Automobile part multidirectional forging process optimization method and system based on finite element analysis

Abstract

The invention discloses an automobile part multidirectional forging process optimization method and system based on finite element analysis, and relates to the technical field of forging, and the method comprises the following steps: based on punch distribution of multidirectional forging, carrying out grid division on an automobile part, and establishing a finite element grid space; forging process parameter simulation is conducted based on the finite element grid space, and all finite element forging parameters are obtained; according to the position distribution of the finite element grid space, positioning comparison is conducted on all the finite element forging parameters, and the grid gradient is obtained; and clustering is conducted according to the grid gradient distribution, a deviation distribution space is recognized, a corresponding punch is matched according to the deviation value of the deviation distribution space to conduct forging parameter regulation and control optimization, and an optimized regulation and control strategy is obtained so that the compensation deviation value can meet the gradient distribution constraint of the finite element grid space. The technical problems that in the multi-direction forging process in the prior art, the forging forming precision is low, and the forging deviation is difficult to effectively control are solved, and the technical effects of improving the forging precision and achieving automatic compensation of the forming deviation are achieved.

Description

Technical Field

[0001] This invention relates to the field of forging technology, and more specifically to a method and system for optimizing the multi-directional forging process of automotive parts based on finite element analysis. Background Technology

[0002] Multi-directional forging is widely used in automotive parts forming. However, due to the asynchronous action of punches in different directions and the complex load distribution, uneven material flow and stress concentration are prone to occur during deformation, resulting in large dimensional deviations in the formed area and making it difficult to guarantee the geometric accuracy and microstructure uniformity of the parts. Traditional forging parameters mainly rely on empirical settings and lack precise analysis and control mechanisms for material response under multi-directional loading, making it difficult to achieve dynamic compensation and high-precision control of forging deviations. Summary of the Invention

[0003] This application provides a method and system for optimizing the multi-directional forging process of automotive parts based on finite element analysis, which is used to address the technical problems of low forging accuracy and difficulty in effectively controlling forging deviations in the existing multi-directional forging process.

[0004] In view of the above problems, this application provides a method and system for optimizing the multi-directional forging process of automotive parts based on finite element analysis.

[0005] The first aspect of this application provides a method for optimizing the multi-directional forging process of automotive parts based on finite element analysis, the method comprising: Based on the punch distribution in multi-directional forging, automotive parts are meshed to establish a finite element mesh space. Forging process parameters are simulated using this finite element mesh space to obtain various finite element forging parameters. According to the positional distribution of the finite element mesh space, the forging parameters are compared and matched to obtain the mesh gradient. Clustering is performed based on the mesh gradient distribution to identify the deviation distribution space. Forging parameters are then adjusted and optimized by matching the corresponding punch according to the deviation amount in the deviation distribution space, resulting in an optimized control strategy that ensures the compensation deviation meets the gradient distribution constraints of the finite element mesh space.

[0006] A second aspect of this application provides a multi-directional forging process optimization system for automotive parts based on finite element analysis, the system comprising: The system comprises the following modules: a mesh generation module for dividing automotive parts into meshes based on the punch distribution in multi-directional forging, establishing a finite element mesh space; a simulation module for simulating forging process parameters based on the finite element mesh space, obtaining various finite element forging parameters; a positioning and comparison module for comparing the positioning of each finite element forging parameter according to the positional distribution in the finite element mesh space, obtaining the mesh gradient; and a control and optimization module for clustering based on the mesh gradient distribution, identifying the deviation distribution space, and matching the corresponding punch according to the deviation amount in the deviation distribution space to perform forging parameter control and optimization, obtaining an optimized control strategy to ensure that the compensation deviation amount meets the gradient distribution constraints of the finite element mesh space.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application, based on the punch distribution in multi-directional forging, meshes automotive parts and establishes a finite element mesh space. Forging process parameters are simulated based on this finite element mesh space to obtain various finite element forging parameters. According to the positional distribution of the finite element mesh space, the forging parameters are compared and matched to obtain the mesh gradient. Clustering is performed based on the mesh gradient distribution to identify the deviation distribution space. Forging parameters are then adjusted and optimized by matching the corresponding punch according to the deviation amount in the deviation distribution space, resulting in an optimized control strategy that ensures the compensation deviation meets the gradient distribution constraints of the finite element mesh space. This invention solves the technical problems of low forging accuracy and difficulty in effectively controlling forging deviations in the multi-directional forging process in existing technologies. By introducing finite element mesh gradient analysis and deviation distribution identification for parameter adjustment and optimization, it achieves the technical effects of improving forging accuracy and realizing automatic compensation for forming deviations. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic diagram of the process optimization method for multi-directional forging of automotive parts based on finite element analysis provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of a multi-directional forging process optimization system for automotive parts based on finite element analysis, provided in an embodiment of this application.

[0010] Figure labeling: Mesh generation module 11, simulation module 12, positioning and comparison module 13, control and optimization module 14. Detailed Implementation

[0011] This application provides a method and system for optimizing the multi-directional forging process of automotive parts based on finite element analysis. It addresses the technical problems of low forging accuracy and difficulty in effectively controlling forging deviations in the existing multi-directional forging process. By introducing finite element mesh gradient analysis and deviation distribution identification, parameter adjustment and optimization are performed to improve forging accuracy and achieve automatic compensation for forming deviations.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown, this application provides a method for optimizing the multi-directional forging process of automotive parts based on finite element analysis, the method comprising: Step S100: Based on the punch distribution of multi-directional forging, mesh the automotive parts and establish a finite element mesh space.

