Segmented stamping thickness gradient distribution monitoring method

Abstract

The invention provides a segmented stamping thickness gradient distribution monitoring method which comprises the following steps: according to a gridding representation form, performing numerical simulation on stamping forming mechanical behaviors in a segmented stamping treatment process by adopting a finite element analysis method, and determining thickness change distribution data of each segmented region after simulation; if the fluctuation of the boundary region of each segment in the simulated thickness change distribution data exceeds a preset threshold value, pressure thinning processing parameters such as the magnitude and direction of force are adjusted, and an optimized parameter set is obtained; for processing path planning, determining a non-interference segmented thinning execution sequence by comparing the difference between the structural stability index of the overall influence evaluation data and the structural strength parameter of the initial geometric model; and according to the determined thinning execution sequence, virtual verification simulation is carried out on the metal material, whether the thicknesses of the segmented areas and the peripheral areas of the segmented areas are kept in the original thickness value within the allowable deviation range or not is judged, and final aluminum sheet local thinning scheme data are obtained.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for monitoring the thickness gradient distribution of segmented stamping. Background Technology

[0002] In the field of materials processing and manufacturing, the optimization of local properties of metallic materials has always been a crucial research direction. This field is directly related to the quality and service life of industrial products, especially in high-demand industries such as aerospace and automotive, where subtle improvements in material properties can often lead to significant safety and efficiency enhancements. Therefore, how to achieve precise control of local properties without sacrificing the overall structure has become a focus of attention for many researchers and engineers. However, current methods in metallic material processing often struggle to find a balance between local and overall performance. Many techniques, when attempting to modify a part of the material, easily trigger performance fluctuations in other areas, even leading to a decrease in the stability of the overall structure. This problem is not simply a technical limitation, but rather stems from a lack of in-depth understanding of the complex interaction between the local and overall properties of the material, thus limiting the control precision and adaptability during processing. Focusing on specific technical challenges, thickness adjustment in local areas becomes a core challenge during metallic material processing. Especially when it is necessary to thin the material in a specific area to improve performance, the uneven distribution of processing forces often affects the original thickness of surrounding areas. This uneven mechanical influence further leads to an imbalance in the internal stress distribution of the material, posing a potential risk to overall strength or durability. For example, when manufacturing a lightweight part, thinning the edge area to reduce weight may lead to stress concentration during processing, causing micro-cracks or deformation in the central area, thus affecting the reliability of the entire part. Similarly, in manufacturing automotive aluminum alloy control arms, locally thinning the connecting end to reduce weight and improve vehicle handling can result in uneven thickness between adjacent support areas due to concentrated hammering or rolling pressure during processing, leading to the accumulation of residual stress within the material. Similar problems are even more pronounced in the production of aerospace titanium alloy wing skins. If high-stress areas are thinned to meet aerodynamic requirements, the thickness deviation in surrounding areas can reach several percentage points, directly affecting the fatigue life of the skin. Therefore, how to precisely thin local areas of metal material while ensuring that the original thickness of other areas remains intact has become a critical issue that urgently needs to be addressed. Solving this problem not only improves the performance of individual parts but also directly affects the stability of the entire manufacturing process and the final application effect of the product. Summary of the Invention

[0003] This invention provides a method for monitoring the thickness gradient distribution of segmented stamping, mainly including: Initial thickness data and overall structural parameters of local areas are extracted from a pre-established three-dimensional geometric model of aluminum material to obtain the boundary conditions of the thinning target area. The aluminum sheet material is then divided into segments according to thickness gradient requirements to obtain a meshed representation. Based on the meshed representation, the finite element method is used to numerically simulate the stamping mechanical behavior during the segmented stamping process, determining the thickness variation distribution data of each segment after simulation. If the fluctuation of each segment boundary region in the simulated thickness variation distribution data exceeds a preset threshold, the thinning processing parameters, such as the magnitude and direction of the force, are adjusted to obtain an optimized parameter set. The parameter set extracts stress imbalance indices corresponding to minimum thickness fluctuations based on stress gradient analysis. A genetic algorithm is used to globally search for these stress imbalance indices to plan segmented stamping processing paths that balance stress distributions. For the processing path planning, the difference between the structural stability index of the overall impact assessment data and the structural strength parameters of the initial geometric model is compared to determine an interference-free segmented thinning execution sequence. Based on the determined thinning execution sequence, a virtual verification simulation is performed on the metal material to determine whether the thickness of each segmented region and its surrounding region remains within the allowable deviation range of the original thickness value, thus obtaining the final aluminum sheet local thinning scheme data. Further, the initial thickness data and overall structural parameters of the local region are extracted from the pre-established three-dimensional geometric model of the aluminum material to obtain the boundary conditions of the thinning target region. The aluminum sheet material is then segmented according to the thickness gradient requirements to obtain a meshed representation, including: The process involves acquiring vertex coordinates and normal vector data, identifying the spatial boundary contour of the target area for thinning using a convex hull algorithm, determining vertex positions based on the relationship between points and polygons, and obtaining initial boundary condition data. The Delaunay triangulation method is used to mesh the aluminum sheet surface. The thickness gradient value of each mesh node is calculated based on the distance from the center of the target area to the boundary. A linear interpolation method is used to establish the thickness transition relationship between adjacent nodes, resulting in a triangular mesh structure with thickness attributes. The thickness difference between adjacent nodes is calculated based on the node thickness values ​​in the triangular mesh structure, and a thickness change threshold is set. Segmented boundary lines are set where the thickness difference exceeds the threshold, generating a meshed representation.

