A method, device and medium for iterative optimization of automobile parts mold surface

Abstract

The present invention discloses a method, device and medium for iterative optimization of the mold surface of an automobile part, relating to the technical field of mold surface optimization, comprising the following steps: establishing a mold finite element model based on design parameters of the automobile part to obtain initial surface data; performing simulation analysis using computer software based on the mold finite element model to obtain springback surface data; solving a surface optimization path using a least squares support vector machine based on the initial surface data and the springback surface data to obtain surface optimization iterative data; iteratively optimizing the mold surface based on the surface optimization iterative data to obtain a final optimized mold surface; the present invention is used to solve the problems of existing mold surface optimization technology in which, when performing mold surface optimization for the springback problem of automobile parts, only the node displacement difference before and after the springback is considered, but the springback path is not considered, resulting in low precision, excessive number of iterations and low efficiency in mold surface optimization iterations.

Description

Technical Field

[0001] The present invention relates to the technical field of mold surface optimization, and in particular to a method, equipment and medium for iterative optimization of the mold surface of an automobile component. Background Art

[0002] Mold surface optimization technology is a method of optimizing the mold's geometry through precise simulation and calculation during the mold design phase. This technology aims to improve mold manufacturing accuracy, reduce production cycle time, lower costs, and ensure that the quality of the final product meets design requirements.

[0003] Some automotive parts, such as covers, will inevitably have springback problems during the production process, resulting in unqualified automotive parts. The springback problem cannot be fundamentally solved by simply changing process parameters and material parameters. Therefore, mold surface optimization is crucial. Mold surface optimization can fundamentally solve the springback problem of parts. The optimization idea is to use springback to allow parts to transition during the deformation process, so that the shape accuracy of the formed parts after springback just meets the requirements, truly avoiding the occurrence of springback. The traditional optimization method is to use repeated trial and error and mold repair, which is time-consuming and labor-intensive, and overly dependent on manual labor. The existing mold surface optimization method is to establish a finite model of the mold through modeling software, and then optimize the mold surface through the geometric compensation method, that is, first obtain the geometric offset of a node before and after springback based on calculation, and according to the obtained offset, apply a vector opposite to the geometric offset vector before the node rebounds. The position of the node after springback compensation is the end point of the vector. By operating this method on each node, the springback compensation of the entire surface can be obtained. , complete the surface optimization; however, in the forming process of automobile parts, springback is a continuous process. During the springback process, each node will displace according to its own path, and finally form the result of springback, which is reflected in the change of curvature. It can be regarded as a rotation at a certain point. Using only the node displacement difference before and after springback for surface optimization will cause the springback amount of automobile parts to be too large. This is because in the process of reverse compensation, the curvature of the compensation surface is too large due to reverse compensation, and the springback amount will increase accordingly. It usually takes many iterative optimizations to achieve a qualified mold surface, and the accuracy is not high. If the springback path is considered in the process of mold surface optimization, the accuracy of mold surface optimization can be improved, the number of mold surface optimization iterations can be reduced, and the efficiency of mold surface optimization can be improved. Therefore, the existing mold surface optimization technology only considers the node displacement difference before and after springback, and does not consider the springback path, which leads to low accuracy, excessive number of iterations and low efficiency of mold surface optimization iterations. Summary of the Invention

[0004] The present invention aims to solve, at least to a certain extent, one of the technical problems in the prior art by establishing a finite element model of a mold, performing simulation analysis using computer software, and solving the surface optimization path using a least squares support vector machine to optimize the mold surface iteratively; in order to solve the problem that the existing mold surface optimization technology only considers the node displacement difference before and after the springback, but does not consider the springback path, resulting in low precision, excessive number of iterations, and low efficiency in the mold surface optimization iterations when optimizing the mold surface for the springback problem of automotive parts.

[0005] To achieve the above objectives, in a first aspect, the present application provides a method for iterative optimization of a mold surface of an automobile component, comprising the following steps:

[0006] Based on the design parameters of automobile parts, a mold finite element model is established to obtain initial surface data;

[0007] Based on the mold finite element model, computer software is used to perform simulation analysis to obtain the springback surface data;

[0008] Based on the initial profile data and the springback profile data, the profile optimization path is solved using the least squares support vector machine to obtain the profile optimization iterative data;

[0009] The mold surface is optimized and iterated based on the surface optimization iteration data to obtain the final optimized mold surface.