[0015] In this embodiment, automotive parts are meshed based on the punch distribution in multi-directional forging. The punch distribution in multi-directional forging refers to the arrangement of multiple punches in a forging machine that simultaneously or sequentially apply pressure to the billet according to different spatial directions and positions. In this process, the positional relationship of the punches and their forging response range are first analyzed to determine independent regions and overlapping connection regions. Then, the interaction strength of the overlapping connection regions is analyzed to obtain the punch interaction strength distribution. Subsequently, based on the independent regions, overlapping connection regions, and the punch interaction strength distribution, meshing is performed to establish a finite element mesh space.

[0016] Furthermore, the method provided in the application embodiments, which involves meshing automotive parts based on the punch distribution in multi-directional forging and establishing a finite element mesh space, also includes: Based on the positional relationship of the punches and the forging response range of the punches, independent regions and overlapping connection regions are determined; the interaction intensity analysis of the overlapping connection regions is performed to obtain the punch interaction intensity distribution of the overlapping connection regions; the independent regions, overlapping connection regions and their punch interaction intensity distributions are used to divide the mesh and establish the finite element mesh space.

[0017] In this embodiment, independent and overlapping regions are first determined according to the distribution and positional relationships of the punches and their forging response ranges. In this process, a blank model is first established based on the three-dimensional geometric data of the automotive part. This blank model is obtained by geometrically simplifying and surface-smoothing the actual part design model and is used to represent the initial shape of the material to be forged in virtual space. Subsequently, based on the installation position and loading direction of each punch in the punch distribution, a spatial correspondence between the punch and the blank is established. The forging response range of the punch refers to the spatial area in which the punch applies effective pressure to the blank and causes plastic deformation during the forging process. This range is determined by the punch end face size, stroke length, loading angle, and blank contact characteristics, and is predetermined in the process design stage through experimental measurement or numerical simulation. After clarifying the forging response range of the punches, the effective range of each punch is projected onto the blank model, and the coverage relationship between the effective areas of different punches is identified through spatial superposition analysis. If a region is within the response range of only a single punch, then that region is classified as an independent region; if a region is within the response range of two or more punches, then it is classified as an overlapping region.

[0018] Next, the interaction intensity of the overlapping and connecting regions is analyzed. This process begins by analyzing the direction of the impact force of the punch in the overlapping and connecting regions to determine the interaction relationship. Then, based on the interaction distance between the punch and each point within the overlapping and connecting regions, the intensity distribution is analyzed to obtain the intensity distribution field. Finally, the interaction relationship is used to correct the intensity distribution, resulting in the punch interaction intensity distribution of the overlapping and connecting regions.

[0019] Finally, mesh generation is performed based on the independent regions, overlapping regions, and the distribution of punch interaction intensity. In this process, a uniform mesh generation method is used for independent regions, with the mesh density determined according to the expected deformation degree of the punch. For overlapping regions, a mesh density ratio is established based on the punch interaction intensity; regions with higher interaction intensity use a higher density mesh generation, and regions with lower interaction intensity use a lower density mesh generation, thus forming a finite element mesh space that matches the punch load distribution characteristics.

[0020] Furthermore, in the method provided in the application embodiments, the interaction intensity analysis of the overlapping connection region to obtain the punch interaction intensity distribution of the overlapping connection region further includes: The direction of the impact force of the punch is analyzed in the overlapping connection area to determine the interaction relationship; the intensity distribution is analyzed according to the action distance between the punch and the overlapping connection area to obtain the intensity distribution; the intensity distribution is corrected using the interaction relationship to obtain the punch interaction intensity distribution in the overlapping connection area, wherein if the interaction relationship is complementary, the intensity distribution is superimposed; if the interaction relationship is interfering, the intensity distribution is reduced.

[0021] In this embodiment, the direction of the punch action is first analyzed in the overlapping connection area. During this process, the loading direction of each punch in three-dimensional space is analyzed within the overlapping connection area, using the overlapping connection area as the calculation range. By extracting the loading direction vector of each punch, the spatial angle between these vectors within the overlapping connection area is calculated one by one to determine the interaction relationship of each punch within this area. When the angle between the loading directions of two punches is less than 30 degrees, it indicates that their loading directions in this area are consistent or close, indicating a complementary relationship; when the angle is greater than 150 degrees, it indicates that their directions are opposite, indicating an interference relationship; when the angle is between 30 and 150 degrees, further judgment is made based on the local geometric features of the billet within the overlapping connection area. If the local surface normal direction is consistent with the loading directions of both punches or the angle is small, it indicates that the force action paths of the two punches at this position are the same, indicating a complementary relationship; if the local surface normal direction is consistent with the direction of one punch and opposite to the direction of the other punch, it indicates that the loading directions are opposing, indicating an interference relationship. Through the above point-by-point calculation, the interaction relationship is determined.