[0004] Furthermore, based on the meshed representation, the finite element method is used to numerically simulate the stamping forming mechanical behavior during the segmented stamping process, determining the thickness variation distribution data of each segment region after simulation, including: A constitutive relation matrix is ​​constructed based on the node coordinates and thickness attributes in the meshed representation. The equivalent stress value of each mesh element is calculated using the von Mises yield criterion. When the stress exceeds the material yield strength, the plastic deformation stage begins. The strain hardening coefficient is obtained through the stress-strain relationship curve. A normal pressure load is applied to the interface between the stamping die and the aluminum sheet to establish a stress field distribution matrix. Using the stress field distribution matrix as the load condition, the node displacement value is solved using the Newton-Raphson iteration method. The element strain tensor is calculated based on the displacement gradient. If the element distortion rate exceeds the threshold, the element is subdivided and new nodes are interpolated to obtain the spatial distribution data of the deformation field. Based on the spatial distribution data of the deformation field, the volumetric strain is calculated by the volume ratio of the element before and after deformation. Under the assumption of material incompressibility, the strain value in the thickness direction is derived based on Poisson's ratio. The thickness strain value of all elements in each segmented region is multiplied by the initial thickness and statistically analyzed to determine the thickness variation distribution data. Furthermore, a normal pressure load is applied to the interface between the stamping die and the aluminum sheet to establish a stress field distribution matrix. This includes: solving for the nodal internal force values ​​using static equilibrium equations, transferring stress layer by layer along the load direction and calculating the attenuation amount to obtain the stress field distribution matrix. Further, using the stress field distribution matrix as the load condition, the thickness strain value is derived. The thickness variation distribution data is determined by multiplying the thickness strain values ​​of all elements in each segmented region by the initial thickness and then statistically analyzing the results. This includes: using the stress field distribution matrix as the load condition, solving for the nodal displacement values ​​using the Newton-Raphson iterative method, calculating the element strain tensor based on the displacement gradient, calculating the volumetric strain using the volume ratio before and after element deformation, deriving the thickness strain value based on Poisson's ratio under the assumption of material incompressibility, and then statistically analyzing the thickness variation distribution data by multiplying the thickness strain values ​​of all elements in each segmented region by the initial thickness.

[0005] Furthermore, if the fluctuations in the segment boundary regions of the simulated thickness variation distribution data exceed a preset threshold, the thinning processing parameters, such as the magnitude and direction of the force, are adjusted to obtain an optimized parameter set, including: Thickness value sequences of each segment boundary region are extracted from the thickness variation distribution data. The absolute value of the thickness difference between adjacent nodes is calculated, and the fluctuation degree value is obtained through the standard deviation. Boundary regions exceeding the threshold are marked as regions to be optimized. The punching force adjustment and angle correction values ​​are calculated based on the deviation between the fluctuation degree value and the target value. Orthogonal experimental design is used to generate a set of candidate parameters. Finite element simulation is re-executed for each set of candidate parameters to calculate the predicted thickness fluctuation value. It is determined whether the difference with the previous iteration is less than the convergence threshold. If it is satisfied, the current parameter is recorded as the optimal parameter set. Otherwise, the iteration continues until the optimized parameter set is determined.

[0006] Furthermore, stress imbalance indices corresponding to minimum thickness fluctuations are extracted from the optimized parameter set based on stress gradient analysis, including: Stress distribution data for each segmented region is extracted from the optimized parameter set. The difference between the maximum and minimum principal stresses between adjacent regions is calculated as the stress gradient value. The stress concentration coefficient is determined by the ratio of the stress gradient value to the average stress, and the stress imbalance index is calculated based on the deviation. Further, a genetic algorithm is used to globally search for the stress imbalance index to plan a segmented stamping processing path that balances the stress distribution. This includes: constructing chromosome encoding based on the stress imbalance index, where each chromosome contains a processing sequence and a load value sequence; evaluating the degree of stress balance using a fitness function; generating offspring using roulette wheel selection and two-point crossover; decoding the processing path from the offspring chromosomes; sequentially applying loads and calculating the stress superposition effect; correcting subsequent loads based on the residual stress of the preceding sequence; randomly mutating gene values ​​if the yield strength is exceeded to obtain a candidate path set; performing virtual stamping from the candidate path set; selecting the path with the smallest thickness fluctuation standard deviation as the optimal solution; and generating a segmented stamping processing path plan.

