manufacturing the asymmetric L-shaped connecting plate. The asymmetric L-shaped connecting plate can be manufactured by the electronic device 100 and the related apparatus for manufacturing the asymmetric L-shaped connecting plate.

[0027] This embodiment also provides a method for manufacturing an asymmetric L-shaped connecting plate. Figure 2This is a flowchart of a method for manufacturing an asymmetric L-shaped connecting plate according to an embodiment of this application. This method is used to improve the quality of the manufactured asymmetric L-shaped connecting plate.

[0028] like Figure 2 As shown, this application provides a method for manufacturing an asymmetric L-shaped connecting plate, comprising: S1 acquires multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and generates the original stamping dataset based on the multi-dimensional original working condition parameters.

[0029] S2 constructs a thermal stratification prediction model based on the original stamping dataset, and combines the dual thresholds of plastic deformation work and interface friction power consumption to perform thermal stratification topology division of the asymmetric L-shaped connecting plate to define the thermal stratification of the asymmetric L-shaped connecting plate. For the thermal stratification, deformation heat weight and friction heat weight are matched differently. The full-time thermal data of the thermal stratification of the asymmetric L-shaped connecting plate during stamping are acquired simultaneously. The thermal prediction parameters are corrected by combining the inter-frame residual iteration method, and the full-domain stratified time-series thermal data of the thermal stratification and primary cooling control parameters are output.

[0030] Based on global layered time-series thermodynamic data, primary cooling control parameters, and springback residual compensation algorithm, S3 establishes a nonlinear correlation mapping relationship between temperature fluctuation, temperature accumulation time, springback angle, and forming curvature of the asymmetric L-shaped connecting plate at different thermodynamic levels. This allows for real-time statistical analysis of the temperature time-series residual of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculation. The goal is to obtain the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold. S4 establishes a thermal asynchronous equilibrium model based on global layered time-series thermal data, zoned temperature control thresholds, and high-precision cooling parameters. It calibrates the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate to obtain the thermal response time difference and thermal deformation misalignment.

[0031] S5 combines the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece. It differentiates the cooling sequence and cooling power parameters of the mold, stamping slide, and asymmetric L-shaped connecting plate to generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, thereby realizing the manufacturing of the asymmetric L-shaped connecting plate.

[0032] In this embodiment, the manufacturing method of the asymmetric L-shaped connecting plate is used to solve the common industry defects caused by the asymmetric, variable cross-section, and stress concentration structural characteristics of the asymmetric L-shaped connecting plate, which make it difficult for traditional stamping forming technology to accurately analyze the thermal distribution, control the springback deviation, and cause the thermal imbalance between the equipment and the workpiece.

[0033] Specifically, the above steps first involve acquiring multi-dimensional raw operating parameters during the continuous stamping process of the asymmetric L-shaped connecting plate. These parameters include the material parameters of the asymmetric L-shaped connecting plate, real-time stamping load, stamping speed, friction coefficient between the die and the plate, plastic deformation work of the plate, initial temperature of the die and the plate, and thermal inertia parameters of the equipment. After abnormal data removal, filtering, noise reduction, and dimensional normalization preprocessing, a standardized raw stamping dataset is generated. Subsequently, a thermodynamic stratification prediction model is constructed based on this raw stamping dataset. Using plastic deformation work and interface friction power consumption as dual thresholds, the asymmetric L-shaped connecting plate is topologically divided into three levels: a nonlinear high-heat zone at the bending root, a gradient medium-heat zone on the folded sidewall, and a linear low-heat zone on the plate end face. Corresponding deformation heat weights and friction heat weights are matched to the differentiated heat generation mechanisms of each thermodynamic level to analyze the thermodynamic characteristics of each region.

[0034] The system synchronously collects millisecond-level full-time thermal data for each thermal level within a single stamping stroke. It extracts the heat accumulation residual and heat dissipation lag residual from the previous stamping cycle to form inter-frame residuals. These residuals are used to iteratively correct the thermal prediction parameters for the next frame and quantify the unique heat accumulation lag effect at the bending root. After the model converges iteratively, it outputs full-domain layered time-series thermal data and primary cooling control parameters adapted to the characteristics of asymmetric structures. This achieves accurate asymmetric thermal perception, providing precise underlying data support for subsequent workpiece springback deviation compensation and coordinated balance control between equipment and workpiece. This avoids potential forming hazards such as localized hot spots, uneven sheet material performance, and warping / necking during asymmetric stamping from the source. Then, using the aforementioned full-domain layered time-series thermal data and primary cooling control parameters as core inputs, and combining them with a springback residual compensation algorithm, a nonlinear correlation mapping relationship is constructed between temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate for each thermal level. The system also statistically analyzes and superimposes the accumulated temperature time-series residuals for each level in the continuous stamping process. Based on the cumulative deviation, the evolution trend of springback deviation at different thermal levels in subsequent stamping processes is deduced. With the goal of offsetting layered springback deviation and unifying sheet metal forming accuracy, the mold partition temperature control threshold and high-precision cooling parameters are iteratively updated to achieve dynamic correction of workpiece springback offset and batch forming accuracy in response to temperature fluctuations.

[0035] Based on the full-domain hierarchical time-series thermal data, the iteratively optimized partition temperature control thresholds, and high-precision cooling parameters, an asynchronous thermal equilibrium model is established. For three types of components with different thermal characteristics—molds, stamping slides, and asymmetric L-shaped connecting plates—the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold are independently calibrated to construct a multi-dimensional heterogeneous thermal parameter library. The thermal response time difference and thermal deformation misalignment between the three types of components are then quantitatively calculated.

[0036] Finally, combining the quantified thermodynamic response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate, and based on the thermodynamic imbalance risk and deformation defect types of each stamping slide and each thermodynamic level of the sheet metal, a differentiated matching of corresponding cooling sequence and cooling power parameters is formed. This results in a stamping temperature adjustment method adapted to the gradient thermodynamic distribution and asynchronous thermodynamic characteristics of multiple stamping slides in the asymmetric L-shaped connecting plate. By using time-sequence misalignment-level temperature control and dynamic closed-loop calibration to offset the thermodynamic deviations of different stamping slides, the method eliminates stamping center offset, die thermal fatigue wear, and forming defects such as sheet metal springback dispersion, warping, and necking. This achieves high-precision, low-loss, and high-consistency mass production of asymmetric L-shaped connecting plates, thereby improving the quality of asymmetric L-shaped connecting plates.

[0037] Based on this, a complete closed-loop technology chain is formed through the sequential coordination and connection of three core technologies: thermal stratification prediction model calculation, rebound residual compensation algorithm correction, and thermal asynchronous equilibrium model control. This chain consists of front-end thermal sensing, mid-level precision compensation, and back-end global steady-state control. Addressing the structural and molding characteristics of the asymmetric L-shaped connecting plate, this solves the problems of the inability of general thermal models to adapt to the nonlinear heat generation of the asymmetric L-shaped connecting plate and the neglect of the time-series cumulative effect of temperature deviation by conventional control algorithms.

