An automobile cover mold aided design method based on digital twinning

Abstract

The application discloses an automobile covering part mold auxiliary design method based on digital twinning, relates to the mold design field, and comprises the following steps: establishing a three-dimensional model of an automobile covering part mold; binding a geometric layer, a material layer and an energy layer according to each spatial coordinate point in the three-dimensional model to obtain a layered digital twinning body of the automobile covering part mold; solving an energy distribution matrix through finite element analysis and calculating an energy gradient field; calculating a thickness increment of the layered digital twinning body to correct the thickness; correcting the angle of the geometric layer of the layered digital twinning body based on an energy field ordered selection strategy by using a differential evolution algorithm; trial-producing the automobile covering part mold, re-correcting the angle and the thickness of the layered digital twinning body, and generating a final automobile covering part mold design result. The layered digital twinning body which is bound by the geometric layer, the material layer and the energy layer is established, and the forming quality and the design efficiency of the automobile covering part mold are improved.

Description

Technical Field

[0001] This invention relates to the field of mold design, and in particular to an auxiliary design method for automotive body panel molds based on digital twins. Background Technology

[0002] With the trend towards lightweighting and increasingly complex styling in automobiles, the design of automotive body panel molds is gradually shifting from traditional geometric modeling to intelligent design involving multiphysics coupling. Currently, the design process for automotive body panel molds generally relies on CAD / CAM software for geometric modeling, followed by static or quasi-static simulation of the stamping process using finite element analysis software to evaluate the stress and deformation of the mold. Some studies also introduce thermo-mechanical coupling analysis to predict the thermal deformation effects of automotive body panel molds during the forming process. Meanwhile, some improvement methods attempt to modify the geometric parameters of automotive body panel molds by combining structural optimization with data-driven correction. For example, multi-objective optimization based on sensitivity analysis or genetic algorithms can be used to balance forming accuracy and material utilization, or energy density can be introduced as an evaluation index to improve the distribution of geometric parameters.

[0003] However, most of these methods lack a hierarchical energy-geometry-material correlation model, making it impossible to achieve a consistent mapping of energy distribution between virtual simulation and physical prototyping. In particular, the energy field calibration and dynamic correction of automotive body panel molds still rely on manual judgment. Therefore, the digital twin system for automotive body panel molds urgently needs to establish a self-consistent mechanism from energy distribution prediction to self-correction of geometric thickness and angle, in order to achieve synchronous updates between virtual and physical models and high-precision assisted design. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a digital twin-based auxiliary design method for automotive body panel molds to solve the problem of not being able to achieve a consistent mapping of energy distribution between virtual simulation and physical prototyping in the automotive body panel mold design stage.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a digital twin-based method for auxiliary design of automotive body panel molds, comprising,

[0008] A three-dimensional model of an automotive body panel mold is established. By defining the geometric layer, material layer, and energy layer of the three-dimensional model, and binding the geometric layer, material layer, and energy layer according to each spatial coordinate point in the three-dimensional model, a layered digital twin of the automotive body panel mold is obtained.

[0009] Multiple loads are applied to the layered digital twin, and the unit volume strain energy density at each spatial coordinate point in the layered digital twin is solved by finite element analysis to generate the energy distribution matrix and calculate the energy gradient field.

[0010] Based on the energy distribution matrix and energy gradient field, the thickness increment of the layered digital twin is calculated, and the thickness of the geometric layers of the layered digital twin is corrected.

[0011] With the optimization objectives of minimizing surface bonding error, minimizing energy distribution variance, and minimizing energy difference between the left and right regions, the differential evolution algorithm is used to correct the angle of the geometric layers of the layered digital twin based on the energy field ordered selection strategy.

[0012] Based on the optimized layered digital twin, a prototype of the automotive body panel mold is manufactured. The energy distribution matrix of the prototype automotive body panel mold and the layered digital twin are compared point by point, and the relative deviation is calculated. The energy distribution matrix of the layered digital twin is calibrated, and the angle and thickness of the layered digital twin are corrected to generate the final automotive body panel mold design result.

[0013] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, wherein: the geometric layer refers to the set of spatial coordinates of each spatial coordinate point in the three-dimensional model;

[0014] The material layer refers to the material parameters at each spatial coordinate point in the three-dimensional model;

[0015] The energy layer refers to the strain energy density and energy gradient at each spatial coordinate point in the three-dimensional model;

[0016] The various loads include impact force, temperature field, and boundary displacement constraints.