[0007] Furthermore, regarding processing path planning, by comparing the differences between the structural stability index of the overall impact assessment data and the structural strength parameters of the initial geometric model, an interference-free segmented thinning execution sequence is determined, including: An overall stiffness matrix is ​​constructed, and after applying boundary conditions, the minimum eigenvalue is calculated through eigenvalue decomposition as a structural stability index. The ratio of the structural stability index to the original yield strength is calculated as a safety factor, and risk level segments are divided into sets. A directed graph of stress propagation is constructed, where nodes are segmented regions and edge weights represent the degree of stress influence. The processing sequence with the minimum cumulative residual stress is determined through topological sorting, and an interference-free segmented thinning execution sequence is generated.

[0008] Furthermore, based on the determined thinning execution sequence, a virtual verification simulation is performed on the metal material to determine whether the thickness of each segment region and its surrounding region remains within the allowable deviation range of the original thickness value, thus obtaining the final aluminum sheet local thinning scheme data, including: Apply stamping loads sequentially according to the thinning execution sequence, calculate the material deformation state at each step using explicit dynamics methods, extract the thickness direction displacement component to obtain the current thickness value, extract the thickness values ​​at the center point and boundary point of each segment, compare them with the initial thickness to calculate the thickness change rate, and determine whether it exceeds the tolerance limit; if all requirements are met, integrate the processing parameters of each segment with the expected thickness distribution map to form the final aluminum sheet local thinning scheme data.

[0009] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses a method for monitoring the thickness gradient distribution in segmented stamping, addressing the unique business scenario of thickness gradient control and structural stability maintenance in the local thinning process of aluminum materials. Specifically, during segmented stamping, excessive thickness fluctuations and stress imbalances in the segment boundary regions due to uneven force distribution pose a risk of overall structural deformation and instability. This method addresses this issue by pre-extracting initial thickness data and overall parameters from a three-dimensional geometric model, meshing it according to thickness gradient requirements, and using finite element analysis to simulate the mechanical behavior of stamping formation. It calculates the stress and deformation fields to obtain the thickness variation distribution data for each segment, thus integrating thickness fluctuation assessment and stress gradient analysis. If the fluctuation exceeds a threshold, iterative adjustments are made to thinning parameters such as force magnitude and direction to optimize the parameter set until thickness fluctuations are minimized. Stress imbalance indices are extracted from these parameters, and a genetic algorithm is used to globally search for segmented stamping processing paths that balance stress distribution. Furthermore, structural mechanics analysis is used to compare the stability indices with the initial structural strength parameters, determining an interference-free execution sequence based on the stress propagation path. Finally, virtual verification simulations are performed to ensure that the thickness deviation in each region is within acceptable limits, ultimately generating reliable data for local thinning schemes of aluminum sheets. The technical advantages of this invention are that it significantly improves the precision and structural integrity of the thinning process, reduces material waste and optimizes manufacturing efficiency, and achieves coordinated control of uniform thickness gradient distribution and stress balance. Attached Figure Description

[0010] Figure 1 This is a flowchart of a segmented stamping thickness gradient distribution monitoring method according to the present invention. Detailed Implementation

[0011] The technical solutions of the present invention will be clearly and thoroughly described below with reference to the accompanying drawings. The described embodiments are merely some embodiments of the present invention.

[0012] like Figure 1 This embodiment of a segmented stamping thickness gradient distribution monitoring method may specifically include: Step S101: Extract the initial thickness data and overall structural parameters of the local area from the pre-established three-dimensional geometric model of the aluminum material, obtain the boundary conditions of the thinning target area, divide the aluminum sheet material into segments according to the thickness gradient requirements, and obtain a meshed representation.

[0013] Vertex coordinates and normal vector data are read from the STL file format of the 3D geometric model of the aluminum material. The spatial boundary contour of the target thinning region is identified using a convex hull algorithm. The positional relationship between points and polygons is used to determine whether each vertex is located inside the target region, obtaining initial boundary condition data containing a set of boundary points and region identifiers. For this initial boundary condition data, the Delaunay triangulation method is used to mesh the aluminum sheet surface. The thickness gradient value of each mesh node is calculated based on the distance from the center of the target region to the boundary. A thickness transition relationship is established between adjacent nodes using linear interpolation, resulting in a triangular mesh structure with thickness attributes. Based on the thickness value of each node in the triangular mesh structure, the thickness difference between adjacent nodes is calculated, and a thickness change threshold is set. When the thickness difference exceeds the threshold, a segmented boundary line is set at the corresponding position. A meshed representation for subsequent simulations is generated based on the set of nodes in each region divided by the segmented boundary lines.