[0038] As one implementation method, the multi-dimensional original working parameters include sheet material parameters, real-time stamping load, stamping speed, friction coefficient of the die-sheet contact surface, sheet plastic deformation work, initial temperature of the die and sheet, and equipment thermal inertia parameters; the multi-dimensional original working parameters during the stamping process of the asymmetric L-shaped connecting plate are obtained, and a stamping original dataset is generated based on these parameters, including: Lock in the complete stamping process of the asymmetric L-shaped connecting plate, screen out unstable operating periods such as equipment start-up and shutdown and idle standby, and simultaneously collect sheet material parameters, real-time stamping load, stamping speed, friction coefficient of the contact surface between the die and the sheet, plastic deformation work of the sheet, initial temperature of the die and the sheet, and thermal inertia parameters of the equipment during continuous and stable stamping operation. Among them, the material parameters of the sheet include the sheet yield strength, thermal conductivity and sheet thickness parameters, and the thermal inertia parameters of the equipment include the mold heat storage coefficient and the heat dissipation coefficient of the stamping slide. For all the original working condition parameters collected, abnormal data removal operations are carried out one by one to remove the jump data caused by the instantaneous impact of stamping. Then, the sampling noise of the acquisition equipment is removed by filtering and noise reduction. Finally, the working condition parameters of different dimensions and different numerical ranges are subjected to unified dimension normalization processing to obtain standardized effective parameter data. All valid working condition data that have completed preprocessing are classified and organized according to parameter type and stamping sequence, and uniformly packaged into a structured data format to construct the original stamping dataset for subsequent model training and calculation.

[0039] In this embodiment, the sheet material parameters are used to characterize the inherent material properties of the asymmetric L-shaped connecting plate. The real-time stamping load reflects the instantaneous pressure applied to the sheet by the die during stamping and is directly related to the intensity of the plastic deformation work generated by the sheet. The stamping speed characterizes the movement rate of the stamping equipment, determining the rate of plastic deformation and the rate of heat accumulation in the sheet. The friction coefficient between the die and the sheet quantifies the intensity of friction between them and is a core parameter for calculating interface frictional power consumption. The sheet plastic deformation work characterizes the energy consumed by the sheet under stamping load during plastic deformation, directly corresponding to the total heat generated by the plastic deformation. The initial temperatures of the die and sheet serve as initial thermodynamic state parameters before the stamping process begins, serving as the benchmark values ​​for subsequent thermodynamic calculations and predictions. The equipment thermal inertia parameters characterize the heat storage and dissipation hysteresis characteristics of the die, stamping slide, and other stamping equipment, reflecting the thermodynamic response law of the stamping equipment. The above seven types of parameters together constitute a working condition parameter system covering all dimensions of workpiece properties, stamping motion, interface effects, and heat generation and dissipation. This provides comprehensive and complete raw data support for the subsequent construction of a thermal stratification prediction model for the asymmetric, variable cross-section, and stress concentration structural characteristics of the asymmetric L-shaped connecting plate.

[0040] In acquiring multi-dimensional original working condition parameters during the stamping process of asymmetric L-shaped connecting plates and generating a stamping raw dataset based on these parameters, the process first involves synchronously collecting all dimensions of parameters during the stable operation period of continuous stamping of the asymmetric L-shaped connecting plates. By eliminating invalid data from unstable working conditions such as stamping start-up and shutdown moments and equipment idle standby, the process ensures that the collected parameters accurately reflect the actual working conditions during the stamping process, preventing invalid data from interfering with the accuracy of subsequent model calculations. Secondly, the collected raw working condition parameters undergo preprocessing operations including outlier removal, filtering and denoising, and dimensional normalization. Outlier removal eliminates jump data caused by factors such as instantaneous stamping impact and equipment vibration; filtering and denoising eliminates sampling noise from the acquisition equipment itself; and dimensional normalization unifies the numerical dimensions and ranges of different types of parameters, resolving model calculation deviations caused by differences in parameter dimensions and magnitudes, thereby obtaining standardized and effective parameter data.

[0041] Finally, all the preprocessed valid working condition data are classified and organized according to parameter type and stamping sequence, and uniformly packaged into a structured data format to construct the original stamping dataset. This original stamping dataset serves as the sole underlying input data source for constructing the thermal stratification prediction model and performing thermal stratification topology division. It is the fundamental prerequisite for realizing the thermal stratification iterative prediction of the asymmetric L-shaped connecting plate. Its data integrity and accuracy directly determine the effectiveness of all subsequent control links such as thermal prediction, springback compensation, and thermal equilibrium. Thus, it can fully characterize the core working condition characteristics of the entire stamping process of the asymmetric L-shaped connecting plate. From the data source, it avoids the problems of thermal calculation distortion and control logic failure caused by missing data dimensions, and provides reliable data guarantee for the subsequent original thermal stratification weighted iterative prediction mechanism.

[0042] As one implementation method, a thermo-stratification prediction model is constructed based on the original stamping dataset. A thermo-stratification topology of the asymmetric L-shaped connecting plate is performed by combining dual thresholds of plastic deformation work and interfacial frictional power consumption to define the thermo-stratification levels of the asymmetric L-shaped connecting plate. Deformation heat weights and friction heat weights are then applied differentially to the thermo-stratification levels, including: The entire original stamping dataset after regularization and encapsulation was used as training input data to complete the parameter training and construction of the thermal stratification prediction model, and to lock the basic operational parameters of the model.

[0043] Based on the sheet metal specifications and stamping conditions of the asymmetric L-shaped connecting plate, fixed plastic deformation work threshold and interface friction power consumption threshold are pre-calibrated, and the two thresholds are used simultaneously as the criteria for classifying the thermal levels of the sheet metal.

[0044] Based on the dual threshold judgment criteria, a point-by-point topological traversal detection is performed on the bending area, side wall area, and outer edge area of ​​the end face of the asymmetric L-shaped connecting plate. According to the relationship between the real-time plastic deformation work value, interface friction power consumption value and the preset threshold at each detection point, three independent thermal levels are defined: the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the bend, and the low-heat zone on the end face of the plate.

[0045] The original stamping data corresponding to each thermal level is retrieved. Based on the differences in the proportion of heat generated by plastic deformation and heat generated by interfacial friction in each level, a unique deformation heat weight and friction heat weight are matched for each type of thermal level. All the differentiated weight coefficients are then embedded into the corresponding level calculation node of the thermal stratification prediction model.

[0046] In this embodiment, the original stamping dataset refers to a set of structural working condition data that, after anomaly removal, filtering and noise reduction, and dimensional normalization, covers sheet material parameters, real-time stamping load, stamping speed, friction coefficient between the die and sheet, sheet plastic deformation work, initial temperature of the die and sheet, and equipment thermal inertia parameters. This dataset serves as the sole input data source for constructing the thermal stratification prediction model in this step, and can fully reflect the working conditions of the entire stamping process of the asymmetric L-shaped connecting plate. The thermal stratification prediction model is a dedicated thermal calculation model specifically constructed for asymmetric, variable cross-section, and stress-concentrated asymmetric L-shaped connecting plates, after abandoning the general simulation logic of homogeneous thermal calculation in the traditional stamping field. It achieves accurate analysis of nonlinear heat generation through topological stratification and weight iteration. After the model is constructed, dual thresholds for plastic deformation work and interface friction power consumption are pre-calibrated based on the actual sheet specifications and stamping conditions of the asymmetric L-shaped connecting plate. Among them, the plastic deformation work threshold is a quantitative basis used to determine the heat intensity generated by the plastic deformation of the sheet under the stamping load. The interface friction power consumption threshold is a quantitative benchmark used to determine the heat generation intensity of friction between the mold and the sheet metal contact surface. These two thresholds together constitute the core criterion for thermal hierarchy topology division. Subsequently, based on these two thresholds, a point-by-point topological traversal detection is performed on the entire forming area of ​​the asymmetric L-shaped connecting plate. According to the relationship between the plastic deformation work and interface friction power consumption at each detection point relative to the preset two thresholds, three thermal levels are accurately defined: the high-heat zone at the bend root, the medium-heat zone on the folded sidewall, and the low-heat zone on the sheet metal end face. The high-heat zone at the bend root corresponds to a region of large plastic deformation and strong friction where both plastic deformation work and interface friction power consumption exceed the thresholds; the medium-heat zone on the folded sidewall corresponds to a region of gradient deformation and friction where a single parameter exceeds the threshold; and the low-heat zone on the sheet metal end face corresponds to a region of low heat generation where both parameters are below the thresholds. This hierarchical division method perfectly matches the structural characteristics and heat generation distribution law of the asymmetric L-shaped connecting plate.