[0022] Subsequently, intensity distribution analysis was performed based on the interaction distance between the punch and the overlapping connection area. In this process, the overlapping connection area was first discretized into a regular three-dimensional mesh with a mesh step size determined, such as 1mm to 2mm, to ensure geometric details were recognizable. Then, the shortest distance from each mesh node to the loading end face of each punch was measured, and the nominal loading pressure, contact area, loading depth, and loading direction of that punch were recorded. Next, a distance-intensity comparison table obtained through unidirectional forging calibration or finite element single-step loading calibration was used to establish several distance intervals and corresponding intensity values; for example, 0 to 10mm corresponds to the high-intensity interval, 10 to 30mm to the medium-intensity interval, and 30 to 60mm to the low-intensity interval. Then, by looking up the table point by point or by linear interpolation within the interval, the shortest distance between each node and each punch was mapped to the intensity value of that node. Spatial interpolation and boundary smoothing were performed on the nodal intensity obtained for the same punch across the entire domain to eliminate discrete jumps and ensure intensity continuity with non-overlapping areas. Finally, the nodal intensity fields of all punches were summarized without superposition or mutual cancellation, forming only a multi-channel spatial intensity record. For example, if the distance from node A to the upper punch is 10mm and the distance to the left punch is 15mm, the intensity values ​​of the two channels are read from the reference table and recorded on node A. At this point, node A only records the intensity values ​​of two channels without combining them. After completing the above process, the intensity distribution is obtained.

[0023] Finally, the intensity distribution is corrected using the interaction relationship. In this process, the grid nodes within the overlapping area are first used as the calculation unit. For each pair of punches, a superposition coefficient or a reduction coefficient is set according to its included angle range. Specifically, a full-scale superposition coefficient is used when the included angle is less than 30 degrees; a decreasing superposition coefficient is used when the included angle is between 30 and 90 degrees; an increasing reduction coefficient is used when the included angle is between 90 and 150 degrees; and a full-scale reduction coefficient is used when the included angle is greater than 150 degrees. This coefficient is fixed through simulation analysis and experimental determination during the process calibration stage and then used for subsequent calculations.

[0024] Subsequently, at the node level, the corresponding coefficient parameters are read according to the previously determined interaction relationships. When the punch relationship at the node is complementary, the superposition coefficient is invoked; when it is interfering, the reduction coefficient is invoked. The intensity values ​​of each punch channel are then weighted and combined. Specifically, channels with complementary relationships are weighted and summed using the superposition coefficient to reflect the synergistic effect of unidirectional loading, while channels with interfering relationships are weighted and subtracted using the reduction coefficient to reflect the weakening effect of reverse loading. During calculation, all punch pairs are traversed sequentially, gradually completing node-level intensity correction to ensure the logical continuity of superposition and cancellation. After completing the node-level correction, the intensity field of the entire overlapping and connecting region is spatially smoothed and its boundaries are made continuous. Local mean correction and boundary transition zone settings are used to eliminate abrupt changes in intensity, ensuring the spatial stability and physical rationality of the intensity distribution. The result after correction and smoothing is the punch interaction intensity distribution.

[0025] Furthermore, the method provided in the application embodiment, which divides the mesh based on the independent regions, overlapping and connecting regions, and their punch interaction intensity distribution, further includes: For the independent regions, a uniform grid is used, and the grid density is determined based on the expected deformation degree of the punch; for the overlapping and connecting regions, the grid density ratio is set according to the punch interaction strength, and the grid is divided based on the grid density ratio.

[0026] In this embodiment, the independent regions are first meshed. Specifically, the size parameters of the mesh cells are determined based on the geometry, material properties, and punch loading conditions of the independent regions. The mesh density is determined based on the expected deformation degree of the punch, which is obtained by analyzing the relationship between the punch loading parameters and the plastic response of the billet. During the process design stage, numerical simulation is used to record the average plastic strain value and deformation depth of the billet in the corresponding stress region under different loading pressures, loading speeds, and contact areas. For example, under the same material conditions, when the punch loading pressure is 150 MPa, the local deformation depth is approximately 5 mm; when the pressure is 80 MPa, the deformation depth is approximately 2 mm. Based on this deformation depth range, the basic size of the mesh is determined so that each mesh cell covers at least one material deformation gradient change. Based on this, when the deformation depth is between 3 mm and 5 mm, the mesh cell side length is 1 mm; when the deformation depth is between 1 mm and 2 mm, the mesh cell side length is 2 mm. By using the above method, the mesh density of independent regions can be matched with the actual loading deformation characteristics, ultimately resulting in a uniform and continuous finite element mesh structure.

[0027] Subsequently, for overlapping and connecting regions, the mesh density ratio is set according to the punch interaction intensity, and meshing is performed based on this ratio. In this process, firstly, based on the previously obtained punch interaction intensity distribution, the intensity gradient range of the region is extracted, and the interaction intensity is divided into high, medium, and low intervals. Then, the mesh density ratio is determined; for example, the mesh density ratio for the high-intensity, medium-intensity, and low-intensity regions is set to 3:2:1, corresponding to element sizes of 0.5mm, 1mm, and 1.5mm, respectively. Then, during spatial partitioning, the gradient direction of the interaction intensity is used as the dominant direction, decreasing the mesh density from the high-intensity region to the low-intensity region. To ensure smooth mesh transition, a transition zone is set at the boundary of the intensity regions, and within this zone, linear density interpolation is used to adjust the mesh size, making the mesh edge length continuously change, for example, gradually transitioning from 0.5mm to 1mm, and then to 1.5mm. This proportional partitioning method ensures that the high-interaction-intensity region has sufficient mesh resolution to capture complex stress changes, while the low-interaction-intensity region maintains a lower density to improve computational efficiency.

[0028] After dividing the independent and overlapping regions, node matching and boundary smoothing are used to achieve a continuous connection between the two mesh structures in terms of geometry and stress transfer. This ultimately establishes the finite element mesh space.