[0014] In one implementation, when reading vertex coordinates and normal vector data from an STL file format of a three-dimensional geometric model of aluminum material, the STL file uses a triangular patch representation, where each triangular patch contains the three-dimensional coordinate values ​​of three vertices and a unit normal vector. By traversing all the triangular patches in the file, the complete geometric information of the aluminum sheet surface is extracted, forming a point cloud data set.

[0015] Specifically, when identifying the spatial boundary contour of the target thinning area, the convex hull algorithm first performs a two-dimensional projection on the extracted point cloud data, projecting the three-dimensional coordinate points onto the main plane of the aluminum sheet. A two-dimensional convex hull is constructed using the Graham scan method, starting from the bottom left corner and connecting the outer points in a counter-clockwise direction to form a closed boundary contour. For each vertex, its positional relationship with the boundary contour is determined using the ray method. Rays are emitted from the vertex in any direction, and the number of intersections between the rays and the boundary contour is counted. An odd number of intersections indicates the point is inside, and an even number indicates the point is outside. During the meshing process of the aluminum sheet surface using the Delaunay triangulation method, the empty circle criterion is followed, meaning that the circumcircle of any triangle does not contain other vertices. This meshing method avoids excessively long and narrow triangles, ensuring mesh quality. When calculating the thickness gradient value of each mesh node, the Euclidean distance from each node to the geometric center of the target area is measured. According to a preset thickness variation function, the distance value is mapped to the corresponding thickness value. Nodes closer to the target area are assigned smaller thickness values, while nodes farther away retain their original thickness. When establishing a thickness transition relationship between adjacent nodes using linear interpolation, the thickness value of any point inside the triangle is calculated using the centroid coordinate method. That is, the thickness values ​​of the three vertices are weighted and averaged according to the distance weights from the point to the three vertices to achieve a smooth thickness transition.

[0016] In one embodiment, the thickness difference is calculated for the two endpoints of each edge. When the thickness difference between adjacent nodes exceeds a preset threshold, a new node is inserted at the midpoint of that edge as a segment boundary point. By connecting all segment boundary points, segment boundary lines are formed that divide the aluminum sheet into different processing areas.

[0017] Preferably, when generating the meshed representation of the node set of each region divided by the segmented boundary line, a unique region identifier is assigned to each region, and the index number, coordinate value and thickness attribute of all nodes in the region are recorded. This data structure is convenient for subsequent finite element analysis to call directly, realizing a seamless connection from the geometric model to the simulation model.

[0018] Step S102: Based on the meshed representation, the finite element analysis method is used to numerically simulate the mechanical behavior of stamping during the segmented stamping process, and to determine the thickness variation distribution data of each segment region after simulation.

[0019] Based on the node coordinates and thickness attributes in the meshed representation, a constitutive relation matrix for the aluminum sheet material is constructed. The von Mises yield criterion is used to calculate the equivalent stress value of each mesh element. When the equivalent stress exceeds the material's yield strength, the plastic deformation stage is determined. The strain hardening coefficient of each element is obtained through the stress-strain relationship curve. Based on the strain hardening coefficient, a normal pressure load is applied to the interface between the stamping die and the aluminum sheet. A load distribution function is established according to the contact point location and pressure magnitude. The internal force values ​​of each node are solved using the static equilibrium equation. The stress is transferred layer by layer along the load direction, and the attenuation is calculated to obtain the stress field distribution matrix of each segmented region. Using the stress field distribution matrix as the load condition, the displacement values ​​of each mesh node are solved using the Newton-Raphson iteration method. The element strain tensor is calculated based on the displacement gradient. If the element distortion rate exceeds a preset threshold, the element is subdivided, and the physical quantities of the new node are interpolated to obtain the spatial distribution data of the deformation field. Based on the spatial distribution data of the deformation field, the volumetric strain is calculated by the volume ratio of the element before and after deformation. Under the assumption of material incompressibility, the strain value in the thickness direction is derived based on the Poisson's ratio. The thickness strain value of all elements in each segment region is multiplied by the initial thickness and statistically analyzed to determine the thickness variation distribution data of each segment region after simulation.

[0020] In one implementation, a power-law hardening model is used to describe the elastoplastic behavior of aluminum sheet materials when constructing the constitutive relation matrix. The constitutive relation matrix includes four key parameters: elastic modulus, Poisson's ratio, yield strength, and hardening exponent. For aerospace aluminum alloys, the elastic modulus is typically 70 GPa, the Poisson's ratio is 0.33, and the yield strength varies from 200 to 400 MPa depending on the alloy grade. The von Mises yield criterion determines whether the material has entered the plastic deformation stage by calculating the quadratic invariant of the principal stress difference; its equivalent stress calculation formula involves a combination of the three principal stress components. When the equivalent stress value exceeds the initial yield strength of the material, the material transitions from elastic deformation to plastic deformation.