[0047] After completing the thermal hierarchy division, this step differentiates the deformation heat generation weights and friction heat generation weights for different thermal hierarchy levels. The deformation heat generation weight is a correction coefficient used to adjust the proportion of heat generated by plastic deformation, while the friction heat generation weight is a correction coefficient used to calibrate the proportion of heat generated by interfacial friction. Since the mechanisms and proportions of plastic and friction heat generation differ significantly in the high-heat zone at the bend root, the medium-heat zone on the folded sidewall, and the low-heat zone on the plate end face, assigning independent weight coefficients to each level completely replaces the traditional uniform parameter calculation mode. This allows for accurate analysis of the differentiated heat generation mechanisms at each thermal hierarchy level. Subsequently, the differentiated deformation heat generation weights and friction heat generation weights are embedded into the corresponding computation nodes of the thermal stratification prediction model, enabling the model to analyze the nonlinear heat generation of the asymmetric L-shaped connecting plate.

[0048] As one implementation method, the full-time thermodynamic data of the thermodynamic levels during the stamping of the asymmetric L-shaped connecting plate are acquired simultaneously, and the thermodynamic prediction parameters are corrected by combining the inter-frame residual iteration method. The full-domain layered time-series thermodynamic data of the thermodynamic levels and the primary cooling control parameters are output, including: The defined high-heat zone at the root of the bend, the medium-heat zone on the side wall of the fold, and the low-heat zone on the end face of the sheet are used as independent temperature measurement and detection units. Taking the complete stamping stroke of a single pressing, holding, and lifting as an independent time sequence cycle, the real-time temperature data of each sheet thermal unit and the mold contact surface at the corresponding bonding position are collected synchronously using a millisecond-level sampling frequency. The data are sorted and integrated according to the time sequence to construct exclusive full-time thermal data for each thermal level.

[0049] The full-time thermal data corresponding to each thermal level are input into the thermal stratification prediction model one by one, and the initial thermal prediction results corresponding to each region are calculated. The thermal prediction results output by the model are compared with the actual measured temperature values ​​on site under the same time series, and the inter-frame residuals between adjacent stamping time series are accurately calculated.

[0050] The inter-frame residuals obtained in real time are used as the only iterative correction variable. The thermal prediction parameters inside the thermal stratification prediction model are replaced and updated one by one through the inter-frame residual iteration method.

[0051] The calculation continues iteratively until the overall prediction error of the model converges to within the preset accuracy threshold. The output is full-domain layered time-series thermal data that fits the actual heat generation state of the board. At the same time, based on the temperature amplitude and time-series change patterns of different thermal levels, the corresponding primary cooling control parameters for each region are generated.

[0052] In this embodiment, for the core operations of inter-frame residual iterative correction and hierarchical adaptive cooling control, all execution actions revolve around the asymmetric, variable cross-section, and stress concentration structural characteristics of the asymmetric L-shaped connecting plate itself. The already defined high-heat zone at the bend root, the medium-heat zone on the folded sidewall, and the low-heat zone on the plate end face are each treated as independent temperature detection units. Because the plastic deformation amplitude and interfacial friction stroke of different regions of the asymmetric L-shaped connecting plate have inherent differences during the stamping process—the bend root is accompanied by large plastic deformation and strong interfacial friction, the folded sidewall by medium gradient deformation and conventional friction, and the plate end face by only weak deformation and slight friction—the heat generation rate and heat accumulation mode of these three types of regions are completely different. Therefore, a unified detection method cannot be used to obtain temperature data; it is necessary to collect data in a zone-independent manner, and then treat a single complete stamping stroke as an independent time cycle. This cycle includes all the process stages of pressing the slide block down, holding the sheet metal under pressure and forming, and lifting the slide block back. It covers the complete thermodynamic change process of the asymmetric L-shaped connecting plate from plastic deformation and rapid heat accumulation to initial heat dissipation, and can completely record the temperature evolution of the sheet metal in the entire forming process.

[0053] Based on this, millisecond-level acquisition frequency is used to synchronously acquire real-time temperature data of each thermal unit and the corresponding mold contact surface. This acquisition frequency is sufficient to capture the instantaneous nonlinear temperature rise caused by large plastic deformation in the high-heat zone at the bending root, the gradient temperature change in the production area of ​​the folded sidewall, and the stable temperature fluctuation in the low-heat zone of the sheet end face, avoiding the loss of thermal change details due to insufficient acquisition frequency. After acquisition, the temperature data is organized and integrated in chronological order to form full-time thermal data for each thermal level. This data is a continuous time-series temperature set, completely recording the temperature rise and fall and heat accumulation state of the corresponding area within a single stamping cycle. Subsequently, the full-time thermal data corresponding to each thermal level is input into the thermal stratification prediction model one by one. The model completes the calculation based on the pre-configured deformation heat weight and friction heat weight, and outputs the initial thermal prediction results for each area. Then, the thermal prediction results output by the model at the same time node are directly compared with the actual temperature values ​​collected on site to calculate the inter-frame residual between adjacent stamping time cycles. The residual directly reflects the deviation between the model's prediction and the actual heat generation state of the plate, including deviation information that traditional homogeneous thermal models cannot identify, such as the residual of heat accumulation at the bend root and the residual of heat dissipation lag in various regions. To eliminate the impact of such deviations on the accuracy of thermal prediction, the inter-frame residuals calculated in real time are used as iterative correction variables. The thermal prediction parameters inside the thermal stratification prediction model are adjusted successively through inter-frame residual iteration. During continuous iteration, the model continuously adapts to the nonlinear heat generation laws of various regions of the asymmetric L-shaped connecting plate, especially quantifying and calibrating the unique heat accumulation lag effect at the bend root, until the overall prediction error of the model converges to within the preset accuracy threshold range.

[0054] The temperature data output by the model at this point is the full-domain layered time-series thermodynamic data. This data fully preserves the heat generation differences of different thermodynamic levels and the time-series change characteristics of the entire cycle, which can truly reflect the actual thermal distribution state of the asymmetric L-shaped connecting plate during the stamping process. At the same time, based on the heat generation intensity, temperature change amplitude, and heat accumulation rate of each thermodynamic level reflected in the full-domain layered time-series thermodynamic data, primary cooling control parameters adapted to different thermodynamic levels are directly matched and generated. These parameters fit the heat generation characteristics of each region and can be directly used to initially control the cooling state of the corresponding region. The execution logic of the entire step completely follows the thermodynamic change law of the asymmetric L-shaped connecting plate. All technical features revolve around accurately capturing, correcting, and outputting the layered thermodynamic data of the asymmetric L-shaped connecting plate. Through the coherent operation of independent acquisition of partitions, full-time recording, and inter-frame residual iterative calibration, the core defects of traditional full-domain homogeneous thermodynamic calculation models, such as the inability to quantify the nonlinear accumulated heat at the bending root and the inability to identify the gradient temperature difference of the plate, are directly solved, allowing the thermodynamic prediction results to fully fit the actual heat generation state of the asymmetric L-shaped connecting plate. Furthermore, the output global hierarchical time-series thermodynamic data possesses both zonal differentiation and temporal continuity. The generated primary cooling control parameters can directly match the heat generation control requirements of each thermodynamic level. The entire process, from data acquisition and model calculation to parameter output, achieves precise adaptation to the thermodynamic characteristics of asymmetric stamping, providing direct and accurate thermodynamic data for subsequent process steps.