[0017] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the specific steps for solving the unit volume strain energy density at each spatial coordinate point in the layered digital twin are as follows:

[0018] The stamping force at each moment is evenly distributed as normal force according to the contact area of ​​the upper and lower dies of the automotive body panel mold. The tangential friction stress is calculated based on the friction coefficient of the upper and lower die contact surfaces. The friction energy per unit volume is calculated based on the tangential friction stress and the stamping time.

[0019] A temperature field is applied to the layered digital twin, and the product of the thermal expansion coefficient and the temperature change of the automotive body panel mold is calculated as the thermal strain tensor.

[0020] At each time step of the load time, the displacement vector of each spatial coordinate point in the layered digital twin is linearized to obtain the small strain tensor.

[0021] Based on the elastic modulus and Poisson's ratio of the automotive body panel mold, a thermoelastic constitutive model of the layered digital twin is established to obtain the stress tensor;

[0022] The strain energy density per unit volume is calculated based on the frictional energy per unit volume, the small strain tensor, the thermal strain tensor, and the stress tensor.

[0023] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the specific steps for calculating the thickness increment of the layered digital twin are as follows:

[0024] Based on the energy distribution matrix, the average energy value and energy distribution variance of the layered digital twins are calculated, and the spatial coordinate points in all layered digital twins are divided into high-energy region points and low-energy region points.

[0025] Based on the elastic modulus of the automotive body panel mold, determine the thickness adjustment ratio coefficient for each spatial coordinate point in the layered digital twin;

[0026] The thickness adjustment amount at each spatial coordinate point is calculated based on the normal component of the energy gradient field and the thickness adjustment ratio coefficient.

[0027] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, wherein: the thickness adjustment value of the high-energy region point is positive, indicating an increase in thickness, and the thickness adjustment value of the low-energy region point is negative, indicating a decrease in thickness.

[0028] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the specific steps for correcting the thickness of the geometric layers of the layered digital twin are as follows:

[0029] The thickness adjustment is superimposed onto the layered digital twin to generate a new thickness distribution;

[0030] During the thickness correction process, the spatial positions of the cavity, shape constraint boundaries and parting line remain unchanged, and the thickness change only occurs in the normal direction of the cavity.

[0031] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the energy field ordered selection strategy refers to arranging individuals in the differential evolution algorithm in ascending order according to the energy distribution variance in the mutation operation of the differential evolution algorithm, and constructing a mutation vector according to the energy ordered selection principle.

[0032] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the specific steps for correcting the angles of the geometric layers of the layered digital twin are as follows:

[0033] The automotive body panel mold is divided into several sections based on the ribs, hole chains, and parting lines. The correction axis of each section is defined as the average direction unit vector of the normal vectors of all spatial coordinate points in the section.

[0034] The rotation center of each partition is defined as the centroid position of all points in the partition;

[0035] The angle correction amount for each partition represents the extent of rotation of the partition as a whole around the correction axis. A positive value indicates rotation in the direction of the main forming process, and a negative value indicates rotation in the opposite direction of the main forming process.

[0036] The spatial coordinate points of each partition are rotated according to the angle correction amount and the rotation axis to complete the angle correction of the spatial coordinate points in each partition.

[0037] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the specific steps for calibrating the energy distribution matrix of the layered digital twin are as follows:

[0038] Strain monitoring equipment was used to record the strain changes at measurement points of the automotive body panel mold during the trial production stage, and the energy distribution matrix of the automotive body panel mold was calculated.

[0039] Extract the unit volume strain energy density of the corresponding spatial coordinate points in the prototype automotive body panel mold from the layered digital twin, and calculate the relative and average deviations of all measurement points of the automotive body panel mold.

[0040] Based on the measurement error of strain at each measurement point of the automotive body panel mold in the trial production, a deviation threshold is set. When the average deviation exceeds the deviation threshold, the energy distribution matrix of the layered digital twin is calibrated regionally.

[0041] As a preferred embodiment of the digital twin-based automotive body panel mold auxiliary design method of the present invention, the punching force is applied to the layered digital twin using a trapezoidal segmentation mode over time, and the temperature field is applied to the layered digital twin using a linear segmentation mode.