[0021] Specifically, the strain hardening coefficient is obtained based on the slope of the true stress versus true strain curve. During the plastic deformation stage, the flow stress of the material increases with the accumulation of plastic strain; this phenomenon is called strain hardening. The hardening curve obtained by fitting uniaxial tensile test data can reflect the degree of strengthening of the material at different strain levels. For aluminum alloy sheets, the hardening index n is typically between 0.15 and 0.25, and the strength coefficient K is in the range of 400-600 MPa. In establishing the load distribution function, the contact interface between the stamping die and the aluminum sheet is discretized into multiple contact element pairs. Each contact pair contains a master surface element and a slave surface element; the master surface is typically the rigid die surface, and the slave surface is the deformable aluminum sheet surface. The normal pressure load exhibits a non-uniform distribution within the contact area according to the die geometry and stamping depth. In the initial stage of stamping, the load is mainly concentrated in the central region of the aluminum sheet; as the stamping depth increases, the load gradually diffuses towards the edges. The local contact pressure is calculated using Hertzian contact theory, and the pressure value exhibits a specific functional relationship with the distance from the contact point to the die center. The static equilibrium equations are solved using the stiffness matrix method, assembling the overall equilibrium equations into a form where the nodal force vector equals the product of the stiffness matrix and the displacement vector. Stress transmission within the material follows Saint-Venant's principle: the stress value decreases exponentially with increasing distance from the load point, and the attenuation coefficient is related to the material's elastic modulus and thickness.

[0022] For example, the Newton-Raphson iterative method transforms the nonlinear problem into a series of linear problems by linearizing the solution of nonlinear finite element equations. In each iteration, the unbalanced force is calculated based on the current displacement state, and then the linear equations are solved to obtain the displacement increment. The nodal displacements are updated until the unbalanced force is less than the convergence tolerance. The displacement gradient tensor is obtained by dividing the difference in displacement between adjacent nodes by the element characteristic length, reflecting the degree of element deformation.

[0023] In one embodiment, the Jacobian determinant value is used as an indicator to determine the element distortion rate. When the Jacobian value of an element is less than 0.3, it indicates that the element has severe distortion and requires mesh subdivision. The subdivision process inserts new nodes at the midpoints of the edges and the centers of the faces of the distorted elements, subdividing a quadrilateral element into four sub-elements. The physical quantities of the new nodes are calculated through shape function interpolation to ensure the continuity of the field variables.

[0024] Preferably, the spatial distribution data of the deformation field includes the displacement components of each node in three directions and the strain tensor components of each element. This data is stored in matrix form, with row indices corresponding to node or element numbers and column indices corresponding to different components.

[0025] Understandably, the calculation of volumetric strain is based on the ratio of the Jacobian determinant before and after element deformation. Under the assumption of small deformation, volumetric strain equals the sum of the three principal strains. For incompressible materials, volumetric strain is zero, meaning the sum of the three principal strains is zero. According to the generalized Hooke's law and Poisson's ratio, when the strains in two in-plane directions are known, the strain in the thickness direction can be derived. The specific derivation process is as follows: let the in-plane strains be ε1 and ε2, and the thickness direction strain be ε3, based on the condition of constant volume... ,get Furthermore, the thickness variation value is obtained by multiplying the strain in the thickness direction by the initial thickness. For each segmented region, the thickness variation values ​​of all elements within that region are statistically analyzed, and the average, standard deviation, maximum, and minimum values ​​are calculated to form the thickness variation distribution characteristics of that region. These statistical data can intuitively reflect the degree of influence of the stamping process on the thickness of different regions.

[0026] Step S103: If the fluctuation of each segment boundary region in the simulated thickness change distribution data exceeds the preset threshold, then adjust the thinning processing parameters such as the magnitude and direction of the force to obtain the optimized parameter set.

[0027] Thickness value sequences of each segment boundary region are extracted from the thickness variation distribution data. The absolute value of the thickness difference between adjacent nodes is calculated, and the fluctuation degree value is obtained by calculating the standard deviation of the thickness difference. If the fluctuation degree value exceeds a preset threshold, the boundary region is marked as a region to be optimized, resulting in a list of regions to be optimized containing region numbers and fluctuation degree values. For each region in the list of regions to be optimized, the impact force adjustment amount is calculated based on the deviation between its fluctuation degree value and the target value. The force application angle correction value is determined by the skewness of the fluctuation distribution. Different force magnitude and direction parameters are combined using an orthogonal experimental design method to generate a candidate parameter set. Finite element simulation is re-executed for each set of parameters in the candidate parameter set to calculate the predicted thickness fluctuation value of each boundary region under the new parameter conditions. It is determined whether the difference between the predicted thickness fluctuation value and the result of the previous iteration is less than the convergence threshold. If the convergence condition is met, the current parameter combination is recorded as the optimal parameter set; otherwise, the parameters are adjusted according to the fluctuation trend, and the iteration continues to determine the optimized parameter set.