[0055] As one implementation method, based on global layered time-series thermodynamic data, primary cooling control parameters, and springback residual compensation algorithms, a nonlinear correlation mapping relationship is established between temperature fluctuations and temperature accumulation time at different thermodynamic levels and the springback angle and forming curvature of the asymmetric L-shaped connecting plate. This allows for real-time statistical analysis of the temperature time-series residuals of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculations. This yields the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control thresholds and high-precision cooling parameters for the mold, including: The global hierarchical time-series thermodynamic data output by the model iteration and the primary cooling control parameters corresponding to each region are uniformly summarized and used as the basic input variables for the algorithm operation, and the rebound residual compensation algorithm is called.

[0056] For the three independent thermal levels of high heat zone at the root of the bend, medium heat zone on the side wall of the fold, and low heat zone on the end face of the plate, independent calculation channels are built for each level, and nonlinear correlation mapping relationship between temperature fluctuation, temperature accumulation time and springback angle and forming curvature of the asymmetric L-shaped connecting plate is established one by one.

[0057] Based on the established nonlinear correlation mapping relationship, the standard forming temperature parameters of the sheet metal and the real-time forming temperature parameters are compared sequentially to obtain the temperature time-series residuals corresponding to each thermal level. At the same time, the deformation deviation generated by multiple consecutive stamping processes is superimposed and accumulated.

[0058] The system records the increase and decrease of the cumulative deviation in real time, and deduces the evolution trend of springback deviation corresponding to different thermal levels in subsequent continuous stamping processes. With the goal of offsetting the springback deviation of sheet metal delamination, the system continuously iterates and corrects the temperature control boundary values ​​of each zone of the mold, and finally outputs the zone temperature control threshold and high-precision cooling parameters of the mold that are adapted to the thermal levels of each sheet metal.

[0059] In this embodiment, to address the issues of springback deviation, inaccurate forming curvature, and poor consistency in batch workpiece precision caused by zoned temperature fluctuations and heat accumulation during continuous stamping of asymmetric L-shaped connecting plates, a springback residual compensation algorithm is introduced based on pre-collected and iteratively corrected global layered time-series thermodynamic data and primary cooling control parameters. This algorithm achieves dynamic closed-loop control of workpiece forming precision by constructing zone-specific nonlinear correlation mapping relationships, statistically analyzing accumulated time-series residuals, predicting springback deviation trends, and iteratively optimizing control parameters. The global layered time-series thermodynamic data serves as the core input, comprehensively encompassing the temperature amplitude changes, temporal rise and fall patterns, and spatiotemporal distribution characteristics of the high-heat zone at the bending root, the middle-heat zone on the folded sidewall, and the low-heat zone on the plate end face throughout the stamping cycle. This accurately reflects the differentiated heat generation and dissipation states of each region of the asymmetric L-shaped connecting plate. The primary cooling control parameters are the initial cooling control parameters generated by matching the heat generation intensity of each thermal level. They provide a control benchmark that fits the actual working conditions for the correlation analysis of temperature and rebound deformation. The rebound residual compensation algorithm abandons the industry's conventional instantaneous single-point correction control logic.

[0060] After the springback residual compensation algorithm is activated, it first constructs a nonlinear correlation mapping relationship between temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate for the three types of thermal levels. Among them, temperature fluctuation is used to characterize the temperature fluctuation amplitude of the corresponding thermal level within a single time segment, directly reflecting the degree of instantaneous temperature change in the region. Temperature accumulation time is used to quantify the duration of heat action inside the sheet during continuous stamping, reflecting the long-term effect of temperature on sheet deformation. Springback angle and forming curvature are the core indicators for measuring the forming accuracy of the asymmetric L-shaped connecting plate. Due to the drastic temperature fluctuations and significant heat accumulation effect in the high-heat zone at the root of the bend, the temperature in the production zone of the folded sidewall exhibits a gradual gradient, while the temperature in the low-heat zone at the end face of the sheet remains relatively stable. The degree and mechanism of influence of temperature changes at different thermal levels on springback deformation and curvature changes are significantly different. Therefore, it is necessary to fit a dedicated three-dimensional nonlinear correlation function for each thermal level and calibrate the springback-sensitive nonlinear coefficient to accurately characterize the inherent coupling law between temperature factors and forming deformation, and avoid analytical deviations caused by a uniform mapping relationship across the entire domain.

[0061] After constructing the nonlinear correlation mapping, real-time forming temperature parameters of each thermal level in the continuous stamping process are continuously collected. These parameters are compared with the standard forming temperature parameters of the corresponding level time-series and point-by-point. The temperature time-series residuals of each level are statistically obtained in real time. These residuals are the difference between the actual temperature and the standard temperature in a single time series, which intuitively reflects the temperature deviation within a single cycle. This method abandons the traditional single-time deviation zeroing process and instead superimposes the temperature time-series residuals generated in each stamping cycle to complete the deviation accumulation calculation. This quantifies the deformation accumulation effect caused by the gradual amplification of small temperature deviations during long-term continuous stamping, which is consistent with the actual working conditions of industrial continuous mass production. Based on the accumulated deviation and the constructed partitioned nonlinear correlation mapping relationship, the evolution trend of springback deviation corresponding to different thermal levels in subsequent stamping processes is deduced. This trend can identify the occurrence pattern, risk points, and development range of sheet metal springback offset and warping deformation in advance.

[0062] Finally, with the goal of offsetting layered springback deviation and unifying the forming accuracy of the asymmetric L-shaped connecting plate, the temperature control boundaries of each area of ​​the mold are continuously iteratively corrected based on the predicted evolution trend of springback deviation. This ultimately outputs mold zone temperature control thresholds and high-precision cooling parameters adapted to different thermal levels. The mold zone temperature control thresholds are temperature control boundaries set differently for the temperature response characteristics and springback sensitivity of different thermal levels. The high-precision cooling parameters are optimized cooling control parameters that, after springback deviation correction, are fully adapted to the real-time deformation control needs of each area. They can directly replace the initial cooling parameters to actively suppress temperature-induced springback. The entire process revolves solely around the coupled control of temperature fluctuations, heat accumulation, and workpiece springback deformation. Through a series of operations including partitioned correlation analysis, time-series residual accumulation, trend prediction, and parameter iterative optimization, the temperature and springback coupling characteristics of each partition of the asymmetric L-shaped connecting plate are precisely adapted. This effectively eliminates the loss of forming accuracy caused by the accumulation of temperature deviations during continuous stamping, stabilizes the springback angle and forming curvature of the workpiece, and significantly improves the forming consistency and assembly accuracy of batch workpieces. At the same time, it provides accurate and reliable precision control parameters for the stable execution of subsequent forming processes, thereby enhancing the stability and reliability of the entire stamping forming process.

[0063] As one implementation method, a thermodynamic asynchronous equilibrium model is established based on global layered time-series thermodynamic data, zoned temperature control thresholds, and high-precision cooling parameters. The heat generation response delay, heat dissipation attenuation coefficient, and critical thermal deformation threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate are calibrated to obtain the thermodynamic response time difference and thermal deformation misalignment, including: The model uses comprehensive hierarchical time-series thermal data, mold zone temperature control thresholds, and high-precision cooling parameters as core input variables. A thermodynamic asynchronous equilibrium model is built based on the heterogeneous thermal characteristics of multi-stamping sliders. The model parameter calibration program is then started to perform parameter calibration for three types of structures: mold, stamping slider, and asymmetric L-shaped connecting plate.

[0064] Specifically, the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold are calibrated for the overall structure of the mold; the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold are calibrated for the reciprocating stamping slide block; and the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold are calibrated layer by layer for the three thermal levels of the asymmetric L-shaped connecting plate.