[0042] The beneficial effects of this invention are as follows: By establishing a layered digital twin with interconnected geometric, material, and energy layers, the energy characteristics of automotive body panel molds can be calculably and traceably described. By uniformly modeling the strain energy density and friction energy under multi-physics load conditions, the energy distribution matrix and energy gradient field of the automotive body panel mold throughout the entire stamping cycle are obtained. Then, the thickness and angle parameters are corrected with the energy field as a guide. The energy distribution of the digital twin is calibrated using strain monitoring data from the trial production stage. A virtual-real closed-loop feedback mechanism is constructed to realize the dynamic correction of the energy field and the real-time update of design parameters. This achieves synergistic optimization between the energy response, geometric shape, and material properties of automotive body panel molds, thereby improving the forming quality and design efficiency of automotive body panel molds. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the 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.

[0044] Figure 1 A flowchart illustrating a digital twin-based method for auxiliary design of automotive body panel molds;

[0045] Figure 2 Flowchart for energy field analysis and energy gradient calculation;

[0046] Figure 3 A flowchart for thickness correction based on energy distribution;

[0047] Figure 4 This is a flowchart for angle correction and virtual-real closed-loop calibration. Detailed Implementation

[0048] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0049] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0050] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0051] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a digital twin-based method for auxiliary design of automotive body panel molds, comprising the following steps:

[0052] S1. Establish a three-dimensional model of the automotive body panel mold. Define the geometry layer, material layer, and energy layer of the three-dimensional model, and bind the geometry layer, material layer, and energy layer according to the spatial coordinates of each spatial coordinate point in the three-dimensional model to obtain a layered digital twin of the automotive body panel mold.

[0053] Based on the surface data of the automotive body panel, such as 3D models in CAD, computer-aided design software, such as CATIA or NX, is used to create a 3D model of the automotive body panel mold. The 3D model includes cavity surfaces, ribs, hole chains, parting lines, and shape constraint boundaries.

[0054] Define the geometry layer, material layer, and energy layer of the 3D model, as follows:

[0055] A geometric layer refers to the set of spatial coordinates of each spatial coordinate point in a 3D model;

[0056] The material layer refers to the material parameters at each spatial coordinate point in the three-dimensional model, including elastic modulus, Poisson's ratio, coefficient of thermal expansion, and yield strength.

[0057] The energy layer refers to the strain energy density and energy gradient at each spatial coordinate point in a three-dimensional model;

[0058] It should be noted that, during initialization, as defined in this case, the settings for strain energy density and energy gradient are only placeholders, and the initial values ​​can be set to 0. The specific values ​​will be calculated in subsequent steps.

[0059] Based on the spatial coordinates of each spatial coordinate point in the 3D model, the geometric layer, material layer, and energy layer are parameter-bound to obtain a layered digital twin.

[0060] S2. Apply various loads to the layered digital twin, solve the unit volume strain energy density at each spatial coordinate point in the layered digital twin by performing finite element analysis, generate the energy distribution matrix, and calculate the energy gradient field.

[0061] Import the layered digital twin into multiphysics simulation software, such as ANSYS Digital Twin and Siemens Digital Twin Platform;

[0062] Based on the stamping time of the automotive body panel mold during the automotive body panel production process, the load time of the layered digital twin is set.

[0063] Based on the load time, various loads are applied to the layered digital twin, including impact force, temperature field, and boundary displacement constraints. The specific steps are as follows:

[0064] For the application of the punching force, a trapezoidal segmented application over time is adopted, including the force climbing stage, the force holding stage, and the force falling stage, to obtain the punching force curve that changes over time.

[0065] A segmented temperature field is applied to the layered digital twin. The application of the temperature field adopts a linear segmented mode, that is, the application of the temperature field is divided into three stages: heating, isothermal and cooling, and the temperature changes linearly in each stage.

[0066] Based on the shape constraint boundary and parting line in the geometric layer, a set of positioning points is defined, including the origin of the three-dimensional coordinate system, linear constraint points, and planar constraint points. Displacement constraints are applied to obtain the displacement vector of each spatial coordinate point. Specifically, in the finite element analysis, each positioning point has three displacement degrees of freedom. In the layered digital twin, the displacement in three directions is constrained at the origin of the three-dimensional coordinate system, the displacement perpendicular to the direction connecting the two points is constrained at the linear constraint points, and the displacement perpendicular to the direction of the plane determined by the first two points is constrained at the planar constraint points, thereby eliminating rotational degrees of freedom. Slip constraints are applied to the shape constraint boundary to prevent unreasonable penetration of edge nodes.