[0028] In one implementation, when extracting the thickness value sequence of each segment boundary region from the thickness variation distribution data, each boundary region contains multiple sampling points distributed along the boundary line. The local gradient value of the thickness variation is obtained by performing a difference operation on the thickness values ​​of adjacent sampling points. The standard deviation is calculated using the Bessel correction formula to reflect the dispersion of the thickness distribution.

[0029] Specifically, the preset threshold is determined based on the allowable thickness tolerance range of the aluminum sheet material. For aerospace aluminum alloy thin plates, the thickness tolerance is usually controlled within ±5% of the initial thickness. When the fluctuation value of a certain boundary area exceeds this threshold, it indicates that the thickness uniformity of that area does not meet the processing requirements and parameter optimization is required. The list of areas to be optimized records the spatial location, fluctuation value, and influence range of each abnormal area. The calculation of the punching force adjustment follows the assumption of a linear relationship between force and deformation. Within the elastic-plastic deformation range, the thickness reduction is proportional to the applied pressure, and the proportionality coefficient depends on the hardening characteristics of the material. By establishing a deviation function between the fluctuation value and the target value, the required pressure correction can be calculated in reverse. The determination of the force application angle correction value is based on the spatial characteristics of the fluctuation distribution. When the thickness fluctuation shows a directional bias, the stress distribution pattern is changed by adjusting the force application angle. The orthogonal experimental design adopts the Taguchi method, and the L9 orthogonal array is selected for parameter combination design. The punching force is divided into three levels, namely 90%, 100%, and 110% of the original value; the force application angle is also set to three levels, namely normal offset. , and Nine sets of experiments were conducted to cover the main parameter combination space, thereby obtaining the parameter influence patterns while ensuring experimental efficiency.

[0030] For example, the iterative process of the finite element simulation uses an incremental step method for solution. The simulation under each set of parameter conditions includes multiple load increment steps, gradually applying the stamping load until the target depth is reached. After the simulation is completed, the thickness distribution data of each boundary region is extracted, and the fluctuation value is recalculated.

[0031] In one embodiment, the convergence threshold is set to 1% of the initial fluctuation value. The optimization process is considered to have converged when the change in fluctuation value between two consecutive iterations is less than this threshold. If the convergence condition is not met, the next set of experimental parameters is determined in the parameter space using the golden section search method based on the difference between the current fluctuation value and the target value.

[0032] Preferably, the optimal parameter set includes the magnitude of the impact force, the angle of force application, and the number of iterations required to achieve convergence for each segmented region.

[0033] Step S104: Extract the stress imbalance index corresponding to the minimum thickness fluctuation from the optimized parameter set based on stress gradient analysis, and use a genetic algorithm to perform a global search on the stress imbalance index to plan the segmented stamping process path with balanced stress distribution.

[0034] Stress distribution data for each segment region is extracted from the optimized parameter set. The difference between the maximum and minimum principal stresses between adjacent regions is calculated as the stress gradient value. The stress concentration coefficient is determined by the ratio of the stress gradient value to the average stress of each region. The stress imbalance index is calculated based on the deviation of the stress concentration coefficient from the target uniform value, resulting in an index matrix containing region numbers and imbalance degrees. Chromosome encoding for a genetic algorithm is constructed based on the index matrix. Each chromosome contains the processing sequence sequence of the segment region and the corresponding load value sequence. The quality of chromosomes is evaluated using a fitness function. The fitness value is calculated based on the stress balance degree of the encoded path. A roulette wheel is used to select individuals with high fitness, and a progeny population is generated through two-point crossover. The processing path is decoded for each chromosome in the progeny population. The loads of each segment are applied sequentially according to the path order, and the stress superposition effect is calculated. The load parameters are corrected based on the residual stress effect of the previous processing on the subsequent regions. If the stress in a certain region exceeds the material yield strength, the gene value at that position is adjusted through random mutation to obtain a set of candidate paths that meet the strength constraints. Virtual stamping is performed from the candidate path set according to the processing order of each path. The thickness distribution of each region after execution is calculated and the standard deviation is taken as the thickness fluctuation value. The path with the smallest fluctuation value is selected as the optimal solution. A segmented stamping processing path plan is generated according to the processing order and load parameters of the optimal path.

[0035] In one implementation, the calculation of the stress gradient involves a precise analysis of the internal stress state of the aluminum sheet material. The maximum and minimum principal stresses are obtained by eigenvalue decomposition of the stress tensor, where the maximum eigenvalue corresponds to the maximum principal stress and the minimum eigenvalue corresponds to the minimum principal stress. The stress gradient between adjacent regions reflects the spatial rate of change of the stress field; a larger gradient value indicates a more uneven stress distribution. The stress concentration factor is quantified by the ratio of local stress peaks to average stress; when this factor exceeds 1.5, it indicates the presence of significant stress concentration.