[0065] After completing the parameter calibration of all stamping slides and all sheet metal areas, the thermal response rate and thermal deformation degree of different stamping slides and different sheet metal thermal regions are compared laterally, and the thermal response time difference and thermal deformation misalignment between stamping slides and sheet metal partitions are quantitatively calculated.

[0066] In this embodiment, addressing the practical problems of asynchronous thermal response and misaligned thermal deformation caused by inherent differences in material properties and working conditions among the die, stamping slide, and asymmetric L-shaped connecting plate during the stamping process, a thermodynamic asynchronous equilibrium model adapted to the heterogeneous thermal characteristics of multiple stamping slides is constructed using full-domain layered time-series thermodynamic data, zoned temperature control thresholds, and high-precision cooling parameters obtained from previous processes as all inputs. By independently calibrating the key thermodynamic parameters of the three core components—die, stamping slide, and asymmetric L-shaped connecting plate—the thermodynamic imbalance state between different components is quantitatively analyzed and corresponding quantitative indicators are output. The full-domain layered time-series thermodynamic data comprehensively records the temperature changes, heat accumulation, and heat dissipation processes of the high-heat zone at the bending root, the middle production zone on the folded sidewall, and the low-heat zone on the plate end face of the asymmetric L-shaped connecting plate during continuous stamping cycles, accurately reflecting the real-time thermal state of each area of ​​the workpiece. The zoned temperature control thresholds are temperature control boundaries set based on the springback sensitivity and heat generation intensity of each thermal level of the workpiece, stably maintaining the temperature range required for workpiece forming. High-precision cooling parameters are control parameters adapted to the precise heat dissipation requirements of each area after springback deviation correction. These parameters together constitute the complete data foundation for model calculation, ensuring that model construction and parameter calibration perfectly match actual stamping conditions.

[0067] After the model was built, calibration was performed on the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold for the three types of components. The heat generation response delay characterizes the lag time after the components are subjected to stamping heat before their internal temperature begins to change significantly, directly reflecting the speed of their thermal response. The heat dissipation attenuation coefficient characterizes the rate at which the components dissipate heat after the heat input is removed, serving as a core indicator of their self-heat dissipation capability. The thermal deformation critical threshold defines the critical temperature at which the temperature of the components rises to cause irreversible geometric deformation, providing a direct basis for assessing the risk of thermal deformation. Because the mold is made of heavy alloy material, it is in long-term contact with the sheet metal and continuously stores heat, exhibiting high thermal inertia, slow response, and slow heat dissipation. Therefore, the three parameters were calibrated based on the mold's actual working conditions and material properties. The stamping slide performs high-frequency reciprocating linear motion, continuously exchanging heat with the air during movement with short single heat generation times, exhibiting rapid heat generation, rapid heat dissipation, and rapid thermal response. Therefore, the three parameters adapted to its motion conditions need to be calibrated independently. The asymmetric L-shaped connecting plate has different zones: a high-heat zone at the root of the bend, a medium-heat zone on the side wall of the fold, and a low-heat zone on the end face of the plate. The thickness, heat generation intensity, and deformation sensitivity of different zones are different. Therefore, it is necessary to calibrate the heat generation response delay, heat dissipation attenuation coefficient, and critical threshold of thermal deformation for each zone in layers to ensure that the parameter calibration results match the asymmetric structural characteristics of the workpiece.

[0068] After completing the parameter calibration of the three types of components, the calibrated parameters are substituted into the thermodynamic asynchronous equilibrium model for horizontal comparison calculation. The thermodynamic response time difference and thermal deformation misalignment between the mold, the stamping slide, and the asymmetric L-shaped connecting plate are calculated respectively. The thermodynamic response time difference is the difference in the heat generation response delay of different stamping slides, which can intuitively quantify the time difference of heating between the stamping slides. Thermal deformation misalignment is the difference in geometric deformation offset caused by different stamping slides reaching the critical threshold of thermal deformation. It directly reflects the degree of structural misalignment caused by thermal imbalance and is a key factor leading to stamping center offset and decreased forming accuracy. The entire process revolves only around the calibration of thermal parameters of multiple stamping slides and the quantification of thermal imbalance, without involving subsequent cooling sequence matching, cooling power adjustment, or other related content. By accurately calibrating the core thermal parameters of three types of components and quantifying the thermal imbalance index, the asynchronous thermal law between the equipment and the workpiece can be clearly identified. This avoids the optimization effect of the previous thermal sensing and workpiece accuracy compensation being negated by equipment thermal deformation and stamping center offset, stabilizing the overall working condition of the stamping operation. It provides accurate and reliable quantitative basis for subsequent coordinated temperature control of the equipment and workpiece, retains the optimization results of the previous process to the greatest extent, and continuously ensures the dimensional accuracy and process stability of the asymmetric L-shaped connecting plate stamping.

[0069] As one implementation method, combining the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece, the cooling sequence and cooling power parameters of the matching mold, stamping slide, and asymmetric L-shaped connecting plate are differentiated, including: The thermal response time difference and thermal deformation misalignment obtained by quantitative calculation are used as the basis for judging the thermal imbalance of the stamping slider. Combined with the forming accuracy constraint of the asymmetric L-shaped connecting plate workpiece, the allowable forming tolerance threshold of each thermal level of the asymmetric L-shaped connecting plate is locked. Based on the thermal characteristics of the mold and the stamping slide, as well as the forming characteristics of the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the fold, and the low-heat zone on the end face of the sheet, the degree of thermal imbalance and the level of deformation risk of different stamping slides and different sheet areas are distinguished. According to the risk level, the cooling start time and cooling duration are matched with the cooling sequence parameters, and the corresponding cooling power parameters are matched with the corresponding heat dissipation power to complete the differential matching of cooling parameters for all stamping slides and sheet metal sections.

[0070] In this embodiment, based on the actual working conditions of the whole machine's coordinated temperature control in the continuous stamping forming of the asymmetric L-shaped connecting plate, and addressing the common industry problem of asynchronous thermal response and superimposed thermal deformation misalignment caused by the essential differences in material properties, stress states, and motion characteristics among the mold, stamping slide, and workpiece, the thermal response time difference and thermal deformation misalignment obtained through the calibration and coupling calculation of the thermal characteristics of multiple stamping slides are used as the core quantitative judgment basis for the thermal imbalance of the whole machine. Simultaneously, considering the asymmetric variable cross-section and stress concentration structural characteristics of the asymmetric L-shaped connecting plate itself, as well as the workpiece forming accuracy constraints determined by subsequent assembly and use requirements, differentiated matching of cooling sequence and cooling power parameters is performed for each thermal region of the mold, stamping slide, and asymmetric L-shaped connecting plate. The thermal response time difference directly reflects the time difference between the mold's heat storage lag, the rapid heating and cooling of the slide, and the gradient heat generation of the workpiece, accurately reflecting the asynchronous characteristics of the thermal response of multiple stamping slides. Thermal deformation misalignment is a numerical offset directly related to the stability of the stamping center, the die fitting accuracy, and the workpiece forming deformation. It is a core imbalance indicator that cooling control needs to address. The above parameters collectively provide an objective and accurate quantitative benchmark for cooling parameter matching. The workpiece forming accuracy constraint is a rigid forming boundary defined by combining the structural characteristics of the asymmetric L-shaped connecting plate and mass production tolerance requirements. This limits the allowable deformation upper limit for each area of ​​the workpiece, ensuring that cooling control always revolves around the forming accuracy target. Cooling sequence is used to define the start point, duration, and intermittent rules of the cooling action. Its core function is to adapt to the thermal response rhythm of different stamping slides, avoiding a disconnect between the cooling action and the heat generation pattern of the stamping slide. Cooling power parameters are used to set the heat dissipation intensity per unit time. Their core function is to match the heat generation rate and thermal deformation risk level of each controlled object, achieving precise heat dissipation rather than homogeneous cooling across the entire area.