[0010] Furthermore, based on the design parameters of the automotive parts, a mold finite element model is established to obtain initial surface data, including the following sub-steps:

[0011] According to the design parameters of automobile parts, the geometric model of automobile parts mold is constructed using CAD software, and then exported and saved to obtain the mold geometric model file;

[0012] Based on DEFORM software, using the pre-processing module of DEFORM software, import the mold geometry model file to obtain the DDEFORM mold model;

[0013] Using DEFORM software, the DEFORM mold model is meshed using tetrahedrons, the number of mesh elements is set to A1, and the mesh of the contact area between the mold and the part in the DEFORM mold model is refined. After completion, the mold finite element model is obtained;

[0014] The mold finite element model is placed in a rectangular coordinate system, and the coordinates of the mesh unit nodes of the mold finite element model surface are obtained and marked as initial surface data. The mold surface corresponding to the initial surface data is the initial surface.

[0015] Furthermore, based on the mold finite element model, computer software is used to perform simulation analysis to obtain the springback profile data, which includes the following sub-steps:

[0016] Using DEFORM software, according to the part design parameters, the material parameters, material properties, and elastic modulus of the part are input, and the material properties of the mold are input. Then, the first simulation production is performed to obtain the first simulated automotive part finite element model, which is marked as the first part simulation model, and the elastic modulus is E.

[0017] The coordinates of the mesh unit nodes on the surface of the simulation model of the first part are extracted and marked as springback profile data. The mold profile corresponding to the springback profile data is the springback profile.

[0018] Furthermore, based on the initial profile data and the springback profile data, the profile optimization path is solved using the least squares support vector machine to obtain the profile optimization iterative data, which includes the following sub-steps:

[0019] Adjust the input elastic modulus to 0.5*E, perform a second simulation, and obtain a finite element model of the second simulated automobile part, which is marked as the second part simulation model. Extract the coordinates of the mesh unit nodes on the surface of the second part simulation model and mark them as the springback intermediate surface data. The mold surface corresponding to the springback intermediate surface data is the springback intermediate surface.

[0020] The least squares support vector machine is used to solve the surface springback path.

[0021] Furthermore, solving the surface springback path using the least squares support vector machine includes the following sub-steps:

[0022] Based on the initial profile data, the rebound profile data, and the rebound intermediate profile data, the grid unit nodes where the profile rebound occurs are screened out, and the coordinates of the grid unit nodes where the profile rebound occurs in the initial profile data are marked as B0(x0i, y0i, z0i), the coordinates of the grid unit nodes corresponding to the rebound profile data are marked as B1(x1i, y1i, z1i), and the coordinates of the grid unit nodes corresponding to the rebound intermediate profile data are marked as B2(x2i, y2i, z2i), where i represents the coordinates of the i-th grid unit node, and the i in B0(x0i, y0i, z0i), B1(x1i, y1i, z1i), and B2(x2i, y2i, z2i) correspond to each other;

[0023] Based on the least squares support vector machine algorithm, x0i, y0i, x1i, y1i, x2i and y2i are used as the input of the least squares support vector machine, and z0i, z1i and z2i are used as the output of the least squares support vector machine. The least squares support vector machine is used for curve fitting to obtain the curve path from B0 to B1, which is marked as the surface rebound path B0~B1;

[0024] The surface optimization path is solved based on the compensation coefficient to obtain the surface optimization iterative data.

[0025] Furthermore, solving the surface optimization similarity path based on the compensation coefficient to obtain the surface optimization iterative data includes the following sub-steps:

[0026] Assume that the compensation coefficient is α, and the coordinates of the mesh unit node on the initial surface corresponding to the mesh unit node coordinates B1(x1i, y1i, z1i) on the surface where the rebound occurs are C1(xui, yui, zui) when the compensation coefficient is α;

[0027] Solve C1(xui, yui, zui): Draw a tangent to line B0B1 through B1(x1i, y1i, z1i) to obtain the first equation. Based on B0(x0i, y0i, z0i) and B1(x1i, y1i, z1i), find the length of line B0B1, labeled LB01. Construct a circle with B0(x0i, y0i, z0i) as the center and α*LB01 as the radius to obtain the second equation.

[0028] Solve the first and second equations simultaneously to find xui and yui. If there are two solutions, calculate the vertical distances between each solution and the springback surface based on the springback surface data, and select the solution with the shorter vertical distance from the springback surface as the final solution; zui = z0i + α*(z1i - z0i). Obtain the solution C1, labeled C1(xci, yci, zci).