[0029] Furthermore, the method provided in the application embodiment, which involves meshing the automotive parts and establishing a finite element mesh space, also includes: Based on historical forging data of automotive parts, a meshing prediction model is trained; according to the geometric characteristics of the automotive parts, the meshing prediction model is used to predict meshing and output a structural meshing scheme; the structural meshing scheme is used as a constraint to optimize and adjust the meshing network that is segmented according to the independent regions, overlapping and connecting regions and their punch interaction intensity distribution.

[0030] In this embodiment, a meshing prediction model is first trained based on historical forging data of automotive parts. Specifically, historical process data of different types of automotive parts during multi-directional forging are obtained from a historical database, including punch loading pressure, loading direction, temperature distribution, billet material properties, strain distribution, and the final meshing result. Then, by organizing and extracting features from the historical data, a correspondence between geometric feature parameters and meshing features is established. Using features such as geometric complexity, wall thickness change rate, curvature distribution, and strain energy density as inputs, and mesh density distribution and element size as outputs, a supervised learning algorithm is used to train and validate the meshing prediction model. Through continuous iterative optimization, the model can predict the optimal meshing distribution pattern based on the geometric features of the part. After training, a meshing prediction model that can be used for meshing prediction is obtained.

[0031] Next, based on the geometric features of the automotive parts, a mesh generation prediction model is used to predict mesh generation. Specifically, the 3D geometric data of the automotive parts is input into the mesh generation prediction model. The model, based on the correspondence between geometric features and mesh density distribution learned during the training phase, outputs mesh generation parameters for different regions. The model automatically generates mesh density levels, element sizes, and boundary transition relationships for different regions based on factors such as the structural complexity, surface curvature, and stress distribution characteristics of the parts. For locations with abrupt geometric changes or concentrated stress, the mesh density is increased; for smooth or uniformly stressed regions, the mesh density is decreased to balance accuracy and efficiency. Through this process, a structural mesh generation scheme matching the geometric features of the automotive parts is obtained.

[0032] Finally, the structural meshing scheme is used as a constraint to optimize and adjust the meshing network segmented based on independent regions, overlapping regions, and the distribution of punch interaction intensity. In this process, the mesh density, element size, and boundary transition characteristics of each region in the structural meshing scheme are used as constraint inputs, and simultaneously optimized in conjunction with the mesh structure obtained from the punch interaction intensity segmentation. During optimization, priority is given to ensuring a dense mesh distribution in high interaction intensity regions, while adjusting the element size in low-intensity regions with reference to the structural meshing scheme, ensuring consistency in both mechanical distribution and geometric structure. Through iterative correction and boundary smoothing, the meshes in different regions are made geometrically continuous and coordinated in stress transfer. After optimization, a meshing network that balances punch interaction intensity characteristics and the constraints of the structural meshing scheme is obtained.

[0033] Step S200: Perform forging process parameter simulation based on finite element mesh space to obtain various finite element forging parameters.

[0034] Furthermore, the method provided in the application embodiments, which obtains various finite element forging parameters by simulating forging process parameters based on finite element mesh space, also includes: Based on the distribution relationship of the punch and the control parameters, the forging process parameters are simulated in the finite element mesh space to obtain the physical field parameters of each finite element, including strain field, temperature field and stress field; the strain field, temperature field and stress field are used as the finite element forging parameters.

[0035] In this embodiment, when simulating forging process parameters based on a finite element mesh space, the forging process parameters are first simulated according to the distribution relationship of the punches and control parameters. In this process, the loading path is determined based on the distribution relationship of the punches in the established finite element mesh space. The punch distribution relationship consists of the position, loading direction, and loading sequence of each punch in space. The loading path is formed by spatially mapping the loading direction vector of each punch to the corresponding nodes on the billet surface. Subsequently, forging control parameters are set, including loading pressure, loading speed, friction coefficient, initial temperature, and thermal conductivity coefficient, and boundary conditions and contact conditions are determined. After setting the loading conditions, the finite element solution of the forging process is executed. The load is applied stepwise according to the time step, the displacement of the mesh nodes is calculated, and the plastic strain distribution of each element is calculated through the displacement gradient between nodes, forming a strain field. The amount of heat generated is calculated based on the plastic work and friction work generated by the material during deformation, and the temperature change distribution of each element is calculated in combination with the thermal conductivity parameters to obtain the temperature field. Based on the stress balance equation under load and the constitutive relation of the material, the internal stress distribution of each mesh element is calculated to obtain the stress field.

[0036] After the solution is completed, the simulation results are extracted and mapped. The strain field, temperature field, and stress field data are spatially interpolated and smoothed according to the coordinates of the finite element mesh nodes to ensure continuous field distribution and boundary consistency. Through this process, the strain field, temperature field, and stress field of each finite element are obtained in the finite element mesh space, and the three together constitute the finite element forging parameters.

[0037] Step S300: Based on the positional distribution of the finite element mesh space, the positioning and comparison of each finite element forging parameter are performed to obtain the mesh gradient.

[0038] Furthermore, in the method provided in the application embodiment, the method for positioning and comparing each finite element forging parameter according to the positional distribution of the finite element mesh space to obtain the mesh gradient also includes: Based on the spatial distribution of each finite element in the finite element mesh space, the difference in physical field parameters between adjacent distributed finite element elements is calculated to obtain the strain field gradient, temperature field gradient, and stress field gradient, respectively. Weighting coefficients are configured for each physical field gradient, and the strain field gradient, temperature field gradient, and stress field gradient are weighted to obtain the composite gradient field of the finite element pair. The composite gradient field is then labeled onto the finite element mesh space according to the edge location of the relational network of the finite element elements to obtain the mesh gradient.