[0067] The stamping force at each moment is evenly distributed as normal pressure according to the contact area of ​​the upper and lower dies of the automotive body panel mold, and the tangential friction stress is calculated based on the friction coefficient of the upper and lower die contact surfaces. The friction coefficient of the upper and lower dies can be obtained from existing databases (such as VDI 3400 Forming Data Catalogue) based on the type of sheet metal, lubrication conditions and mold surface treatment status. For example, the value is 0.12 to 0.18.

[0068] In any contact area, the tangential frictional stress and the tangential relative slip rate are integrated to calculate the frictional power density of the tangential frictional force on the boundary tangential displacement. Based on the frictional power density, the frictional energy density is obtained by integrating over the entire stamping time. The frictional energy density is then standardized to obtain the frictional energy per unit volume. The tangential relative slip rate refers to the tangential motion velocity component of the automotive body panel surface relative to the automotive body panel mold surface. The velocity difference is obtained by differentiating the displacement of each spatial coordinate point in the layered digital twin with time, and then projected along the tangential direction.

[0069] According to the linear elastic thermo-coupling theory, when the temperature of the automotive body panel mold rises from the reference temperature, free expansion occurs. The reference temperature can be defined as the room temperature in the production workshop. The free expansion caused by the temperature rise of the automotive body panel mold is represented by the thermal strain, which is the product of the thermal expansion coefficient of the automotive body panel mold and the temperature change. The thermal strain tensor is obtained by calculating the thermal strain.

[0070] At each time step of the loading time, the displacement vector of each spatial coordinate point in the layered digital twin is linearly approximated to obtain the small strain tensor. Specifically, all spatial coordinate points in the layered digital twin are discretized by finite element method. Each spatial coordinate point and its adjacent spatial coordinate points form a local neighborhood. The displacement vector of the spatial coordinate points in the local neighborhood is linearly fitted according to the spatial coordinate difference between adjacent time steps to obtain the local displacement gradient. In the local neighborhood of each spatial coordinate point, the displacement gradients in the three directions are linearized to form a displacement gradient matrix. The small strain tensor is obtained by solving the average value of the displacement gradient matrix and the transpose of the displacement gradient matrix.

[0071] Based on the elastic modulus and Poisson's ratio of the automotive body panel mold, a thermoelastic constitutive model of the layered digital twin is established, and the stress tensor is obtained. The calculation formula is as follows:

[0072] ;

[0073] ;

[0074] ;

[0075] ;

[0076] In the formula, For stress tensor, It is a fourth-order elastic stiffness tensor. For small strain tensors, For thermal strain tensor, Let Lamé constant be . It is a second-order identity tensor. It is a symmetric projection operator. Shear modulus For elastic modulus, Poisson's ratio, express With stress tensor Multiply each component one by one in each direction and then sum them up.

[0077] The formula for calculating the strain energy density per unit volume is as follows:

[0078] ;

[0079] In the formula, The strain energy density per unit volume. The frictional energy per unit volume This is the friction energy conversion factor. Friction energy is partially converted into heat energy and partially retained on the surface of the automotive body panel mold in the form of microscopic plastic deformation. An exemplary value range is... ;

[0080] The energy distribution matrix is ​​obtained by averaging the strain energy density per unit volume at each spatial coordinate point in the layered digital twin with respect to the load time and summarizing the results.

[0081] The energy gradient field is obtained by solving the second-order central difference of the strain energy density per unit volume at each spatial coordinate point in the energy distribution matrix.

[0082] S3. Based on the energy distribution matrix and energy gradient field, calculate the thickness increment of the layered digital twin and correct the thickness of the geometric layers of the layered digital twin.