[0029] If the compensation coefficient α<1, the surface optimization similarity path is: starting point B0 (x0i, y0i, z0i) along the surface springback path B0~B1 to C1 (xci, yci, zci) marked as B0~C1;

[0030] If the compensation coefficient α>1, the surface optimization similarity path is: the surface springback path B0~B1 plus the straight line B1C1, marked as B0~B1C1;

[0031] If the compensation coefficient α=1, then C1=B1, and the surface optimization similarity path is: surface springback path B0~C1;

[0032] The surface optimization path is solved based on the surface optimization similarity path to obtain the surface optimization iteration data.

[0033] Furthermore, solving the profile optimization path based on the profile optimization similarity path and obtaining the profile optimization iteration data includes the following sub-steps:

[0034] Based on B0(x0i, y0i, z0i), the normal vector of point B0(x0i, y0i, z0i) is obtained, marked as B0H, and based on C1(xci, yci, zci), the normal vector of C1(xci, yci, zci) is obtained, marked as C1H;

[0035] Move the profile optimization similarity path so that the end point of the profile optimization similarity path coincides with the starting position C1 of the profile optimization similarity path, and then rotate the profile optimization similarity path around the starting position C1 until B0H coincides with C1H. Mark the profile optimization similarity path at this time as the profile optimization path, obtain the position coordinates D (xdi, ydi, zdi) of the end point of the profile optimization path, and mark it as the profile optimization iteration data.

[0036] Furthermore, the mold surface is optimized and iterated based on the surface optimization iteration data to obtain the final optimized mold surface, which includes the following sub-steps:

[0037] Modifying the surface of the mold finite element model based on the surface optimization iteration data to obtain the optimized mold surface, and performing a third simulation production to obtain the third simulated automobile part finite element model, which is marked as the third part simulation model, and outputting the parameters of the third part simulation model. Based on the design parameters of the automobile part, it is determined whether the parameters of the third part simulation model are within the error range. If they are within the error range, the surface of the modified mold finite element model is marked as the final optimized mold surface;

[0038] If it is not within the error range, the coordinates of the grid unit nodes on the surface of the simulation model of the third part are extracted as new rebound surface data. Based on the initial surface data and the rebound surface data, the least squares support vector machine is used to solve the surface optimization path to obtain a new surface optimization path. The mold optimization surface is optimized again based on the new surface optimization path to obtain a new mold optimization surface. Simulation production is performed again to determine whether the parameters of the simulated model are within the error range. Until the parameters of the simulated model are within the error range, the surface of the mold finite element model after the last modification is marked as the final optimized surface of the mold.

[0039] In a second aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the above method are performed.

[0040] In a third aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the above method are performed.

[0041] The beneficial effects of the present invention are as follows: first, based on the design parameters of the automobile parts, the present invention establishes a mold finite element model to obtain initial profile data; then, based on the mold finite element model, computer software is used to perform simulation analysis to obtain springback profile data; then, based on the initial profile data and the springback profile data, a least squares support vector machine is used to solve a profile optimization path to obtain profile optimization iteration data; then, based on the profile optimization iteration data, the mold profile is optimized and iterated to obtain a final optimized mold profile; the mold profile is optimized for the springback problem of the automobile parts, and the profile optimization path is calculated and fitted based on the difference in node displacement before and after springback, thereby increasing the accuracy of the mold profile optimization, reducing the number of profile optimization iterations, and improving the profile optimization iteration efficiency;

[0042] The present invention establishes a finite element model of the mold and accurately simulates the forging process through finite element simulation technology. It can quickly iterate and verify different mold surface solutions, shorten the process preparation cycle, reduce material waste due to improper design, improve production efficiency, and reduce production costs; the least squares support vector machine is used to solve the surface optimization path. The least squares support vector machine has good performance for complex curve fitting tasks. When faced with data with nonlinear relationships, it can more accurately capture its internal laws. Compared with the support vector machine, it no longer requires the solution of quadratic programming, which improves computing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the steps of the method of the present invention;

[0044] Figure 2 This is a schematic diagram of the rebound intermediate profile of the present invention;

[0045] Figure 3 Schematic diagram of the surface rebound path of the present invention;

[0046] Figure 4 This is a schematic diagram of profile optimization of the present invention;

[0047] Figure 5 It is a schematic diagram of the secondary surface optimization of the present invention;

[0048] Figure 6 Schematic diagram of the structure of the electronic device of the present invention. DETAILED DESCRIPTION

[0049] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0050] Example 1, please refer to Figure 1 As shown, in the first aspect, the present application provides a method for iterative optimization of the mold surface of an automobile component, comprising the following steps:

