Hot stamping die cooling water channel optimization design method based on numerical simulation and machine learning

Abstract

The invention relates to the technical field of hot stamping dies, in particular to a hot stamping die cooling water channel optimization design method based on numerical simulation and machine learning, and aims to solve the technical problem of unbalanced cooling of a hot stamping part with a complex structural surface. According to the method, a closed loop from data collection, simulation analysis and machine learning modeling to decision optimization is realized, and the aim is to improve the thermal management effect of the die, reduce the production risk and improve the production efficiency through accurate design optimization of the hot stamping die cooling system. According to the method, mechanical simulation, temperature field analysis, machine learning prediction and multi-objective optimization are combined, and a scientific and efficient optimization scheme is provided for cooling water channel design.

Description

Technical Field

[0001] The present invention relates to the technical field of hot stamping dies, and in particular to a method for optimizing the design of cooling channels of hot stamping dies based on numerical simulation and machine learning. Background Art

[0002] Lightweighting of automobiles has become a major trend in the development of the global automobile manufacturing industry. In order to achieve lightweight design while ensuring the structural strength of the automobile body, high-strength steel is widely used in the manufacturing process of automobile parts. Given the relatively poor forming performance of high-strength steel, traditional stamping processes can no longer meet the production needs of high-strength automotive structural parts. Therefore, in current industrial production, hot stamping technology has become the main means of production.

[0003] As a high-strength steel, 22MnB5 manganese-boron steel has been widely used in industrial production. Before sheet metal is stamped, its initial phase structure is mainly composed of ferrite and pearlite. The mechanical strength of the part is closely related to the martensite content, and the formation of martensite is mainly achieved during the pressure-holding quenching process in the mold. By accurately controlling the cooling rate during the quenching process, the mechanical properties of the parts after hot stamping can be effectively adjusted. When the cooling rate exceeds the critical value, a complete martensitic microstructure can be formed, and the comprehensive mechanical properties of the component can be significantly enhanced. In order to achieve rapid cooling of hot-formed parts and promote the formation of high-strength martensitic structure, cooling water channels need to be arranged inside the mold. The rapid cooling of the mold and hot stamping parts to the target temperature is accelerated by the circulation of cooling water.

[0004] The geometric structure and layout of the cooling water channel have a significant impact on the uniformity of the cooling temperature and rate during the martensitic phase transformation. In addition, the design of the cooling water channel also has an important impact on the structural strength and service life of the mold. Therefore, for hot stamping dies, it is particularly important to implement a reasonable cooling water channel design. At present, the design of cooling water channels mainly depends on the practical experience of engineers. When dealing with stamping parts with complex shapes, the uneven distribution of temperature fields has become a significant problem, which makes it difficult to effectively layout the water channel structure in specific areas, resulting in poor cooling effects in certain areas. Therefore, it is urgent to develop a hot stamping die cooling water channel design technology suitable for cooling parts with complex structural surfaces. The main purpose of this technology is to reduce the time cost in the mold design process and ensure that the designed cooling water channel structure can achieve a more balanced cooling effect to ensure the overall structural strength of the stamping parts. Summary of the invention

[0005] The purpose of the present invention is to propose a method for optimizing the design of cooling channels for hot stamping dies based on numerical simulation and machine learning. This technical solution can solve the technical problem of uneven cooling of hot stamping parts with complex structural surfaces.

[0006] To achieve the above object, the present invention provides a method for optimizing the design of cooling channels of hot stamping dies based on numerical simulation and machine learning, comprising: Obtain the structural parameters and cooling channel layout information of the hot stamping die and build the initial 3D model of the die; Implement mechanical simulation and temperature field simulation of hot stamping process in simulation software to obtain contact pressure distribution data and temperature distribution data of die contact surface; Map the contact pressure to the solid model surface of the mold without water channel, perform static simulation analysis, determine the stress distribution characteristics of the mold, and mark the high-risk areas and general risk areas; Simplify the mechanical model, calculate the yield risk of the main structural parameters of the cooling water channel under different contact pressures, determine the value range of the structural parameters in each area, and establish a structural parameter data set; According to the mold temperature distribution data and the divided areas, the water channels in each area are arranged in combination with the water channel center distance to generate the cooling water channel layout and density data set; The structural parameter data set is fused with the cooling water channel layout and density data set to generate a fused data set; the optimal hypercube sampling strategy is used to sample the fused data set, a simulation parameter combination experiment is designed, and a continuous production simulation experiment is conducted; Extract simulation results, obtain the mold surface unit data at the end of the stable period, and calculate the maximum temperature, average temperature and temperature uniformity of the mold surface; The selected simulation parameter combination and the maximum temperature, average temperature and temperature uniformity of the mold surface are used as training sets and input into the pre-built machine learning model to train the machine learning prediction model; The fused data set is input into the trained machine learning model, and a predictive analysis is performed on each set of cooling water channel structure design parameter combinations to obtain the key evaluation indicators corresponding to each set of design parameter combinations and determine the optimal parameter combination for comprehensive evaluation.