[0036] Specifically, the stress imbalance index is calculated using a weighted average method, assigning different weights to each region based on its area and location importance. Regions near the center of the aluminum sheet are given higher weights due to their greater impact on overall structural stability. The index matrix is ​​stored as a two-dimensional array, with rows representing different segmented regions and columns containing four attributes: region number, center coordinates, imbalance degree value, and weight coefficient. The chromosome encoding design of the genetic algorithm directly determines the structure of the search space and the algorithm's efficiency. Each chromosome employs a two-layer encoding structure: the first layer encodes the processing order, using integers to represent the processing sequence of each segmented region; the second layer encodes the load parameters, using real values ​​to represent the impact pressure magnitude of the corresponding region. The chromosome length is equal to twice the total number of segmented regions. The fitness function design comprehensively considers both stress balance and processing efficiency. Stress balance is measured by calculating the coefficient of variation of stress values ​​across all regions; a smaller coefficient of variation indicates a more uniform stress distribution. Processing efficiency is evaluated based on the total path length and the number of mold changes. Fitness is calculated by multiplying stress balance weight by the balance score, plus efficiency weight multiplied by the efficiency score. The sum of these two weights is 1, typically set to 0.7 and 0.3. Roulette wheel selection determines the selection probability based on the proportion of an individual's fitness value to the total population fitness; individuals with higher fitness have a greater probability of being selected. Two-point crossover randomly selects two crossover points on the chromosome, exchanging gene segments from the parent chromosomes between these two points. Maintaining the legality of the processing order requires a partially mapped crossover strategy.

[0037] For example, the path decoding process converts chromosomes into specific processing schemes. Following the processing sequence determined by sequential encoding, corresponding loads are applied to each region in turn. Residual stresses generated by previous processing affect the stress state of subsequent regions; this effect is calculated using the principle of stress superposition.

[0038] In one embodiment, the influence range of residual stress is described using an exponential decay model, where the influence of residual stress decreases with distance from the processing point. The decay coefficient is determined based on the material's elastic modulus and Poisson's ratio. When the cumulative stress in a certain region approaches 90% of the material's yield strength, the load parameters at that location need to be adjusted.

[0039] Preferably, the random mutation operation employs a Gaussian mutation strategy, adding a normally distributed random perturbation to the original gene values. The mutation amplitude gradually decreases with the number of iterations, initially setting the standard deviation of the mutation to 10% of the parameter range, and later reducing it to 1%.

[0040] Understandably, virtual stamping simulation rapidly evaluates path effectiveness based on a simplified mechanical model. By establishing a load and deformation response database, interpolation methods are used to predict thickness distribution under different load conditions. Thickness fluctuation values ​​are obtained by calculating the standard deviation of thickness values ​​at all measurement points; a smaller standard deviation indicates a more uniform thickness distribution. Furthermore, the selection of the optimal path not only considers minimizing thickness fluctuation but also meets production efficiency requirements. A multi-objective optimization function is set to find a balance between thickness uniformity and processing time.

[0041] For example, for an aluminum sheet comprising 12 segmented regions, a genetic algorithm with a population size of 100 typically converges to a satisfactory solution after 200 generations of iteration. Optimal path planning includes a detailed processing sequence table, a load parameter table for each region, and a map showing the expected thickness distribution.

[0042] Step S105: For the processing path planning, by comparing the differences between the structural stability index of the overall impact assessment data and the structural strength parameters of the initial geometric model, an interference-free segmented thinning execution sequence is determined.

[0043] To construct the overall stiffness matrix of the aluminum sheet structure for processing path planning, the structure is divided into multiple elements using the finite element method (FEM). The stiffness matrices of each element are assembled to obtain the overall stiffness matrix. After applying boundary conditions, the nodal displacements are solved. The minimum eigenvalue of the overall stiffness matrix is ​​calculated using eigenvalue decomposition as a structural stability index. Original structural strength parameters, including yield strength, elastic modulus, and ultimate strength, are extracted from the initial geometric model. The ratio of the structural stability index to the original yield strength is calculated as a safety factor. If the safety factor is lower than a preset threshold, the region is marked as a high-risk region. Segment sets with different risk levels are divided according to the safety factor. A directed stress propagation graph is constructed based on these segment sets, with each segment region serving as a node in the graph. If processing in region A will affect the stress in region B, a directed edge is established from A to B. The edge weight represents the degree of stress influence. The stress attenuation law with distance is calculated using material mechanics formulas to obtain the stress influence matrix between regions. Based on the stress influence matrix, regions with influence weight values ​​less than the preset interference threshold are identified. A topology sorting algorithm is used to determine the processing order that satisfies the dependency relationship, so that the cumulative residual stress experienced by each region during processing is minimized, and an interference-free segmented thinning execution sequence is generated.