[0051] Step S1: Based on the design parameters of the automobile parts, a mold finite element model is established to obtain initial surface data. Step S1 includes the following sub-steps:

[0052] Step S101: constructing a geometric model of an automobile component mold using CAD software according to automobile component design parameters, and exporting and saving the model to obtain a mold geometric model file;

[0053] Step S102: Based on DEFORM software, using the pre-processing module of DEFORM software, import the mold geometry model file to obtain the DDEFORM mold model; DEFORM software is mainly composed of a pre-processing module, a solver module and a post-processing module when performing finite element processing, wherein the pre-processing module can import the CAD geometry model established by other modeling software, and then perform relevant settings, mainly including: mesh division, material model selection, solver and iterative algorithm selection, boundary conditions and other simulation parameters definition; the solver module is a finite element solution calculation module, which generally does not require user intervention in the process of solving the calculation task, but can observe the information of the intermediate results of the solution calculation in real time and can terminate the solution calculation process at any time; the post-processing module can conveniently and intuitively display the results of the solution calculation, so that users can analyze and process the material simulation process;

[0054] Step S103: Using DEFORM software, mesh the DEFORM mold model using tetrahedrons. The number of mesh elements is set to A1. In this embodiment, the number of mesh elements is set to 128,000. The mesh of the contact area between the mold and the part in the DEFORM mold model is refined to improve the accuracy of subsequent production simulation results. After completion, a finite element model of the mold is obtained.

[0055] Step S104: placing the mold finite element model in a rectangular coordinate system, obtaining the coordinates of the mesh unit nodes of the mold finite element model surface, marking them as initial surface data, and the mold surface corresponding to the initial surface data is the initial surface;

[0056] During the specific implementation process, since the finite element software DEFORM does not have the ability to perform three-dimensional solid modeling, but it has the ability to connect to data from other three-dimensional solid modeling software, it can easily import the three-dimensional geometric models generated by other modeling software into DEFORM to complete the finite element analysis simulation. Therefore, the geometric model of the automobile parts mold is constructed by CAD software and then imported into DEFORM for subsequent processing; the number of grid units needs to be set according to the actual application scenario. Setting the number of grid units too small will affect the accuracy of the production simulation results. Setting the number of grid units too large will cause the production simulation time to be too long, affecting the efficiency of the surface optimization iteration.

[0057] Step S2, performing simulation analysis using computer software based on the mold finite element model to obtain springback profile data; Step S2 includes the following sub-steps:

[0058] Step S201: Using DEFORM software, input the material parameters, material properties, and elastic modulus of the part according to the part design parameters, input the material properties of the mold, and the elastic modulus is E;

[0059] Material parameters include density, yield strength, tensile strength, hardness, coefficient of thermal expansion, Poisson's ratio, shear modulus, and bulk modulus. Material properties include rigidity and plasticity. Elastic modulus is a physical quantity that describes a material's ability to resist elastic deformation under external force. It is also called Young's modulus and is the ratio of normal stress to the corresponding normal strain during the elastic deformation phase of a material. It reflects the material's ability to resist elastic deformation. From a macroscopic perspective, the elastic modulus measures the object's ability to resist elastic deformation. From a microscopic perspective, it reflects the bond strength between atoms, ions, or molecules. The unit of elastic modulus is usually Pascal (Pa). In this real-time example, the elastic modulus E is 70 GPa.

[0060] Step S202, then performing a first simulation production to obtain a first simulated automobile component finite element model, which is marked as a first component simulation model;

[0061] Step S203, extracting the coordinates of the grid unit nodes on the surface of the first part simulation model, marking them as springback profile data, and the mold profile corresponding to the springback profile data is the springback profile;

[0062] In the specific implementation process, a rigid body refers to an ideal object whose shape and size remain unchanged under the action of external forces; in reality, any solid will have a certain degree of deformation after being subjected to force, but this deformation is extremely small relative to the geometric dimensions of the object itself, so it can be ignored in many applications. In the actual production process, the mold surface should remain unchanged to ensure the accuracy of the produced parts, so the material properties of the mold should be set to a rigid body; a plastic body refers to an object that will undergo permanent deformation under the action of external forces; unlike a rigid body, the shape of a plastic body will not completely return to its original state after the external force is removed; in the actual production process, the blank must be shaped by the mold to become a suitable part, so the material properties of the part must be set to a plastic body.