[0007] Beneficial effects of the basic solution: The present invention collects the structural parameters of the original hot stamping die and the layout information of the cooling water channel, pre-processes the acquired data and creates a standardized data set and three-dimensional modeling, thereby ensuring the accuracy and completeness of the data.

[0008] The hot stamping process is simulated using numerical simulation technology to obtain the contact pressure distribution and temperature field distribution information of the die contact interface, thereby revealing potential thermal stress concentration phenomena and material fatigue problems, and identifying high-risk areas in the die. The mechanical models of high-risk areas and general risk areas are simplified, and the yield risk of the main structural parameters of the cooling water channel is calculated to ensure the safety and economy of the design. By determining the value range of the structural parameters, the blindness and arbitrariness in the design process are avoided, and the rationality and reliability of the design are improved.

[0009] The layout and density of cooling channels are designed in combination with temperature distribution data to generate a fusion data set, which contains all relevant parameters and simulation results, providing data support for subsequent simulation parameter combination design experiments. The optimal hypercube sampling strategy is used for simulation parameter combination design experiments to systematically explore the impact of different parameter combinations on mold performance, avoiding the limitations of traditional design methods. The simulation result data of the mold part surface unit in the continuous generation simulation experiment is extracted to evaluate the rationality of the cooling channel structure design parameters and ensure the reliability and efficiency of the mold in actual production.

[0010] By using machine learning models to predict key evaluation indicators, the advantages and disadvantages of different design parameter combinations can be quickly evaluated, improving the efficiency and accuracy of the design. Thus, the best cooling water channel optimization scheme can be formulated. Therefore, the present invention comprehensively considers the structural parameters, arrangement mode, and heat dissipation effect of the cooling water channel through numerical simulation and machine learning methods for comprehensive optimization, avoiding the limitations of traditional design methods, improving the cooling efficiency of the hot stamping die and the rationality of the cooling water channel layout, and extending the service life of the die.

[0011] As an implementable preferred solution, the structural parameters of the hot stamping die and the layout information of the cooling water channel, including the structural dimensions of the die, the length, diameter, and direction of the water channel, are obtained through design drawings, CAD models, or actual measurements; and a model is created through 3D modeling software.

[0012] As an implementable preferred solution, mechanical simulation and temperature field simulation of the hot stamping process are implemented in the simulation software to obtain the contact pressure distribution data and temperature distribution data of the die contact surface, including the following contents: Define contact pairs: define the contact pairs between the mold and the workpiece in the simulation model, and set related contact properties, including friction coefficient, contact stiffness, and heat transfer coefficient; According to the actual conditions of the hot stamping process, set the simulation parameters, including stamping speed, stamping force, and die preheating temperature; The simulation is performed using thermal-structural coupling analysis and an explicit dynamics solver; during the simulation, the software calculates the contact pressure and temperature distribution between the mold and the workpiece.

[0013] As an implementable preferred solution, the stress distribution characteristics of the water channel-free solid model during the stamping process are determined, and the areas with stress concentration are regarded as high-risk areas, while other areas are general risk areas.

[0014] As an implementable preferred solution, the main structural parameters of the cooling water channel include center distance, diameter, and distance. By calculating the main structural parameters of the cooling water channel under different contact pressures, the value range of the structural parameters of each region is determined, and a structural parameter data set is established.