[0039] In this embodiment, when locating and comparing the forging parameters of each finite element according to the positional distribution of the finite element mesh space, the difference in physical field parameters between adjacent finite element elements is first calculated according to the spatial distribution of each finite element element in the finite element mesh space. Specifically, in the established finite element mesh space, each finite element element corresponds to a specific three-dimensional coordinate position and node topology. By reading the strain field, temperature field, and stress field values ​​at adjacent element nodes, and comparing their changes at spatially adjacent positions, the parameter difference between nodes is calculated. To ensure the continuity of the calculation, only elements with shared nodes or geometric proximity less than a preset threshold are selected for difference analysis. By organizing the parameter differences between adjacent elements, the strain field gradient, temperature field gradient, and stress field gradient are obtained respectively.

[0040] Next, the weighting coefficients for each physical field gradient are configured. To comprehensively reflect the influence of different physical fields during the forging process, weighting coefficients are set based on the importance of the strain field, temperature field, and stress field to the plastic flow and microstructure evolution of the metal. For example, in the high-temperature plastic deformation stage, the temperature field gradient has a significant impact on material flowability, so its weight can be relatively increased; in the dimensional accuracy control stage, the weight of the strain field gradient can be increased accordingly. By applying the corresponding weighting coefficients to each physical field gradient, the strain field gradient, temperature field gradient, and stress field gradient are weighted and calculated to obtain the composite gradient field of the finite element pair.

[0041] After the composite gradient field calculation is completed, edge positioning is performed according to the relationship network between finite element elements, and the composite gradient field is labeled to the finite element mesh space. In this process, based on the element boundaries of the finite element mesh, the composite gradient values ​​between each pair of elements are assigned to the corresponding boundary nodes. By interpolating and smoothing the gradient continuity between nodes, the composite gradient field achieves spatial mapping throughout the entire finite element mesh space. After this process, the mesh gradient is obtained.

[0042] Step S400: Cluster according to the grid gradient distribution, identify the deviation distribution space, and match the corresponding punch according to the deviation amount in the deviation distribution space to optimize the forging parameters and obtain the optimized control strategy so that the compensation deviation amount meets the gradient distribution constraint of the finite element grid space.

[0043] In this embodiment, clustering is first performed based on the grid gradient distribution. In this process, feature vectors, including gradient magnitude, gradient direction, and spatial distribution characteristics, are extracted from the composite gradient field. Clustering and spatial aggregation analysis are then performed on the gradient data to identify the gradient spatial distribution characteristics. Finally, based on a forged gradient threshold, the gradient spatial distribution is traversed to determine regions where the gradient is greater than a set threshold as the deviation distribution space.

[0044] Subsequently, forging parameters are optimized by matching the corresponding punches according to the deviation amount in the deviation distribution space. In this process, the working domain of each punch in three-dimensional space is calculated by combining the geometric parameters, positional parameters, and kinematic constraints of the punches. A force distribution model is established based on the punch shape and loading direction, forming a punch influence space model. Then, the control range of each punch within the deviation distribution space is calculated, and the deviation area and the undevised area are balanced and optimized according to gradient distribution constraints to obtain an optimized control strategy. After obtaining the optimized control strategy, the parameters of the forging process are verified and corrected to ensure that the compensation deviation meets the gradient distribution constraints of the finite element mesh space. Specifically, the optimized control strategy is input into the forging simulation model, and the loading pressure, loading speed, and loading sequence of each punch are iteratively adjusted. Based on the gradient differences in each region of the deviation distribution space, the strain field, temperature field, and stress field responses corresponding to changes in the punch loading parameters are calculated in real time, and the gradient distribution of the finite element mesh space is regenerated. When the deviation between the newly calculated mesh gradient distribution and the target gradient distribution gradually decreases, and the compensation deviation remains consistent with the gradient change direction, it indicates that the optimized control strategy can effectively coordinate the stress and deformation of each punch during multi-directional forging. In this state, the high-gradient region in the deviation distribution space is effectively weakened, while the low-gradient region is moderately enhanced, thus ensuring that the compensation deviation satisfies the gradient distribution constraints of the finite element mesh space within the overall spatial range.

[0045] Through the above process, the final optimized control strategy achieves precise matching and coordinated control of the punch parameters, making the deformation and temperature distribution of the material more uniform during the forging process, improving the overall forging accuracy, and meeting the gradient distribution requirements determined by finite element simulation.

[0046] Furthermore, the method provided in the application embodiments, which clusters based on the grid gradient distribution to identify the deviation distribution space, further includes: Feature vectors in the composite gradient field are extracted, including gradient magnitude, gradient direction, and spatial distribution features. Gradient clustering is performed according to gradient magnitude and gradient direction, and spatial aggregation is performed according to the spatial distribution features of the clusters to identify the gradient spatial distribution. Based on the forging gradient threshold, the gradient spatial distribution is traversed to identify the deviation distribution space, where the deviation distribution space is the gradient distribution region where the gradient is greater than the forging gradient threshold.