[0083] Based on the energy distribution matrix, the average energy value and energy distribution variance of the layered digital twin are calculated. The average energy value reflects the energy level of the overall stress area of ​​the automotive body panel mold, while the energy distribution variance characterizes the non-uniformity of energy distribution in each area. Based on the average energy value, the spatial coordinate points of all layered digital twins are divided into two categories: spatial coordinate points with energy not lower than the average value belong to high-energy areas, indicating potential stress concentration or deformation risk; spatial coordinate points with energy lower than the average value belong to low-energy areas, indicating relatively insufficient energy in force transmission.

[0084] Based on the elastic modulus of the automotive body panel mold, the thickness adjustment ratio coefficient for each spatial coordinate point in the layered digital twin is determined. Specifically, the magnitude of the thickness adjustment ratio coefficient is determined by two parts: the first part is the elastic modulus of the corresponding spatial coordinate point in the material layer, and the second part is the non-uniformity of the entire energy field. The larger the elastic modulus, the less likely the material is to deform. Therefore, the thickness adjustment should be smaller under the same energy gradient conditions. The greater the energy non-uniformity, the smaller the adjustment coefficient should be to avoid overcompensation under severe energy gradient conditions. The calculation formula is as follows:

[0085] ;

[0086] In the formula, This is the thickness adjustment ratio factor. For reference, the elastic modulus is selected to be in the same order of magnitude as common automotive body panel mold steels, such as H13 and SKD61. This allows the thickness adjustment in the calculation to maintain a consistent ratio with the material stiffness, ensuring relative balance of the same type of steel in different regions. This is the average energy value. The variance of the energy distribution;

[0087] Based on the normal component of the energy gradient field and the thickness adjustment ratio coefficient, the thickness adjustment amount at each spatial coordinate point is calculated. That is, the thickness adjustment amount is equal to the product of the component of the energy gradient in the normal direction at each spatial coordinate point and the thickness adjustment ratio coefficient. Among them, the thickness adjustment amount at high energy region points is positive, indicating an increase in thickness, and the thickness adjustment amount at low energy region points is negative, indicating a decrease in thickness.

[0088] Once the thickness adjustment amount for all spatial coordinate points is determined, the thickness adjustment amount is superimposed on the layered digital twin to generate a new thickness distribution. During the update process, the spatial positions of the cavity, shape constraint boundary, and parting line remain unchanged, that is, the coordinates of all boundary points remain fixed, while the thickness change only occurs in the normal direction of the cavity.

[0089] S4. With the optimization objectives of minimizing surface bonding error, minimizing energy distribution variance, and minimizing energy difference between the left and right regions, the differential evolution algorithm is used to correct the angle of the geometric layers of the layered digital twin based on the energy field ordered selection strategy.

[0090] Surface bonding error refers to the degree of spatial geometric consistency between the geometric layer of the layered digital twin and the surface of the automotive body panel. It reflects the degree of deviation of the geometric layer from the surface of the automotive body panel in three-dimensional space. The smaller the value, the closer the geometric shape is to the surface of the automotive body panel. The larger the value, the larger the geometric deviation. Specifically, on the surface of the automotive body panel, the minimum distance between each spatial coordinate point in the geometric layer is calculated to obtain the distance from the spatial coordinate point to the surface. The distances from all spatial coordinate points to the surface are averaged on the surface of the automotive body panel to obtain the mean square error term. The mean square error term is scaled and integrated over the entire surface area of ​​the automotive body panel and normalized through an exponential function to obtain the surface bonding error.

[0091] The layered digital twin is divided into left and right regions by the main fractal line. The average energy difference between the left and right regions is calculated to obtain the energy difference between the left and right regions. The energy difference between the left and right regions is then normalized to obtain the target energy difference.

[0092] The energy variance target is obtained by normalizing the energy distribution variance.

[0093] The total objective function value of the layered digital twin is obtained by weighted summation of the surface bonding target error, energy difference target, and energy variance target.

[0094] Using the angle correction as an optimization variable, the differential evolution algorithm is employed to iteratively optimize the overall objective function. The specific steps are as follows:

[0095] Specifically, based on the springback deviation and draft angle safety of automotive body panels, boundary constraints are applied to the angle correction amount, for example, limiting the range of values ​​for the angle correction amount to [specific value]. ;

[0096] Initialize the parameters of the differential evolution algorithm, including the population size, maximum number of iterations, and random seed, for example, the population size is 40 and the maximum number of iterations is 60;

[0097] In the initial population generation stage, the angle correction value is uniformly and randomly selected according to the range of the angle correction value to generate several individuals. Here, an individual refers to the angle correction value of the hierarchical digital twin.