[0063] Step S3, based on the initial profile data and the springback profile data, using the least squares support vector machine to solve the profile optimization path to obtain the profile optimization iterative data; Step S3 includes the following sub-steps:

[0064] Step S301, please refer to Figure 2 As shown, the input elastic modulus is adjusted to 2*E, and a second simulation production is performed to obtain a finite element model of the automobile parts for the second simulation, which is marked as the second part simulation model;

[0065] When solving the surface optimization path, in addition to the starting point and the end point, an intermediate point is also required. The elastic modulus of a part is inversely proportional to the springback amount, and the elastic modulus affects the degree of springback of the part. Therefore, when solving the surface optimization path, when the elastic modulus of the material is increased, the intermediate surface obtained is located between the design surface and the springback surface. However, in order to ensure the accuracy of solving the surface optimization path, the intermediate surface cannot be too close to the design surface or the springback surface. Therefore, the elastic modulus after adjustment cannot be too close to the elastic modulus before adjustment, nor can it be too large or close to the elastic modulus before adjustment.

[0066] Step S302: extracting the coordinates of the mesh unit nodes on the surface of the second part simulation model and marking them as springback intermediate profile data. The mold profile corresponding to the springback intermediate profile data is the springback intermediate profile.

[0067] Step S303, using the least squares support vector machine to solve the surface springback path; Step S303 includes the following sub-steps:

[0068] Step S3031, filtering out mesh unit nodes where surface springback occurs based on the initial surface data, the springback surface data, and the springback intermediate surface data;

[0069] Step S3032: The coordinates of the mesh unit node where the surface rebound occurs in the initial surface data are labeled as B0(x0i, y0i, z0i), the coordinates of the mesh unit node corresponding to the rebound surface data are labeled as B1(x1i, y1i, z1i), and the coordinates of the mesh unit node corresponding to the rebound intermediate surface data are labeled as B2(x2i, y2i, z2i), where i represents the coordinates of the i-th mesh unit node, and the i in B0(x0i, y0i, z0i), B1(x1i, y1i, z1i), and B2(x2i, y2i, z2i) correspond to each other; for example, the point (x02, y03, z01) on the initial surface corresponds to the point (x12, y13, z11) on the rebound surface, and corresponds to the point (x22, y23, z21) on the rebound intermediate surface.

[0070] Step S3033, please refer to Figure 3 As shown, based on the least squares support vector machine algorithm, x0i, y0i, x1i, y1i, x2i and y2i are used as the input of the least squares support vector machine, z0i, z1i and z2i are used as the output of the least squares support vector machine, and the least squares support vector machine is used to perform curve fitting to obtain the curve path from B0 to B1, which is marked as the surface rebound path B0~B1; in this embodiment, the curve fitting is performed by LS-SVMlab-v1.8 in the MATLAB software toolbox to obtain the curve path from B0 to B1;

[0071] Step S304, solving the profile optimization path based on the compensation coefficient to obtain profile optimization iteration data; Step S304 includes the following sub-steps:

[0072] Step S3041, set the compensation coefficient to α, and set the mesh unit node coordinates B1 (x1i, y1i, z1i) on the initial surface corresponding to the mesh unit node coordinates on the surface where springback occurs to C1 (xui, yui, zui) when the compensation coefficient is α; because the part springback problem itself is a nonlinear problem, the springback amount of the part may increase to a certain extent after the mold surface is compensated. Therefore, the problem of increased springback amount after compensation should be considered, and the compensation amount should be redesigned. The compensation amount should also be adaptively increased or decreased. Therefore, the compensation coefficient is introduced into the optimization iteration of the mold surface. Usually, in actual production, the compensation coefficient has a value range of [0.8, 1.2]. After the compensation coefficient is added, the springback path will also change with the change of the compensation coefficient, so the change of the springback path must be determined;

[0073] Step S3042: Solve C1(xui, yui, zui): Draw a tangent line to the line B0B1 through B1(x1i, y1i, z1i) to obtain the first equation, which is as follows:

[0074] yui=[f(x0i,x2i)+f(x0i,x2i,x1i,)(2x1i-x2i-x0i)](xui-x1i)+y1i, where f(x0i,x2i)+f(x0i,x2i,x1i,)(2x1i-x2i-x0i) is the slope of the surface springback path B0-B1 at point B1;

[0075] Step S3043, based on B0(x0i, y0i, z0 i ) and B1(x1i, y1i, z1i), we get the length of the line B0B1, labeled LB01. With B0(x0i, y0i, z0i) as the center and α*LB01 as the radius, we can construct a circle to get the second equation: α*LB01=(xui-x0i) 2 +(yui-y0i) 2 ;

[0076] Step S3044: Solve xui and yui by simultaneously solving the first and second equations. If there are two solutions, calculate the vertical distances between the two solutions and the springback surface based on the springback surface data, and select the solution with the shorter vertical distance from the springback surface as the final solution.