[0047] In this embodiment, the composite gradient field is first processed within the finite element mesh space to extract feature vectors. Specifically, the gradient change rate is calculated for each finite element in the three-dimensional coordinate direction to obtain the gradient magnitude; the main change direction is determined by comparing the relative proportions of the gradient components in the three directions to obtain the gradient direction; and then, based on the spatial distribution characteristics of the gradient changes between adjacent elements, the gradient distribution pattern of the elements in space is extracted, thereby forming a feature vector containing gradient magnitude, gradient direction, and spatial distribution characteristics.

[0048] Subsequently, gradient clustering was performed based on gradient magnitude and gradient direction. By calculating the numerical distance and directional angle differences between the eigenvectors of each finite element, a similarity-based clustering analysis method was used to group finite element elements with similar gradient magnitudes and consistent gradient directions into the same cluster group, thereby achieving grouping and classification based on gradient change trends and obtaining preliminary gradient partitioning results.

[0049] Then, spatial aggregation is performed according to the spatial distribution characteristics of the clusters. By analyzing the relative positional relationships of finite element elements within each cluster in the finite element mesh space, adjacent and connected element regions are identified, and connectivity judgment methods are used to merge these elements into continuous spatial regions, forming gradient distribution regions with spatial integrity, thereby identifying the gradient spatial distribution.

[0050] Finally, the gradient space distribution is traversed based on the forging gradient threshold. The forging gradient threshold, determined according to material deformation characteristics and process control parameters, is used to distinguish between normal and abnormal deformation zones. In the finite element mesh space, the gradient magnitude of each element is compared point-by-point with the forging gradient threshold. Continuous regions with gradients greater than the forging gradient threshold are identified as deviation distribution spaces, ultimately obtaining the spatial range of the deviation distribution space and its location distribution within the finite element mesh space.

[0051] Furthermore, in the method provided in the application embodiments, matching the corresponding punch according to the deviation amount in the deviation distribution space to optimize forging parameters and obtain an optimized control strategy, it further includes: Based on the punch's geometric parameters, position parameters, and kinematic constraints, the working domain of each punch in three-dimensional space is calculated. Based on the punch's shape and working direction, a force distribution model of the punch's action on the material is established. Combined with the working domain, a punch influence space model is generated. Based on the punch influence space model, the control range of each punch in the deviation distribution space is calculated. Based on gradient distribution constraints, the balance optimization of deviation and non-deviation regions is performed according to the control range of each punch to obtain the optimized control strategy.

[0052] In this embodiment, the working domain of each punch in three-dimensional space is first calculated based on the punch's geometric parameters, position parameters, and kinematic constraints. During this process, the punch's movement path is spatially discretized along the loading direction using the punch's end face dimensions, shape curvature, and installation direction. Contact detection is performed at each discrete position with the workpiece surface to determine the area of ​​effective contact. By spatially superimposing the effective contact areas of all discrete positions, the effective working area of ​​the punch throughout the entire loading stroke is obtained, i.e., the working domain.

[0053] Subsequently, based on the punch shape and working direction, a force distribution model of the punch's effect on the material is established, and a punch influence space model is generated in conjunction with the working domain. In this process, data such as the contact distance from the punch end face to the blank surface, the local contact angle, and the loading direction are used to determine the force distribution law. Within the working domain, the force intensity is calculated from the center of the punch end face to the edge along the loading direction; the force intensity decreases linearly or piecewise with increasing distance from the center. The contact force at the edge is corrected using a preset coefficient. The calculated spatial force distribution is superimposed on the aforementioned working domain to obtain the force influence distribution of the punch in three-dimensional space, i.e., the punch influence space model.

[0054] Next, based on the impact space model of the punches, the control range of each punch in the deviation distribution space is calculated. In this process, the impact space model of the punches and the deviation distribution space are spatially superimposed in three-dimensional coordinates to obtain their overlapping area. The intensity values ​​of the overlapping area are then filtered, retaining only the areas where the effect intensity is higher than a preset threshold, thus forming the effective control range of each punch within the deviation distribution space.

[0055] Finally, based on gradient distribution constraints, the deviation amount and the unbiased region are balanced and optimized according to the control range of each punch to obtain an optimized control strategy. During the process, the loading parameters are corrected for regions with excessively high gradients in the deviation distribution space, using the grid gradient distribution as a constraint. By adjusting the loading pressure, loading speed, or loading stroke of the punch, the gradient amplitude in the deviation region is gradually reduced while ensuring that the gradient change in the unbiased region does not exceed the allowable range, so that the compensation deviation amount is spatially consistent with the grid gradient distribution. For example, when the gradient in a certain region exceeds a set threshold of 10%, the loading pressure of the punch corresponding to that region is reduced proportionally, and moderately increased in adjacent regions to maintain overall gradient balance. Through iterative calculation and constraint correction, an optimized control strategy that satisfies the gradient distribution constraints of the finite element grid space is obtained.

[0056] In summary, the embodiments of this application have at least the following technical effects: This application, based on the punch distribution in multi-directional forging, meshes automotive parts and establishes a finite element mesh space. Forging process parameters are simulated based on this finite element mesh space to obtain various finite element forging parameters. According to the positional distribution of the finite element mesh space, the forging parameters are compared and matched to obtain the mesh gradient. Clustering is performed based on the mesh gradient distribution to identify the deviation distribution space. Forging parameters are then adjusted and optimized by matching the corresponding punch according to the deviation amount in the deviation distribution space, resulting in an optimized control strategy that ensures the compensation deviation meets the gradient distribution constraints of the finite element mesh space. This invention solves the technical problems of low forging accuracy and difficulty in effectively controlling forging deviations in the multi-directional forging process in existing technologies. By introducing finite element mesh gradient analysis and deviation distribution identification for parameter adjustment and optimization, it achieves the technical effects of improving forging accuracy and realizing automatic compensation for forming deviations.