[0015] As an implementable optimal solution, based on the mold temperature distribution data and the divided areas, the water channels in each area are arranged in combination with the water channel center distance to generate a cooling water channel layout and density data set, including the following: The mold temperature data is converted into a cloud map to show the temperature distribution of the mold; the cooling water channel layout and density data set includes the location, length and diameter of the water channel.

[0016] As an implementable optimal solution, the optimal hypercube sampling strategy is used to sample and select the fusion data set, design simulation parameter combination experiments, and conduct continuous production simulation experiments; including the following contents: According to the experimental requirements, the number of samples is determined to be N groups to ensure that the dimension matches the number of variables; The optimal hypercube sampling strategy is used to sample and select the fusion data set to generate 50 sets of simulation parameter combinations; A continuous hot stamping production simulation experiment was carried out in the simulation software ANSYS, and the simulation results of the temperature distribution of the mold during the continuous production process were calculated.

[0017] As an implementable preferred solution, Calculate the maximum temperature, average temperature and temperature uniformity of the mold surface, including the following: Obtain the maximum temperature T of the mold surface by direct reading max ; Average temperature T of mold part surface ave The formula is as follows:

[0018] Uniformity of mold part surface T Dev Characterized by temperature standard deviation, the formula is as follows:

[0019] In the formula, is the temperature of the mold surface element i, is the volume of the mold surface element i, is the total number of mold surface elements, and the calculation results and simulation parameters are corresponded and recorded.

[0020] As an implementable preferred solution, the machine learning model uses an artificial neural network for supervised learning. In supervised learning, the neural network learns through the relationship between input data and corresponding target outputs, and the ultimate goal is to enable the network to make accurate predictions for new input data.

[0021] As an implementable optimization scheme, the optimal parameter combination for comprehensive evaluation is determined, including the following: The good-bad solution distance method is used to rank the evaluation index data set, compare the advantages and disadvantages of different parameter combinations, determine the ideal optimal solution and the worst solution for each indicator, calculate the relative distance between each combination and these solutions, and derive a comprehensive evaluation score based on this; based on the score ranking, the optimal parameter combination is selected and the best cooling water channel optimization plan is formulated. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A logical schematic diagram of a hot stamping die cooling channel optimization design method based on numerical simulation and machine learning.

[0023] Figure 2 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the technical solution and advantages of the present application clearer, the technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings. It will be understood that the specific embodiments described herein are only partial embodiments of the present invention, which are only used to explain the present application, rather than to limit the present application. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated, and they can be combined with each other to achieve better technical effects. The same reference numerals appearing in the drawings of the following embodiments represent the same features or components, which can be applied to different embodiments.

[0025] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of the present invention should have the common meanings understood by those skilled in the art in the art to which the present invention belongs.

[0026] Reference numerals: electronic device 500 includes processor 501 , communication interface 502 , memory 503 , and bus 504 .

[0027] The present invention is further described in detail below in conjunction with the accompanying drawings: Reference Figure 1 , a hot stamping die cooling channel optimization design method based on numerical simulation and machine learning, including: Step S1, obtain the structural parameters and cooling water channel layout information of the hot stamping die, including but not limited to the structural dimensions of the die (such as length, width, height), the length, diameter, direction of the water channel and other detailed information to ensure the integrity and accuracy of all initial data. Obtained through design drawings, CAD models or actual measurements. Perform preprocessing steps on the collected data, including data cleaning, removal of redundant information, unified format, etc., to generate a standardized data set to ensure data consistency and accuracy. On this basis, use professional 3D modeling software (such as UG, SolidWorks, etc.) to build a 3D model of the initial die according to the preprocessed standard digital data set.

[0028] Step S2, implementing mechanical simulation and temperature field simulation of the hot stamping process in the simulation software to obtain contact pressure distribution data and temperature distribution data of the die contact surface, including: Step S21, defining contact pairs, defining contact pairs between the mold and the workpiece in the simulation model, and setting related contact properties, such as friction coefficient, contact stiffness, thermal conductivity, etc.

[0029] Step S22, setting simulation parameters such as stamping speed, stamping force, mold preheating temperature, etc. according to actual conditions of the hot stamping process.