[0098] Based on the energy field ordered selection strategy, the mutation vector is constructed. Specifically, in each generation of calculation, individuals are arranged in ascending order according to their energy distribution variance, and the mutation vector is constructed according to the energy ordered selection principle. Among them, individuals with lower energy distribution variance are selected as the differential base individuals, and differential reference individuals are selected from individuals with higher energy distribution variance to form a differential vector with energy-oriented characteristics. The generated mutation vector essentially represents the adjustment trend of high energy imbalance relative to low energy balance solution, so that the newly generated candidate individuals are more likely to evolve along the energy balance direction, further making the evolution direction of candidate solutions consistent with the natural trend of energy balance, and the optimization path more in line with physical laws, thereby improving the physical rationality of angle correction convergence.

[0099] Within the range of angle correction values, cross-operation is performed between the mutation vector and the original individual to generate the experimental vector;

[0100] Calculate the total objective function value of the experimental vector and the original individual respectively. If the total objective function value of the experimental vector is smaller, then proceed to the next generation; otherwise, retain the original individual.

[0101] In each iteration, physical parameters such as load time, impact force, temperature field, thickness and material layer are kept constant. The angle of the geometric layer is corrected according to the angle correction amount. The energy distribution matrix and energy gradient field are recalculated and substituted into the calculation of the total objective function value. The optimal individual is selected to enter the next generation.

[0102] After reaching the maximum number of iterations, the individual with the smallest total objective function value among all generations is output as the optimal result, thus obtaining the optimal angle correction amount.

[0103] The automotive body panel mold is divided into several zones based on the ribs, hole chains, and parting lines. The correction axis of each zone is defined as the average direction unit vector of the normal vectors of all spatial coordinate points in the zone. The unit normal vectors of all spatial coordinate points in the same zone are weighted and averaged to serve as the main forming direction of each zone.

[0104] The rotation center of each partition correction axis is defined as the centroid position of all spatial coordinate points of the partition;

[0105] The angle correction amount of each partition represents the extent of rotation of the partition as a whole around the correction axis. If the angle correction value is positive, it means that the angle correction direction is in the same direction as the main forming direction. If the angle correction value is negative, it means that the angle correction direction is opposite to the main forming direction.

[0106] Each spatial coordinate point is rotated in the direction of angular correction, and then translated back to its original spatial position. This process updates the coordinates of the spatial coordinate points within each partition and smooths the boundary regions of adjacent partitions, resulting in a new geometric layer. This new geometric layer maintains the original thickness and boundary constraints in space, with only attitude adjustments in the normal direction. This allows the geometric layer to better conform to the distribution trend of the energy field. The process of rotating and then translating back to the original spatial position does not mean canceling the rotation, but rather transferring the rotation operation from the local coordinate system back to the global coordinate system. This is because when a spatial coordinate point rotates around any non-origin (i.e., the rotation center), all points within the partition must first be vector-translated so that the rotation center of the partition, as the local coordinate origin, coincides with the global coordinate origin. Therefore, after the complete angular correction of each partition, the spatial coordinate points within the partition must be translated back to their original spatial positions.

[0107] S5. Based on the optimized layered digital twin, trial production of the automotive body panel mold is carried out. The energy distribution matrix of the trial automotive body panel mold and the layered digital twin is compared point by point, and the relative deviation is calculated. The energy distribution matrix of the layered digital twin is calibrated, and the angle and thickness of the layered digital twin are corrected to generate the final automotive body panel mold design result.

[0108] The strain change process of the automotive body panel mold was recorded at the measurement points during the trial production stage using strain monitoring equipment. The energy distribution matrix of the automotive body panel mold was calculated. The arrangement of the measurement points was based on the principle of balanced spatial distribution. According to the surface grid division of the geometric layer, the cavity surface, ribs, hole chains and parting lines were taken as the priority objects for point placement. In the planar area, the measurement points were distributed in a grid pattern along the principal stress direction and the secondary stress direction. In the curvature change area, the density of measurement points should be increased accordingly so that the measurement results can reflect the local deformation gradient.