[0077] Step S3045, solve zui = z0i + α*(z1i - z0i); obtain the solved C1, marked as C1(xci, yci, zci);

[0078] Step S3046: If the compensation coefficient α is less than 1, then C1 is located on the springback path B0-B1, so the springback path becomes B0-C1. The surface optimization similarity path is: starting point B0 (x0i, y0i, z0i) along the surface springback path B0-B1 to C1 (xci, yci, zci) is marked as B0-C1.

[0079] Step S3047: If the compensation coefficient α>1, then C1 is outside the springback path B0-B1, and the profile optimization similarity path is: the profile springback path B0-B1 plus the straight line B1C1, marked as B0-B1C1;

[0080] Step S3048: If the compensation coefficient α=1, then C1=B1, and the surface optimization similarity path is: surface springback path B0-C1;

[0081] Step S305, please refer to Figure 4 As shown, Figure 4 The compensation coefficient α>1, the surface optimization path is solved based on the surface optimization similarity path to obtain the surface optimization iteration data; step S305 includes the following sub-steps:

[0082] Step S3051, based on B0(x0i, y0i, z0i), obtain the normal vector of point B0(x0i, y0i, z0i), marked as B0H;

[0083] Step S3052: obtain the normal vector of C1(xci, yci, zci) based on C1(xci, yci, zci), marked as C1H;

[0084] Step S3053, moving the profile optimization similarity path so that the end point of the profile optimization similarity path coincides with the starting position C1 of the profile optimization similarity path;

[0085] Step S3054: Then, the profile optimization similarity path is rotated around the starting position C1 until B0H and C1H coincide with each other, and the profile optimization similarity path at this time is marked as the profile optimization path;

[0086] Step S3055, obtaining the position coordinates D (xdi, ydi, zdi) of the end point of the profile optimization path, and marking them as profile optimization iteration data;

[0087] During the implementation process, the least squares support vector machine is a machine learning algorithm based on statistical theory. Its main advantages are high computational efficiency, easy implementation and strong generalization ability. The least squares support vector machine has good performance for complex curve fitting tasks. When faced with data with nonlinear relationships, it can more accurately capture its internal laws, and compared with the support vector machine, it no longer requires the solution of quadratic programming, which improves the computational efficiency and fitting accuracy. Because the parts produced before and after mold surface optimization only change in curvature, their complexity does not change; so the shape of the mold surface rebound path is similar before and after surface optimization, so in this embodiment, the shape of the surface optimization similar path is used instead of the surface optimization path.

[0088] Step S4, iteratively optimizing the mold surface based on the surface optimization iteration data to obtain the final optimized mold surface; Step S4 includes the following sub-steps:

[0089] Step S401, modifying the profile of the mold finite element model based on the profile optimization iteration data to obtain the optimized mold profile, and performing a third simulation production to obtain a third simulated automobile component finite element model, which is marked as the third component simulation model;

[0090] Step S402: Output the parameters of the third part simulation model. Based on the design parameters of the automobile part, determine whether the parameters of the third part simulation model are within the error range. For example, if the maximum design length of a certain automobile panel is 520 mm ± 3 mm, the error range is [517 mm, 523 mm]. That is, if the length is less than 517 mm or greater than 523 mm, it will be unqualified.

[0091] Step S403: If the error is within the range, the modified mold finite element model surface is marked as the final optimized mold surface;

[0092] Step S404: if the error is not within the error range, the coordinates of the mesh unit nodes on the surface of the simulation model of the third part are extracted as new springback surface data;

[0093] Step S405, please refer to Figure 5 If it is not within the error range, the coordinates of the mesh unit nodes on the surface of the simulation model of the third part are extracted as new springback surface data. Based on the initial surface data and the springback surface data, the surface optimization path is solved by the least squares support vector machine to obtain a new surface optimization path. The mold optimization surface is optimized again based on the new surface optimization path to obtain a new mold optimization surface. The simulation production is performed again and it is judged whether the parameters of the simulated model are within the error range. Until the parameters of the simulated model are within the error range, the surface of the mold finite element model after the last modification is marked as the final optimized surface of the mold.