[0030] Step S23, using thermal-structural coupling analysis and explicit dynamics solver to perform simulation. During the simulation, the software will calculate the contact pressure distribution and temperature distribution between the mold and the workpiece.

[0031] Step S24: After the simulation is finished, the post-processing phase is entered, where contact pressure and temperature are selected as result variables, and corresponding distribution diagrams are generated.

[0032] Step S25, exporting the simulation result data for subsequent analysis.

[0033] In this embodiment, the above simulation result data is obtained by performing thermal-structural coupled analysis (Thermal-Structural Coupled AnalySiS) and explicit dynamics solver (Explicit DynamicS) using ANSYS.

[0034] In step S3, the contact pressure is mapped to the surface of the solid model of the water channel-free mold, and a static simulation analysis is performed to determine the stress distribution characteristics of the water channel-free solid model during the stamping process. The post-processing functions of the finite element software (such as contour maps and stress cloud maps) can be used to intuitively identify stress concentration areas. The stress concentration areas are considered high-risk and marked as area 1; other areas are general risks and marked as area 2.

[0035] Step S4, simplify the mechanical model of area 1 and area 2. The simplified model should be able to reflect the main stress conditions of the mold during the stamping process. Calculate the yield risk of the main structural parameters of the cooling water channel (center distance P, diameter D, distance H) under different contact pressures. The yield risk refers to the possibility of yielding (i.e., plastic deformation) of the water channel structure under stress. By calculating the yield risk, determine the value range of the structural parameters of each region, and establish a structural parameter data set.

[0036] Step S5, convert the mold temperature data obtained in step S2 into a cloud map to intuitively display the temperature distribution of the mold; and divide the area according to the water channel center distance P determined in step S4, reasonably layout the water channels in each area, and the water channel layout should ensure that the cooling water can flow evenly through the mold surface to reduce the temperature of the mold and improve its temperature distribution uniformity; generate a cooling water channel layout and density data set, including information such as the location, length and diameter of the water channel.

[0037] Step S6, performing data fusion and simulation experiments, including: Step S61, fusing the structural parameter data set obtained in step S4 with the cooling water channel layout and density data set obtained in step S5 to generate a fused data set, which includes information such as structural parameters, water channel layout and density of the mold.

[0038] Step S62, adopt the optimal hypercube sampling strategy to sample and select the fusion data set, so that the selected sample data can effectively cover the entire parameter space, thereby improving the accuracy and reliability of the simulation experiment. Design a simulation parameter combination experiment, including: Step S62-1, according to the experimental requirements, determine the number of samples to be N groups. In this embodiment, the number of samples is set to 50 groups to ensure that the dimension matches the number of variables.

[0039] Step S62-2, using the optimal hypercube sampling strategy to sample and select the fused data set, generating 50 sets of simulation parameter combinations to comprehensively reflect the cooling performance of the mold.

[0040] Step S62-3, performing a continuous hot stamping production simulation experiment in the simulation software ANSYS, and calculating the simulation results of the temperature distribution of the mold during the continuous production process.

[0041] In the embodiment of the present invention, steps S3 to S5 mainly involve dividing the regions according to the stress distribution and temperature distribution, and marking the specific regions. Based on comprehensive consideration of factors such as the structural strength, cooling efficiency, and temperature distribution uniformity of the hot stamping die, parameter design and reasonable pipeline layout are performed for different regions to generate corresponding cooling water channel layout and density data sets.

[0042] Step S7: Extract the simulation results of step S6 and obtain the mold surface unit data at the end of the stable period. Calculate the maximum temperature T of the mold surface max , average temperature T ave And temperature uniformity T Dev , and record it as the basis for evaluating the rationality of the cooling water channel design.

[0043] Maximum temperature of mold surface T max Can directly read; the average temperature T of the mold part surface ave The formula is as follows:

[0044] Uniformity of mold part surface T Dev Characterized by temperature standard deviation, the formula is as follows:

[0045] In the formula, is the temperature of the mold surface element i (°C), is the volume of the mold surface unit i (㎥), is the total number of mold surface elements, and the calculation results and simulation parameters are recorded accordingly. Temperature uniformity It reflects the degree of discreteness of the mold surface temperature. The smaller the value, the more uniform the temperature distribution.