[0057] Example 2 is based on the same inventive concept as the finite element analysis-based multi-directional forging process optimization method for automotive parts in the previous examples, such as... Figure 2 As shown, this application provides a multi-directional forging process optimization system for automotive parts based on finite element analysis. The system and method embodiments in this application are based on the same inventive concept. The system includes: The mesh generation module 11 is used to mesh automotive parts based on the punch distribution in multi-directional forging, and establish a finite element mesh space; the simulation module 12 is used to simulate forging process parameters based on the finite element mesh space to obtain various finite element forging parameters; the positioning comparison module 13 is used to position and compare the various finite element forging parameters according to the position distribution of the finite element mesh space to obtain the mesh gradient; the control and optimization module 14 is used to cluster according to the mesh gradient distribution, identify the deviation distribution space, and match the corresponding punch according to the deviation amount in the deviation distribution space to control and optimize the forging parameters, so as to obtain an optimized control strategy to ensure that the compensation deviation amount meets the gradient distribution constraint of the finite element mesh space.

[0058] Furthermore, the system is also used to implement the following functions: Based on the positional relationship of the punches and the forging response range of the punches, independent regions and overlapping connection regions are determined; the interaction intensity analysis of the overlapping connection regions is performed to obtain the punch interaction intensity distribution of the overlapping connection regions; the independent regions, overlapping connection regions and their punch interaction intensity distributions are used to divide the mesh and establish the finite element mesh space.

[0059] Furthermore, the system is also used to implement the following functions: Based on historical forging data of automotive parts, a meshing prediction model is trained; according to the geometric characteristics of the automotive parts, the meshing prediction model is used to predict meshing and output a structural meshing scheme; the structural meshing scheme is used as a constraint to optimize and adjust the meshing network that is segmented according to the independent regions, overlapping and connecting regions and their punch interaction intensity distribution.

[0060] Furthermore, the system is also used to implement the following functions: The direction of the impact force of the punch is analyzed in the overlapping connection area to determine the interaction relationship; the intensity distribution is analyzed according to the action distance between the punch and the overlapping connection area to obtain the intensity distribution; the intensity distribution is corrected using the interaction relationship to obtain the punch interaction intensity distribution in the overlapping connection area, wherein if the interaction relationship is complementary, the intensity distribution is superimposed; if the interaction relationship is interfering, the intensity distribution is reduced.

[0061] Furthermore, the system is also used to implement the following functions: For the independent regions, a uniform grid is used, and the grid density is determined based on the expected deformation degree of the punch; for the overlapping and connecting regions, the grid density ratio is set according to the punch interaction strength, and the grid is divided based on the grid density ratio.

[0062] Furthermore, the system is also used to implement the following functions: Based on the distribution relationship of the punch and the control parameters, the forging process parameters are simulated in the finite element mesh space to obtain the physical field parameters of each finite element, including strain field, temperature field and stress field; the strain field, temperature field and stress field are used as the finite element forging parameters.

[0063] Furthermore, the system is also used to implement the following functions: Based on the spatial distribution of each finite element in the finite element mesh space, the difference in physical field parameters between adjacent distributed finite element elements is calculated to obtain the strain field gradient, temperature field gradient, and stress field gradient, respectively. Weighting coefficients are configured for each physical field gradient, and the strain field gradient, temperature field gradient, and stress field gradient are weighted to obtain the composite gradient field of the finite element pair. The composite gradient field is then labeled onto the finite element mesh space according to the edge location of the relational network of the finite element elements to obtain the mesh gradient.

[0064] Furthermore, the system is also used to implement the following functions: Feature vectors in the composite gradient field are extracted, including gradient magnitude, gradient direction, and spatial distribution features. Gradient clustering is performed according to gradient magnitude and gradient direction, and spatial aggregation is performed according to the spatial distribution features of the clusters to identify the gradient spatial distribution. Based on the forging gradient threshold, the gradient spatial distribution is traversed to identify the deviation distribution space, where the deviation distribution space is the gradient distribution region where the gradient is greater than the forging gradient threshold.

[0065] Furthermore, the system is also used to implement the following functions: Based on the punch's geometric parameters, position parameters, and kinematic constraints, the working domain of each punch in three-dimensional space is calculated. Based on the punch's shape and working direction, a force distribution model of the punch's action on the material is established. Combined with the working domain, a punch influence space model is generated. Based on the punch influence space model, the control range of each punch in the deviation distribution space is calculated. Based on gradient distribution constraints, the balance optimization of deviation and non-deviation regions is performed according to the control range of each punch to obtain the optimized control strategy.

[0066] It should be noted that the order of the embodiments described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0067] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing the multi-directional forging process of automotive parts based on finite element analysis, characterized in that, include: Based on the punch distribution in multi-directional forging, the automotive parts are meshed and a finite element mesh space is established. Forging process parameters were simulated based on finite element mesh space to obtain various finite element forging parameters. Based on the positional distribution of the finite element mesh space, the positioning and comparison of each finite element forging parameter are performed to obtain the mesh gradient; Clustering is performed based on the grid gradient distribution to identify the deviation distribution space. Then, the forging parameters are adjusted and optimized according to the deviation amount in the deviation distribution space and the corresponding punch. An optimized control strategy is obtained so that the compensation deviation amount meets the gradient distribution constraint of the finite element grid space.

2. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 1, characterized in that, Based on the punch distribution in multi-directional forging, the automotive parts are meshed, and a finite element mesh space is established, including: Based on the distribution and position of the punches, and the forging response range of the punches, determine the independent areas and overlapping areas; The interaction intensity of the overlapping connection area is analyzed to obtain the punch interaction intensity distribution of the overlapping connection area; The finite element mesh space is established by dividing the independent regions, overlapping and connecting regions and their punch interaction intensity distribution.

3. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 2, characterized in that, Meshing automotive parts and establishing a finite element mesh space also includes: A segmentation prediction model was trained based on historical forging data of automotive parts. Based on the geometric features of the automotive parts, the meshing prediction model is used to predict the meshing and output a structural meshing scheme. Using the aforementioned structural mesh partitioning scheme as a constraint, the partitioning network, which is divided according to the distribution of independent regions, overlapping and connecting regions, and their punch interaction intensity, is optimized and adjusted.

4. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 2, characterized in that, The interaction intensity of the overlapping connection region is analyzed to obtain the punch interaction intensity distribution of the overlapping connection region, including: The direction of the impact force in the overlapping and connecting areas is analyzed to determine the interaction relationship. Intensity distribution is obtained by performing an intensity distribution analysis based on the interaction distance between the punch and the overlapping connection area; The intensity distribution is corrected using the interaction relationship to obtain the punch interaction intensity distribution in the overlapping connection area. If the interaction relationship is complementary, the intensity distribution is superimposed; if the interaction relationship is interfering, the intensity distribution is reduced.

5. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 4, characterized in that, Mesh division is performed based on the independent regions, overlapping and connecting regions, and their punch interaction intensity distribution, including: A uniform grid is used for the independent region, and the grid density is determined based on the expected deformation of the punch. For the overlapping and connecting areas, the grid density ratio is set according to the punch interaction intensity, and the grid is divided based on the grid density ratio.

6. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 1, characterized in that, Forging process parameters were simulated using a finite element mesh space to obtain various finite element forging parameters, including: Based on the distribution relationship of the punch and the control parameters, the forging process parameters are simulated in the finite element mesh space to obtain the physical field parameters of each finite element, including strain field, temperature field and stress field. The strain field, temperature field, and stress field are used as the finite element forging parameters.

7. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 6, characterized in that, Based on the positional distribution of the finite element mesh space, the positioning and comparison of each finite element forging parameter are performed to obtain the mesh gradient, including: Based on the spatial distribution of each finite element in the finite element mesh space, the difference in physical field parameters between adjacent distributed finite element elements is calculated, and the strain field gradient, temperature field gradient and stress field gradient are obtained respectively. By configuring the weighting coefficients of each physical field gradient, the strain field gradient, temperature field gradient, and stress field gradient are weighted and calculated to obtain the composite gradient field of the finite element pair. The relationship network edges are located according to the finite element elements, and the composite gradient field is labeled to the finite element mesh space to obtain the mesh gradient.

8. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 7, characterized in that, Clustering is performed based on the grid gradient distribution to identify the spatial distribution of deviations, including: Extract feature vectors from the composite gradient field, including gradient magnitude, gradient direction, and spatial distribution features; Gradient clustering is performed based on gradient magnitude and gradient direction, and spatial aggregation is performed based on the spatial distribution characteristics of the clusters to identify the spatial distribution of gradients. Based on the forging gradient threshold, the gradient space distribution is traversed to identify the deviation distribution space, where the deviation distribution space is the gradient distribution region where the gradient is greater than the forging gradient threshold.

9. The method for optimizing the multi-directional forging process of automotive parts based on finite element analysis according to claim 8, characterized in that, According to the deviation amount in the aforementioned deviation distribution space, the forging parameters are adjusted and optimized by matching the corresponding punch, resulting in an optimized control strategy, including: Based on the geometric parameters, position parameters, and kinematic constraints of the punches, calculate the working domain of each punch in three-dimensional space; Based on the shape and working direction of the punch, a force distribution model of the punch on the material is established, and combined with the working domain, a space model of the punch's influence is generated. Based on the aforementioned punch influence space model, calculate the control range of each punch in the deviation distribution space; Based on gradient distribution constraints, the optimized control strategy is obtained by balancing the deviation amount and the unbiased region according to the control range of each punch.

10. A multi-directional forging process optimization system for automotive parts based on finite element analysis, characterized in that, The system is used to execute the finite element analysis-based multi-directional forging process optimization method for automotive parts as described in any one of claims 1-9, and the system includes: The mesh generation module is used to generate meshes for automotive parts based on the punch distribution in multi-directional forging, and to establish a finite element mesh space. The simulation module is used to simulate forging process parameters based on finite element mesh space and obtain various finite element forging parameters. The positioning and comparison module is used to position and compare each finite element forging parameter according to the positional distribution of the finite element mesh space to obtain the mesh gradient; The control and optimization module is used to cluster according to the grid gradient distribution, identify the deviation distribution space, and match the corresponding punch according to the deviation amount in the deviation distribution space to control and optimize the forging parameters, so as to obtain an optimized control strategy so that the compensation deviation amount meets the gradient distribution constraint of the finite element grid space.

Images