[0094] In the specific implementation process, it is often difficult to achieve the ideal effect with one surface optimization iteration, so a second and more surface optimization iterations are required. However, surface optimization iterations based on the surface optimization path can reduce the number of surface optimization iterations, increase the accuracy of single surface optimization, and improve the efficiency of surface optimization iterations compared to traditional surface optimization iterations based on geometric displacement.

[0095] Example 2, please refer to Figure 6 As shown, Figure 6 A schematic diagram of the structure of an electronic device is provided. The electronic device may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions from the memory. When the computer-readable instructions are executed by the processor, the steps of a method for iterative optimization of the mold surface of an automotive component are executed to achieve the following functions: establishing a mold finite element model based on the design parameters of the automotive component to obtain initial surface data; performing simulation analysis using computer software based on the mold finite element model to obtain springback surface data; solving a surface optimization path using a least squares support vector machine based on the initial and springback surface data to obtain surface optimization iteration data; and iteratively optimizing the mold surface based on the surface optimization iteration data to obtain a final optimized mold surface.

[0096] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0097] Example 3. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method for iterative optimization of the mold surface of an automobile part are executed to achieve the following functions: based on the design parameters of the automobile part, a mold finite element model is established to obtain initial surface data; based on the mold finite element model, simulation analysis is performed using computer software to obtain rebound surface data; based on the initial surface data and the rebound surface data, the surface optimization path is solved using a least squares support vector machine to obtain surface optimization iterative data; based on the surface optimization iterative data, the mold surface is optimized iteratively to obtain the final optimized mold surface.

[0098] Through the description of the above embodiments, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on this understanding, the above technical solutions, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment.

[0099] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for iterative optimization of automobile parts mold surface, characterized in that: The steps include: Based on the design parameters of automobile parts, a mold finite element model is established to obtain initial surface data; Based on the mold finite element model, computer software is used to perform simulation analysis to obtain the springback surface data; Based on the initial profile data and the springback profile data, the profile optimization path is solved using the least squares support vector machine to obtain the profile optimization iterative data; Optimize the mold surface iteratively based on the surface optimization iterative data to obtain the final optimized mold surface; Based on the design parameters of the automotive parts, the mold finite element model is established to obtain the initial surface data, which includes the following sub-steps: According to the design parameters of automobile parts, the geometric model of automobile parts mold is constructed using CAD software, and then exported and saved to obtain the mold geometric model file; Based on DEFORM software, using the pre-processing module of DEFORM software, import the mold geometry model file to obtain the DDEFORM mold model; Using DEFORM software, the DEFORM mold model is meshed using tetrahedrons, the number of mesh elements is set to A1, and the mesh of the contact area between the mold and the part in the DEFORM mold model is refined. After completion, the mold finite element model is obtained; The mold finite element model is placed in a rectangular coordinate system, and the coordinates of the mesh unit nodes of the mold finite element model surface are obtained and marked as initial surface data. The mold surface corresponding to the initial surface data is the initial surface; Based on the mold finite element model, computer software is used to perform simulation analysis to obtain the springback surface data, which includes the following sub-steps: Using DEFORM software, according to the part design parameters, the material parameters, material properties, and elastic modulus of the part are input, and the material properties of the mold are input. Then, the first simulation production is performed to obtain the first simulated automotive part finite element model, which is marked as the first part simulation model, and the elastic modulus is E. Extracting the coordinates of the mesh unit nodes on the surface of the simulation model of the first part and marking them as springback profile data; the mold profile corresponding to the springback profile data is the springback profile; Based on the initial profile data and the springback profile data, the profile optimization path is solved using the least squares support vector machine. The profile optimization iterative data is obtained, which includes the following sub-steps: Adjust the input elastic modulus to 0.5*E, perform a second simulation, and obtain a finite element model of the second simulated automobile part, which is marked as the second part simulation model. Extract the coordinates of the mesh unit nodes on the surface of the second part simulation model and mark them as the springback intermediate surface data. The mold surface corresponding to the springback intermediate surface data is the springback intermediate surface. Least squares support vector machine is used to solve the surface springback path; Optimizing the mold surface iteratively based on the surface optimization iterative data to obtain the final optimized mold surface includes the following sub-steps: Modifying the surface of the mold finite element model based on the surface optimization iteration data to obtain the optimized mold surface, and performing a third simulation production to obtain the third simulated automobile part finite element model, which is marked as the third part simulation model, and outputting the parameters of the third part simulation model. Based on the design parameters of the automobile part, it is determined whether the parameters of the third part simulation model are within the error range. If they are within the error range, the surface of the modified mold finite element model is marked as the final optimized mold surface; If it is not within the error range, the coordinates of the grid unit nodes on the surface of the simulation model of the third part are extracted as new rebound surface data. Based on the initial surface data and the rebound surface data, the least squares support vector machine is used to solve the surface optimization path to obtain a new surface optimization path. The mold optimization surface is optimized again based on the new surface optimization path to obtain a new mold optimization surface. Simulation production is performed again to determine whether the parameters of the simulated model are within the error range. Until the parameters of the simulated model are within the error range, the surface of the mold finite element model after the last modification is marked as the final optimized surface of the mold.