[0046] Step S7 is to extract the simulation results in the previous step and obtain the mold part surface data at the end of the stable period. The maximum temperature T of the mold surface max , the average temperature T of the mold part surface ave And the uniformity of the mold part surface temperature T Dev These three parameters are important references for evaluating whether the design is reasonable.

[0047] In step S8, the 50 sets of simulation parameter combinations selected in step S6 and the calculation result data generated in step S7 are used as training sets and input into a pre-built machine learning model, in which artificial neural network (ANN) supervised learning is used.

[0048] In supervised learning, a neural network learns the relationship between input data and the corresponding target output (label), with the ultimate goal of enabling the network to make accurate predictions for new input data.

[0049] The model aims to predict the key evaluation indicator (T max 、T ave 、T Dev ). Through this process, the training of the machine learning prediction model is completed.

[0050] Step S9, input the fused data set of step S6 into the machine learning model that has been trained and has efficient prediction capability in step S8. The data set covers all combinations of optimization parameters. The powerful calculation and analysis functions of the machine learning model are used to predict and analyze each group of cooling water channel structure design parameter combinations, and obtain the key evaluation indicators corresponding to each group of design parameter combinations, including the maximum temperature (T max ), average temperature (T ave ) and temperature deviation (T Dev ). The forecast results are collated and summarized to form a comprehensive key evaluation indicator data set to provide data support for further analysis and optimization decisions.

[0051] Step S10, use the TOPSIS method to sort the evaluation index data set in step S9. This method can systematically compare the advantages and disadvantages of various parameter combinations. Determine the ideal optimal solution and the worst solution for each evaluation index, and then calculate the relative distance between each parameter combination and the ideal optimal solution and the worst solution. By comparing these distances, the comprehensive evaluation score of each parameter combination can be obtained. Sort by score to determine the parameter combination with the best comprehensive evaluation. Perform final processing on the optimal parameter combination, and then formulate the best cooling water channel optimization plan.

[0052] The disclosed embodiment also provides a hot stamping die cooling water channel optimization design system based on numerical simulation and machine learning. The system uses a hot stamping die cooling water channel optimization design method based on numerical simulation and machine learning.

[0053] The embodiment of the present disclosure also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, all steps of the above-mentioned method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning can be implemented.

[0054] A person of ordinary skill in the art can understand that the implementation of all or part of the process in a method for optimizing the design of cooling channels for hot stamping dies based on numerical simulation and machine learning can be completed by instructing related hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include a process of various embodiments of a method for optimizing the design of cooling channels for hot stamping dies based on numerical simulation and machine learning. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0055] The embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the above-mentioned method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning are implemented. In the embodiment of the present application, the processor is the control center of the computer system, which can be a processor of a physical machine or a processor of a virtual machine.

[0056] Reference Figure 2 , the electronic device 500 includes: at least one processor 501, at least one communication interface 502, at least one memory 503 and at least one bus 504. The bus 504 is used to realize the connection and communication between these components, the communication interface 502 is used to communicate signaling or data with other node devices, and the memory 503 stores machine-readable instructions executable by the processor 501. When the electronic device 500 is running, the processor 501 communicates with the memory 503 through the bus 504, and when the machine-readable instructions are called by the processor 501, the steps of the above-mentioned method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning are executed.

[0057] The above contents are only embodiments of the present invention. The common sense such as the known specific structures and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or the priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for the ordinary technicians in the relevant field to implement this application. It should be pointed out that for the technicians in this field, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to explain the content of the claims.