[0109] Extract the unit volume strain energy density of the corresponding spatial coordinate points in the prototype automotive body panel mold from the layered digital twin, and calculate the relative and average deviations of all measurement points;

[0110] When the average deviation exceeds the deviation threshold, where the deviation threshold is set based on the measurement error of the strain at each measurement point of the automotive body panel mold in the trial production, such as 0.05 to 0.15, the energy distribution matrix of the layered digital twin is calibrated regionally. Specifically, the measurement points are divided into multiple natural regions according to the structure of the automotive body panel mold. Specifically, the main parting line is used as the first dividing boundary to divide the automotive body panel mold into upper and lower parts. Ribs, hole chains, and shape constraint boundaries are used as secondary dividing lines to divide the cavity surfaces of the upper and lower parts of the automotive body panel into multiple geometrically continuous natural regions. If there are different materials inside a certain natural region, such as steels of different hardness or surface-strengthened areas, the material interface is used as the new boundary to divide again to ensure that the material property parameters inside each region are consistent. The average value of the relative deviation is calculated for each natural region. Based on the magnitude of the relative deviation, the energy distribution matrix of each natural region is linearly proportionally calibrated.

[0111] Based on the calibrated energy distribution matrix, the thickness and angle are corrected again, and the final automotive body panel mold design result, namely the layered digital twin, is output, including the geometric layer, material layer and energy layer of the automotive body panel mold.

[0112] This embodiment also provides a computer device applicable to the case of an auxiliary design method for automotive body panel molds based on digital twins, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the auxiliary design method for automotive body panel molds based on digital twins as proposed in the above embodiment.

[0113] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0114] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the digital twin-based automotive body panel mold-aided design method proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0115] In summary, this invention achieves calculable and traceable description of the energy characteristics of automotive body panel molds by establishing a layered digital twin with interconnected geometric, material, and energy layers. Through unified modeling of strain energy density and friction energy under multi-physics load conditions, the energy distribution matrix and energy gradient field of the automotive body panel mold throughout the entire stamping cycle are obtained. Then, thickness and angle parameters are corrected guided by the energy field. Strain monitoring data from the trial production stage are used to calibrate the energy distribution of the digital twin, constructing a virtual-real closed-loop feedback mechanism to achieve dynamic correction of the energy field and real-time updating of design parameters. This enables synergistic optimization between the energy response, geometric shape, and material properties of the automotive body panel mold, improving the forming quality and design efficiency of the mold.

[0116] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A digital-twin-based method for assisting in the design of an automobile panel mold, characterized in that: The application relates to a method for establishing a three-dimensional model of an automobile panel mold, and the method comprises the following steps: a three-dimensional model of an automobile panel mold is established, a geometric layer, a material layer and an energy layer of the three-dimensional model are defined, the geometric layer, the material layer and the energy layer are bound according to each spatial coordinate point in the three-dimensional model, and a split-layer digital twin of the automobile panel mold is obtained; a plurality of loads are applied to the split-layer digital twin, unit volume strain energy density of each spatial coordinate point in the split-layer digital twin is solved through finite element analysis, an energy distribution matrix is generated, and an energy gradient field is calculated; the unit volume strain energy density of each spatial coordinate point in the split-layer digital twin is solved, and the specific steps are as follows, a stamping force at each moment is uniformly distributed as a normal pressure according to a contact area of an upper mold and a lower mold of the automobile panel mold, tangential frictional stress is calculated according to a friction coefficient of a contact surface of the upper mold and the lower mold, and unit volume friction energy is calculated according to the tangential frictional stress and a stamping time; a temperature field is applied to the split-layer digital twin, and a product of a thermal expansion coefficient of the automobile panel mold and a temperature change amount is calculated as a thermal strain tensor; a first-order linearization process is performed on a displacement vector of each spatial coordinate point in the split-layer digital twin at each time step of the load time, and a small strain tensor is obtained; a thermoelastic constitutive of the split-layer digital twin is established according to an elastic modulus and a Poisson's ratio of the automobile panel mold, and a stress tensor is obtained; unit volume strain energy density is calculated based on the unit volume friction energy, the small strain tensor, the thermal strain tensor and the stress tensor; the unit volume strain energy density of each spatial coordinate point in the split-layer digital twin is averaged at the load time, and is summarized, and an energy distribution matrix is obtained; the energy gradient field is obtained by solving the unit volume strain energy density of each spatial coordinate point in the energy distribution matrix through a second-order central difference; the thickness increment of the split-layer digital twin is calculated according to the energy distribution matrix and the energy gradient field, and the geometric layer of the split-layer digital twin is corrected in thickness; a surface fitting error, an energy distribution variance and a left-right regional energy difference are minimized as optimization targets, an energy field ordered selection strategy is used, and a differential evolution algorithm is used to correct the angle of the geometric layer of the split-layer digital twin; the automobile panel mold is trial-produced according to the optimized split-layer digital twin, the energy distribution matrix of the trial-produced automobile panel mold and the split-layer digital twin is compared point by point, a relative deviation is calculated, the energy distribution matrix of the split-layer digital twin is calibrated, the angle and the thickness of the split-layer digital twin are re-corrected, and a final automobile panel mold design result is generated.

2. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized in that: The geometric layer refers to a spatial coordinate set of each spatial coordinate point in the three-dimensional model; The material layer refers to material parameters of each spatial coordinate point in the three-dimensional model; The energy layer refers to strain energy density and an energy gradient of each spatial coordinate point in the three-dimensional model; The plurality of loads include stamping force, temperature field and boundary displacement constraint.

3. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized in that: The thickness increment of the split-layer digital twin is calculated, and the specific steps are as follows, the average energy value and the energy distribution variance of the split-layer digital twin are calculated according to the energy distribution matrix, and all spatial coordinate points in the split-layer digital twin are divided into high-energy area points and low-energy area points; According to the elastic modulus of the automobile panel die, the thickness adjustment proportion coefficient of each spatial coordinate point in the sub-layer digital twin is determined; Based on the normal component of the energy gradient field and the thickness adjustment proportion coefficient, the thickness adjustment amount of each spatial coordinate point is calculated.

4. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 3, characterized in that: The value of the thickness adjustment amount of the high-energy area point is positive, indicating that the thickness increases, and the value of the thickness adjustment amount of the low-energy area point is negative, indicating that the thickness decreases.

5. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized by: The specific steps of the thickness correction of the geometric layer of the sub-layer digital twin are as follows, The thickness adjustment amount is superimposed into the sub-layer digital twin to generate a new thickness distribution; During the thickness correction process, the spatial positions of the cavity, the contour constraint boundary and the parting line remain unchanged, and the thickness change is only in the normal direction of the cavity.

6. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized by: The energy field ordered selection strategy refers to arranging the individuals in the differential evolution algorithm in ascending order according to the energy distribution variance in the mutation operation of the differential evolution algorithm, and constructing a mutation vector according to the energy ordered selection principle.

7. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized by: The specific steps of the angle correction of the geometric layer of the sub-layer digital twin are as follows, According to the automobile panel die, a plurality of partitions are divided according to the rib line, hole chain and parting line, and the correction axis of each partition is defined as the average direction unit vector of the normal vector of all spatial coordinate points in the partition; The rotation center of each partition is defined as the centroid position of all point coordinates in the partition; The angle correction amount of each partition represents the rotation amplitude of the partition as a whole around the correction axis, and a positive value represents rotation in the main forming direction, and a negative value represents rotation in the reverse main forming direction; The spatial coordinate points in the partition are rotated according to the angle correction amount and the rotation axis to complete the angle correction of the spatial coordinate points in each partition.

8. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 1, characterized by: The specific steps of the energy distribution matrix correction of the calibrated sub-layer digital twin are as follows, The strain monitoring equipment is used to record the strain change of the measurement points of the automobile panel die during the trial production stage, and the energy distribution matrix of the automobile panel die is calculated; The unit volume strain energy density of the corresponding spatial coordinate points in the trial production automobile panel die is extracted from the sub-layer digital twin, and the relative deviation and average deviation of all automobile panel die measurement points are calculated; According to the measurement error of the strain of each measurement point of the trial production automobile panel die, a deviation threshold is set, and when the average deviation exceeds the deviation threshold, the energy distribution matrix of the sub-layer digital twin is regionally calibrated.

9. The digital-twin-based auxiliary design method of an automobile panel mold according to claim 2, characterized by: The stamping force is applied to the sub-layer digital twin in a trapezoidal segmentation mode over time, and the temperature field is applied to the sub-layer digital twin in a linear segmentation mode.

Images