2. The method for iterative optimization of the mold surface of an automobile component according to claim 1, characterized in that: Solving the surface springback path using the least squares support vector machine includes the following sub-steps: Based on the initial profile data, the rebound profile data, and the rebound intermediate profile data, the grid unit nodes where the profile rebound occurs are screened out, and the coordinates of the grid unit nodes where the profile rebound occurs in the initial profile data are marked as B0(x0i, y0i, z0i), the coordinates of the grid unit nodes corresponding to the rebound profile data are marked as B1(x1i, y1i, z1i), and the coordinates of the grid unit nodes corresponding to the rebound intermediate profile data are marked as B2(x2i, y2i, z2i), where i represents the coordinates of the i-th grid unit node, and the i in B0(x0i, y0i, z0i), B1(x1i, y1i, z1i), and B2(x2i, y2i, z2i) correspond to each other; Based on the least squares support vector machine algorithm, x0i, y0i, x1i, y1i, x2i and y2i are used as the input of the least squares support vector machine, and z0i, z1i and z2i are used as the output of the least squares support vector machine. The least squares support vector machine is used for curve fitting to obtain the curve path from B0 to B1, which is marked as the surface rebound path B0~B1; The surface optimization path is solved based on the compensation coefficient to obtain the surface optimization iterative data.

3. The method for iterative optimization of the mold surface of an automobile component according to claim 2, characterized in that: Solving the similar path of profile optimization based on the compensation coefficient and obtaining the profile optimization iterative data includes the following sub-steps: Assume that the compensation coefficient is α, and the coordinates of the mesh unit node on the initial surface corresponding to the mesh unit node coordinates B1(x1i, y1i, z1i) on the surface where the rebound occurs are C1(xui, yui, zui) when the compensation coefficient is α; Solve C1(xui, yui, zui): Draw a tangent to line B0B1 through B1(x1i, y1i, z1i) to obtain the first equation. Based on B0(x0i, y0i, z0i) and B1(x1i, y1i, z1i), find the length of line B0B1, labeled LB01. Construct a circle with B0(x0i, y0i, z0i) as the center and α*LB01 as the radius to obtain the second equation. Solve the first and second equations simultaneously to find xui and yui. If there are two solutions, calculate the vertical distances between each solution and the springback surface based on the springback surface data, and select the solution with the shorter vertical distance from the springback surface as the final solution; zui = z0i + α*(z1i - z0i). Obtain the solution C1, labeled C1(xci, yci, zci). If the compensation coefficient α<1, the surface optimization similarity path is: starting point B0 (x0i, y0i, z0i) along the surface springback path B0~B1 to C1 (xci, yci, zci) marked as B0~C1; If the compensation coefficient α>1, the surface optimization similarity path is: the surface springback path B0~B1 plus the straight line B1C1, marked as B0~B1C1; If the compensation coefficient α=1, then C1=B1, and the surface optimization similarity path is: surface springback path B0~C1; The surface optimization path is solved based on the surface optimization similarity path to obtain the surface optimization iteration data.

4. The method for iterative optimization of the mold surface of an automobile component according to claim 3, characterized in that: Solving the surface optimization path based on the surface optimization similarity path and obtaining the surface optimization iterative data includes the following sub-steps: Based on B0(x0i, y0i, z0i), the normal vector of point B0(x0i, y0i, z0i) is obtained, marked as B0H, and based on C1(xci, yci, zci), the normal vector of C1(xci, yci, zci) is obtained, marked as C1H; Move the profile optimization similarity path so that the end point of the profile optimization similarity path coincides with the starting position C1 of the profile optimization similarity path, and then rotate the profile optimization similarity path around the starting position C1 until B0H coincides with C1H. Mark the profile optimization similarity path at this time as the profile optimization path, obtain the position coordinates D (xdi, ydi, zdi) of the end point of the profile optimization path, and mark it as the profile optimization iteration data.

5. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method according to any one of claims 1 to 4 are executed.

6. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are executed.

Images