Claims

1. A method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning, characterized in that: include: Obtain the structural parameters and cooling channel layout information of the hot stamping die and build the initial 3D model of the die; Implement mechanical simulation and temperature field simulation of hot stamping process in simulation software to obtain contact pressure distribution data and temperature distribution data of die contact surface; Map the contact pressure to the solid model surface of the mold without water channel, perform static simulation analysis, determine the stress distribution characteristics of the mold, and mark the high-risk areas and general risk areas; Simplify the mechanical model, calculate the yield risk of the main structural parameters of the cooling water channel under different contact pressures, determine the value range of the structural parameters in each area, and establish a structural parameter data set; According to the mold temperature distribution data and the divided areas, the water channels in each area are arranged in combination with the water channel center distance to generate the cooling water channel layout and density data set; The structural parameter data set is fused with the cooling water channel layout and density data set to generate a fused data set; the optimal hypercube sampling strategy is used to sample the fused data set, a simulation parameter combination experiment is designed, and a continuous production simulation experiment is conducted; Extract simulation results, obtain the mold surface unit data at the end of the stable period, and calculate the maximum temperature, average temperature and temperature uniformity of the mold surface; The selected simulation parameter combination and the maximum temperature, average temperature and temperature uniformity of the mold surface are used as training sets and input into the pre-built machine learning model to train the machine learning prediction model; The fused data set is input into the trained machine learning model, and a predictive analysis is performed on each set of cooling water channel structure design parameter combinations to obtain the key evaluation indicators corresponding to each set of design parameter combinations and determine the optimal parameter combination for comprehensive evaluation.

2. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: The structural parameters of the hot stamping die and the layout information of the cooling water channel are obtained through design drawings, CAD models or actual measurements, including the structural dimensions of the die, the length, diameter and direction of the water channel; and the model is created through 3D modeling software.

3. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: Implement mechanical simulation and temperature field simulation of hot stamping process in simulation software to obtain contact pressure distribution data and temperature distribution data of die contact surface, including the following contents: Define contact pairs: define the contact pairs between the mold and the workpiece in the simulation model, and set related contact properties, including friction coefficient, contact stiffness, and heat transfer coefficient; According to the actual conditions of the hot stamping process, set the simulation parameters, including stamping speed, stamping force, and die preheating temperature; Simulate using thermal-structural coupling analysis and explicit dynamics solvers; During the simulation, the software calculates the contact pressure distribution and temperature distribution between the mold and the workpiece.

4. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: The stress distribution characteristics of the solid model without water channel during the stamping process are determined. The areas with stress concentration are regarded as high-risk areas, and other areas are general risk areas.

5. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: The main structural parameters of the cooling water channel include center distance, diameter, and distance. By calculating the main structural parameters of the cooling water channel under different contact pressures, the value range of the structural parameters in each area is determined, and a structural parameter data set is established.

6. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: According to the mold temperature distribution data and the divided areas, the water channels in each area are arranged in combination with the water channel center distance, and the cooling water channel layout and density data set is generated, including the following: The mold temperature data is converted into a cloud map to show the temperature distribution of the mold; the cooling water channel layout and density data set includes the location, length and diameter of the water channel.

7. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: The optimal hypercube sampling strategy is used to sample and select the fusion data set, design simulation parameter combination experiments, and conduct continuous production simulation experiments; including the following contents: According to the experimental requirements, the number of samples is determined to be N groups to ensure that the dimension matches the number of variables; The optimal hypercube sampling strategy is used to sample and select the fusion data set to generate 50 sets of simulation parameter combinations; A continuous hot stamping production simulation experiment was carried out in the simulation software ANSYS, and the simulation results of the temperature distribution of the mold during the continuous production process were calculated.

8. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: Calculate the maximum temperature, average temperature and temperature uniformity of the mold surface, including the following: Obtain the maximum temperature T of the mold surface by direct reading max ; Average temperature T of mold part surface ave The formula is as follows: Uniformity of mold part surface T Dev Characterized by temperature standard deviation, the formula is as follows: In the formula, is the temperature of the mold surface element i, is the volume of the mold surface element i, is the total number of mold surface elements, and the calculation results and simulation parameters are corresponded and recorded.

9. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: The machine learning model uses an artificial neural network for supervised learning. In supervised learning, the neural network learns through the relationship between input data and corresponding target outputs, with the ultimate goal of enabling the network to make accurate predictions for new input data.

10. The method for optimizing the cooling channel of a hot stamping die based on numerical simulation and machine learning according to claim 1, characterized in that: Determine the optimal parameter combination for comprehensive evaluation, including the following: The good-bad solution distance method is used to rank the evaluation index data set, compare the advantages and disadvantages of different parameter combinations, determine the ideal optimal solution and the worst solution for each indicator, calculate the relative distance between each combination and these solutions, and derive a comprehensive evaluation score based on this; based on the score ranking, the optimal parameter combination is selected and the best cooling water channel optimization plan is formulated.

Images