[0071] In the specific control and execution process, the mold, as the core pressure-bearing forming component of stamping, is made of high thermal inertia alloy material. It is in close contact with the sheet metal for a long time without active heat dissipation conditions. The heat generation response delay is long and the heat dissipation attenuation coefficient is low. Heat is easy to accumulate inside and cause gradual thermal deformation. Therefore, based on its thermal lag and slow heat accumulation characteristics, a delayed start and long-term continuous cooling sequence is configured, and a stable cooling power that is adapted to its heat capacity is matched. Heat dissipation is smoothly intervened before the heat accumulation in the mold reaches the critical range of thermal deformation, which not only continuously dissipates the internal heat accumulation, but also avoids the internal stress and dimensional deviation of the mold caused by sudden temperature changes. As a power transmission component of high-frequency reciprocating motion, the stamping slide has a short motion cycle, short single contact time, and good heat exchange conditions with the air. There is only a risk of instantaneous temperature rise and no long-term heat accumulation. Therefore, a synchronous intermittent cooling sequence is matched, and cooling is precisely started during the intervals when the slide stops moving. A moderate and constant cooling power is configured to quickly suppress the instantaneous temperature rise without interfering with the normal stamping motion rhythm of the slide.

[0072] The asymmetric L-shaped connecting plate, as the formed workpiece, is divided into three thermal levels: a high-heat zone at the bending root, a medium-heat zone on the folded sidewall, and a low-heat zone on the plate end face. Each region exhibits significant differences in heat generation intensity and thermal deformation sensitivity. The high-heat zone at the bending root experiences intense plastic deformation and concentrated frictional heat generation, with the highest risk of hot spots, warping, and necking. Therefore, a pre-start, continuous cooling sequence with high-power cooling parameters is configured to quickly dissipate nonlinear accumulated heat. The medium-heat zone on the folded sidewall exhibits a gradient heat generation and moderate thermal deformation risk, requiring a timely start, medium-duration cooling sequence with medium-power cooling parameters to balance heat dissipation needs and plate temperature control stability. The low-heat zone on the plate end face generates minimal heat and has low deformation sensitivity, requiring a delayed start, short-duration, low-power cooling sequence to maintain a stable regional temperature. The entire control process strictly adheres to the thermodynamic characteristics of multiple stamping slides and the zoned heat generation characteristics of the asymmetric workpiece. By precisely adapting the timing misalignment and power gradation configuration, the overall thermal imbalance caused by thermal response time difference and thermal deformation misalignment is directly resolved. The results of precision optimization of pre-heating layer sensing, temperature, and springback residual compensation are fully solidified, effectively preventing problems such as equipment thermal deformation and stamping center offset from negating the effects of previous control measures. At the same time, the workpiece forming precision constraints are strictly adhered to throughout the process, ensuring that the forming dimensions and deformation states of each area of ​​the asymmetric L-shaped connecting plate meet the mass production precision requirements. From the perspective of equipment and workpiece collaborative temperature control, the closed-loop control logic of stamping forming is improved, significantly enhancing the stability of the overall process operation and the consistency of workpiece forming precision under continuous mass production conditions. This provides solid overall collaborative temperature control support for the efficient, high-precision, and high-consistency mass production of asymmetric L-shaped connecting plates.

[0073] As one implementation method, a method for adjusting the stamping temperature of an asymmetric L-shaped connecting plate is provided to manufacture the asymmetric L-shaped connecting plate, comprising: The cooling timing and cooling power parameters of the different thermal zones of the mold, stamping slide, and asymmetric L-shaped connecting plate are retrieved and matched. All cooling parameters are unified as the process execution benchmark and integrated to form an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate.

[0074] During the continuous stamping process, the system simultaneously calls up the pre-completed thermal hierarchy topology division results, the springback deviation evolution trend of each region, and the thermal parameter calibration data of the multi-stamping slider, and monitors the stamping conditions and the thermal state of the sheet metal in real time.

[0075] Based on real-time operating conditions, the cooling sequence and cooling power parameters corresponding to each stamping slide and each thermal zone of the sheet metal are dynamically fine-tuned to continuously calibrate the stamping temperature of the sheet metal and complete the batch stamping manufacturing of asymmetric L-shaped connecting plates.

[0076] In this embodiment, the cooling timing parameters are determined based on the high thermal inertia heat storage rhythm of the mold, the high-frequency reciprocating motion characteristics of the stamping slide, and the differences in the heat generation timing of each thermal level of the sheet metal. The cooling power parameters are the heat dissipation intensity index per unit time configured by combining the heat dissipation capacity of each stamping slide and the heat generation intensity of each area of ​​the sheet metal. All of the above parameters are in line with the actual thermal change law of the equipment and workpiece in the stamping system. In the entire process of continuous stamping forming operation, this adjustment method abandons the traditional fixed parameter temperature control mode, and synchronously links the thermal level topology division results generated by the previous process, the springback deviation evolution trend data of each area of ​​the asymmetric L-shaped connecting plate, and the thermal parameter calibration results of the mold, stamping slide, and sheet metal, and captures the stamping condition fluctuations, sheet metal thermal state changes, and equipment operating state changes in real time. Based on this, the cooling sequence and cooling power parameters for the mold, stamping slide, high-heat zone at the root of the sheet metal bend, mid-heat zone on the side wall of the folded edge, and low-heat zone on the end face of the sheet metal are dynamically fine-tuned. To address the slow heat accumulation and temperature rise trend caused by the mold's continuous contact with the sheet metal, the cooling duration is extended and the cooling power is slightly optimized. To address the instantaneous temperature rise and rapid heat dissipation characteristics caused by the high-frequency reciprocating motion of the stamping slide, the intermittent nodes for cooling initiation are precisely matched and the cooling power output is stabilized. To address the fluctuations in real-time heat generation intensity at each thermal level of the sheet metal, the timing of cooling intervention and the heat dissipation intensity in the corresponding areas are adjusted synchronously. Through continuous dynamic stamping temperature control, various forming hazards such as hot spots at the root of the bend, temperature difference deformation of sheet metal zones, and thermal misalignment of the mold and slide are continuously counteracted and eliminated during the stamping process.

[0077] By stabilizing the stamping temperature of each thermal level of the sheet metal within the range suitable for the forming process, and with the support of stable and controllable temperature conditions, precise temperature control effectively suppresses forming defects such as springback dispersion, warping and necking caused by abnormal temperature. At the same time, it reduces the fatigue wear of the mold caused by alternating hot and cold temperatures, ensuring the long-term stable operation of the stamping equipment. Ultimately, relying on this closed-loop stamping temperature adjustment logic, it enables the mass production of asymmetric L-shaped connecting plates with high precision, low loss and high consistency.

[0078] like Figure 3 As shown, this application also includes a manufacturing apparatus for an asymmetric L-shaped connecting plate, comprising: an acquisition unit, a processing unit, and a control unit.

[0079] The acquisition unit is used to acquire multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and to generate the original stamping dataset based on the multi-dimensional original working condition parameters.

[0080] The processing unit is used to construct a thermal stratification prediction model based on the original stamping dataset. It performs thermal stratification topology partitioning of the asymmetric L-shaped connecting plate by combining plastic deformation work and interface friction power consumption as dual thresholds to define the thermal stratification levels of the asymmetric L-shaped connecting plate. It differentiates the deformation heat weight and friction heat weight for each thermal stratification level, simultaneously acquires full-time-series thermal data of the asymmetric L-shaped connecting plate during stamping, and corrects the thermal prediction parameters using an inter-frame residual iteration method. It outputs full-domain stratified time-series thermal data and primary cooling control parameters. Based on the full-domain stratified time-series thermal data, primary cooling control parameters, and springback residual compensation algorithm, it establishes a thermal stratification... The nonlinear correlation mapping between temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate is established to statistically analyze the temperature time-series residual of the asymmetric L-shaped connecting plate in real time and complete the deviation accumulation calculation. This allows for the acquisition of the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold. Based on the global layered time-series thermodynamic data, partitioned temperature control threshold, and high-precision cooling parameters, a thermodynamic asynchronous equilibrium model is established to calibrate the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate, respectively, to obtain the thermodynamic response time difference and thermal deformation misalignment.

[0081] The control unit combines the thermal response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece. It differentiates the cooling sequence and cooling power parameters of the mold, stamping slide, and asymmetric L-shaped connecting plate to generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, thereby realizing the manufacturing of the asymmetric L-shaped connecting plate.

[0082] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for manufacturing an L-shaped connecting plate, characterized in that, include: Obtain multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and generate the original stamping dataset based on the multi-dimensional original working condition parameters. A thermal stratification prediction model is constructed based on the original stamping dataset. The thermal stratification topology of the asymmetric L-shaped connecting plate is divided by combining the dual thresholds of plastic deformation work and interface friction power consumption to define the thermal stratification of the asymmetric L-shaped connecting plate. The deformation heat weight and friction heat weight are matched differently for the thermal stratification. The full-time thermal data of the thermal stratification of the asymmetric L-shaped connecting plate during stamping are obtained simultaneously. The thermal prediction parameters are corrected by combining the inter-frame residual iteration method. The full-domain stratified time-series thermal data and primary cooling control parameters of the thermal stratification are output. Based on the global layered time-series thermodynamic data, the primary cooling control parameters, and the springback residual compensation algorithm, a nonlinear correlation mapping relationship is established between the temperature fluctuation, temperature accumulation time, and springback angle and forming curvature of the asymmetric L-shaped connecting plate at the thermodynamic level. This allows for real-time statistical analysis of the temperature time-series residual of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculation. This enables the acquisition of the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold. A thermodynamic asynchronous equilibrium model is established based on the global layered time-series thermodynamic data, the zoned temperature control threshold, and the high-precision cooling parameters. The heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate are calibrated to obtain the thermodynamic response time difference and thermal deformation misalignment. Combining the thermodynamic response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece, the cooling sequence and cooling power parameters of the mold, stamping slide, and asymmetric L-shaped connecting plate are differentially matched to generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, so as to realize the manufacturing of the asymmetric L-shaped connecting plate.

2. The manufacturing method of an L-shaped connecting plate according to claim 1, characterized in that, The multi-dimensional original working condition parameters include sheet material parameters, real-time stamping load, stamping speed, friction coefficient of the die-sheet contact surface, sheet plastic deformation work, initial temperature of the die and sheet, and equipment thermal inertia parameters; the acquisition of multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and the generation of a stamping original dataset based on the multi-dimensional original working condition parameters, includes: Lock in the complete stamping process of the asymmetric L-shaped connecting plate, screen out unstable operating periods such as equipment start-up and shutdown and idle standby, and simultaneously collect sheet material parameters, real-time stamping load, stamping speed, friction coefficient of the contact surface between the die and the sheet, plastic deformation work of the sheet, initial temperature of the die and the sheet, and thermal inertia parameters of the equipment during continuous and stable stamping operation. Among them, the material parameters of the sheet include the sheet yield strength, thermal conductivity and sheet thickness parameters, and the thermal inertia parameters of the equipment include the mold heat storage coefficient and the heat dissipation coefficient of the stamping slide. For all the original working condition parameters collected, abnormal data removal operations are carried out one by one to remove the jump data caused by the instantaneous impact of stamping. Then, the sampling noise of the acquisition equipment is removed by filtering and noise reduction. Finally, the working condition parameters of different dimensions and different numerical ranges are subjected to unified dimension normalization processing to obtain standardized effective parameter data. All valid working condition data that have completed preprocessing are classified and organized according to parameter type and stamping sequence, and uniformly packaged into a structured data format to construct the original stamping dataset for subsequent model training and calculation.

3. The method for manufacturing an L-shaped connecting plate according to claim 2, characterized in that, The thermal stratification prediction model is constructed based on the original stamping dataset, and a thermal stratification topology is performed on the asymmetric L-shaped connecting plate by combining the dual thresholds of plastic deformation work and interface friction power consumption to define the thermal stratification of the asymmetric L-shaped connecting plate. Deformation heat weights and friction heat weights are differentially matched for each thermal stratification, including: The entire original stamping dataset after regular packaging was used as training input data to complete the parameter training and construction of the thermal stratification prediction model and lock the basic operation parameters of the model. Based on the sheet metal specifications and stamping conditions of the asymmetric L-shaped connecting plate, fixed plastic deformation work threshold and interface friction power consumption threshold are pre-calibrated, and the two thresholds are used simultaneously as the criterion for classifying the thermal levels of the sheet metal. Based on the dual threshold judgment criteria, the bending area, side wall area, and outer edge area of ​​the asymmetric L-shaped connecting plate are subjected to point-by-point topological traversal detection. According to the relationship between the real-time plastic deformation work value, interface friction power consumption value and the preset threshold at each detection point, three independent thermal levels are defined: the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the bend, and the low-heat zone on the end face of the plate. The original stamping data corresponding to each thermal level is retrieved. Based on the differences in the proportion of heat generated by plastic deformation and heat generated by interfacial friction in each level, a unique deformation heat weight and friction heat weight are matched for each type of thermal level. All the differentiated weight coefficients are then embedded into the corresponding level calculation node of the thermal stratification prediction model.

4. The method for manufacturing an L-shaped connecting plate according to claim 3, characterized in that, The process involves synchronously acquiring full-time thermal data of the asymmetric L-shaped connecting plate during stamping, and correcting the thermal prediction parameters using an inter-frame residual iteration method. The output includes global layered time-series thermal data and primary cooling control parameters for the thermal hierarchy. The defined high-heat zone at the root of the bend, the medium-heat zone on the side wall of the bend, and the low-heat zone on the end face of the sheet are used as independent temperature measurement and detection units. The complete stamping stroke of a single pressing, holding, and lifting is used as an independent time sequence cycle. The real-time temperature data of each sheet thermal unit and the mold contact surface at the corresponding bonding position are collected synchronously using a millisecond-level sampling frequency. The data are sorted and integrated according to the time sequence to construct exclusive full-time thermal data for each thermal level. The full-time thermal data corresponding to each thermal level are input into the thermal stratification prediction model one by one, and the initial thermal prediction results corresponding to each region are calculated. The thermal prediction results output by the model are compared with the actual measured temperature values ​​on site under the same time series, and the inter-frame residual between adjacent stamping time series cycles is accurately calculated. The inter-frame residuals obtained in real time are used as the only iterative correction variable, and the thermal prediction parameters inside the thermal stratification prediction model are replaced and updated one by one through the inter-frame residual iterative method. The calculation continues iteratively until the overall prediction error of the thermal stratification prediction model converges to within the preset accuracy threshold. The output is full-domain stratified time-series thermal data that fits the actual heat generation state of the board. At the same time, based on the temperature amplitude and time-series change patterns of different thermal levels, the corresponding primary cooling control parameters for each region are generated.

5. A method for manufacturing an L-shaped connecting plate according to claim 4, characterized in that, Based on the global layered time-series thermodynamic data, the primary cooling control parameters, and the springback residual compensation algorithm, a nonlinear correlation mapping relationship is established between the temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate at each thermodynamic level. This allows for real-time statistical analysis of the temperature time-series residual of the asymmetric L-shaped connecting plate and completion of deviation accumulation calculation. This yields the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partitioned temperature control threshold and high-precision cooling parameters of the mold, including: The global stratified time-series thermal data and the primary cooling control parameters corresponding to each region are uniformly summarized from the iterative output of the thermal stratification prediction model and used as the basic input variables for the algorithm operation, and the rebound residual compensation algorithm is called. For the three independent thermal levels of high heat zone at the root of the bend, medium heat zone on the side wall of the bend, and low heat zone on the end face of the plate, independent calculation channels are built for each level, and nonlinear correlation mapping relationship between temperature fluctuation, temperature accumulation time and springback angle and forming curvature of the asymmetric L-shaped connecting plate is established one by one. Based on the established nonlinear correlation mapping relationship, the standard forming temperature parameters of the sheet metal and the real-time forming temperature parameters are compared sequentially to obtain the temperature time-series residuals corresponding to each thermal level. At the same time, the deformation deviation generated by multiple consecutive stamping processes is superimposed and accumulated. The system records the increase and decrease of the cumulative deviation in real time, and deduces the evolution trend of springback deviation corresponding to different thermal levels in subsequent continuous stamping processes. With the goal of offsetting the springback deviation of sheet metal delamination, the system continuously iterates and corrects the temperature control boundary values ​​of each zone of the mold, and finally outputs the zone temperature control threshold and high-precision cooling parameters of the mold that are adapted to the thermal levels of each sheet metal.

6. A method for manufacturing an L-shaped connecting plate according to claim 5, characterized in that, The process involves establishing an asynchronous thermal equilibrium model based on the global layered time-series thermal data, the zoned temperature control thresholds, and the high-precision cooling parameters. This model calibrates the heat generation response delay, heat dissipation attenuation coefficient, and critical thermal deformation threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate, respectively, to obtain the thermal response time difference and thermal deformation misalignment. This includes: The model's core input variables are summarized from the overall hierarchical time-series thermal data, mold zone temperature control thresholds, and high-precision cooling parameters. A thermodynamic asynchronous equilibrium model is built based on the heterogeneous thermal characteristics of multi-stamping sliders. The model parameter calibration program is started to perform parameter calibration for three types of structures: mold, stamping slider, and asymmetric L-shaped connecting plate. Among them, the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the overall structure of the mold are calibrated, the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the reciprocating stamping slider are calibrated, and the heat generation response delay, heat dissipation attenuation coefficient and thermal deformation critical threshold corresponding to the three thermal levels of the asymmetric L-shaped connecting plate are calibrated in layers. After completing the parameter calibration of all stamping slides and all sheet metal areas, the thermal response rate and thermal deformation degree of different stamping slides and different sheet metal thermal regions are compared laterally, and the thermal response time difference and thermal deformation misalignment between stamping slides and sheet metal partitions are quantitatively calculated.

7. A method for manufacturing an L-shaped connecting plate according to claim 6, characterized in that, The method, combining the thermodynamic response time difference, thermal deformation misalignment, and the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece, differentiates and matches the cooling sequence and cooling power parameters corresponding to the mold, stamping slide, and asymmetric L-shaped connecting plate, including: The thermal response time difference and thermal deformation misalignment obtained by quantitative calculation are used as the basis for judging the thermal imbalance of the stamping slider. Combined with the forming accuracy constraint of the asymmetric L-shaped connecting plate workpiece, the allowable forming tolerance threshold of each thermal level of the asymmetric L-shaped connecting plate is locked. Based on the thermal characteristics of the mold and the stamping slide, as well as the forming characteristics of the high-heat zone at the root of the bend, the medium-heat zone on the side wall of the fold, and the low-heat zone on the end face of the sheet, the degree of thermal imbalance and the level of deformation risk of different stamping slides and different sheet areas are distinguished. According to the risk level, the cooling start time and cooling duration are matched with the cooling sequence parameters, and the corresponding cooling power parameters are matched with the corresponding heat dissipation power to complete the differential matching of cooling parameters for all stamping slides and sheet metal sections.

8. A method for manufacturing an L-shaped connecting plate according to claim 7, characterized in that, The method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate to achieve the manufacturing of the asymmetric L-shaped connecting plate includes: The cooling timing and cooling power parameters of the different thermal zones of the mold, stamping slide, and asymmetric L-shaped connecting plate are retrieved and matched. All cooling parameters are unified as the process execution benchmark and integrated to form an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate. During the continuous stamping process, the system simultaneously calls up the pre-completed thermal hierarchy topology division results, the springback deviation evolution trend of each region, and the calibration data of the thermal parameters of multiple stamping slides to monitor the stamping conditions and changes in the thermal state of the sheet metal in real time. Based on real-time operating conditions, the cooling sequence and cooling power parameters corresponding to each stamping slide and each thermal zone of the sheet metal are dynamically fine-tuned to continuously calibrate the stamping temperature of the sheet metal and complete the batch stamping manufacturing of asymmetric L-shaped connecting plates.

9. An apparatus for manufacturing an L-shaped connecting plate for performing the manufacturing method of an L-shaped connecting plate as described in any one of claims 1 to 8, characterized in that, include: The acquisition unit is used to acquire multi-dimensional original working condition parameters during the stamping process of the asymmetric L-shaped connecting plate, and generate a stamping original dataset based on the multi-dimensional original working condition parameters. The processing unit is used to construct a thermal stratification prediction model based on the original stamping dataset, and to perform thermal stratification topology partitioning of the asymmetric L-shaped connecting plate by combining plastic deformation work and interface friction power consumption as dual thresholds to define the thermal stratification of the asymmetric L-shaped connecting plate. For the thermal stratification, deformation heat weights and friction heat weights are differentially matched, and the full-time thermal data of the thermal stratification of the asymmetric L-shaped connecting plate during stamping is acquired simultaneously. The thermal prediction parameters are corrected using an inter-frame residual iteration method, and the full-domain stratified time-series thermal data and primary cooling control parameters of the thermal stratification are output. Based on the full-domain stratified time-series thermal data, the primary cooling control parameters, and the springback residual compensation algorithm, a model is established regarding the... The nonlinear correlation mapping relationship between the temperature fluctuation, temperature accumulation time, and the springback angle and forming curvature of the asymmetric L-shaped connecting plate at the described thermal level is used to statistically analyze the temperature time-series residual of the asymmetric L-shaped connecting plate in real time and complete the deviation accumulation calculation to obtain the springback deviation evolution trend of the asymmetric L-shaped connecting plate during stamping, thereby iteratively outputting the partition temperature control threshold and high-precision cooling parameters of the mold; a thermal asynchronous equilibrium model is established based on the global layered time-series thermal data, the partition temperature control threshold, and the high-precision cooling parameters, and the heat generation response delay, heat dissipation attenuation coefficient, and thermal deformation critical threshold of the mold, stamping slide, and asymmetric L-shaped connecting plate are calibrated to obtain the thermal response time difference and thermal deformation misalignment; The control unit is used to combine the thermal response time difference and thermal deformation misalignment amount with the forming accuracy constraints of the asymmetric L-shaped connecting plate workpiece, and to differentiate and match the cooling sequence and cooling power parameters corresponding to the mold, stamping slide and asymmetric L-shaped connecting plate, and generate an adjustment method for adjusting the stamping temperature of the asymmetric L-shaped connecting plate, so as to realize the manufacturing of the asymmetric L-shaped connecting plate.

10. An electronic device, characterized in that, include: A memory, wherein the memory stores program instructions; A processor, when executing the program instructions stored in the memory, implements the manufacturing method of the L-shaped connecting plate according to any one of claims 1 to 8.

Images