[0044] In one implementation, constructing the overall stiffness matrix of the aluminum sheet structure requires first meshing the structure, discretizing the continuum into a finite number of elements. The stiffness matrix of each element reflects its resistance to deformation under stress. The element stiffness matrix in the local coordinate system is transformed to the global coordinate system through coordinate transformation, and then the elements are stacked and assembled according to the node numbering rules. For a four-node quadrilateral element, the element stiffness matrix is ​​an 8×8 symmetric matrix, while the dimension of the overall stiffness matrix depends on the total number of nodes.

[0045] Specifically, the eigenvalue decomposition process solves the generalized eigenvalue problem, with the minimum eigenvalue corresponding to the first buckling mode of the structure. When the minimum eigenvalue approaches zero, it indicates that the structure is close to instability. The physical meaning of the eigenvalue is the critical load factor required for the structure to buckle; the larger the eigenvalue, the better the stability of the structure. The first few minimum eigenvalues ​​of a large-scale sparse matrix can be efficiently solved using subspace iteration or the Lanzos method, avoiding the computational burden of calculating all eigenvalues. The structural stability index is expressed as the ratio of the minimum eigenvalue to a reference value, which is usually taken as the eigenvalue in the unprocessed state, thus intuitively reflecting the degree of influence of processing on structural stability. The calculation of the safety factor involves a comprehensive evaluation of multiple strength parameters. Yield strength characterizes the stress level at which the material begins to undergo plastic deformation, elastic modulus reflects the stiffness characteristics of the material, and ultimate strength is the maximum stress that the material can withstand before failure. The safety factor is defined as the ratio of allowable stress to working stress; in the processing of aluminum alloy thin sheets, a safety factor of not less than 1.5 is typically required.

[0046] For example, the construction of the directed stress propagation graph is based on Saint-Venant's principle and stress diffusion theory. When a concentrated load is applied to a region of an aluminum sheet, the stress diffuses outwards and gradually decays. The influence range is typically 3-5 times the characteristic size of the load-bearing region. The establishment of directed edges follows the principle of causality; if processing region A will generate residual stress in region B, then a directed edge is established from A to B. The edge weights are calculated using a stress influence function that considers factors such as distance decay, material properties, and geometry. The stress influence function uses an exponential decay form, with the decay rate related to the material's Poisson's ratio and elastic modulus. For aerospace aluminum alloys, the Poisson's ratio is approximately 0.33, and the stress influence decay with distance can be expressed as the initial value multiplied by a negative power of the natural exponent, where the coefficient of the power is proportional to the distance. By traversing and calculating all possible region pairs, a complete stress influence matrix is ​​formed, where the matrix elements represent the degree of stress influence of the row-indexed region on the column-indexed region.

[0047] In one embodiment, the preset interference threshold is determined based on the material's fatigue limit. When the residual stress is below 10% of the fatigue limit, it is considered that it will not significantly interfere with subsequent processing.

[0048] Preferably, the application of the topology sorting algorithm ensures that the processing order satisfies dependency constraints. In a directed graph of stress propagation, if a directed path exists from node A to node B, then A must be processed before B. Topology sorting is implemented through depth-first search or breadth-first search. First, all nodes with an in-degree of zero are identified as candidate starting nodes. The node with the least stress influence is selected to begin processing. Then, the connectivity of the graph is updated, and this process is repeated until all nodes are sorted.

[0049] Understandably, the calculation of cumulative residual stress needs to consider the stress superposition effect. Each completed processing area generates a residual stress field within its influence range, and the residual stress fields from multiple processed areas will superimpose. The combined stress at each point is calculated using the principle of vector superposition. When the cumulative stress at a point approaches the material's yield strength, subsequent processing parameters need to be adjusted or the processing sequence changed. Furthermore, the interference-free segmented thinning execution sequence must not only meet stress constraints but also consider processing efficiency. By introducing a path length penalty term, under the premise of meeting the interference-free condition, the scheme of continuous processing of adjacent areas is prioritized to reduce equipment movement time and tool changes. The execution sequence is output in the form of a processing instruction table, including the processing sequence number, spatial coordinates, process parameters, and expected completion time for each segment.

[0050] Step S106: Based on the determined thinning execution sequence, perform virtual verification simulation on the metal material to determine whether the thickness of each segment area and its surrounding area remains within the allowable deviation range of the original thickness value, and obtain the final aluminum sheet local thinning scheme data.

[0051] According to the determined thinning execution sequence, corresponding stamping loads are applied sequentially in the virtual simulation environment according to the processing order. The material deformation state after each processing step is calculated using explicit dynamic methods. The displacement components of each segment node in the thickness direction are extracted, and the current thickness value is obtained by subtracting the displacement from the initial thickness, thus obtaining a simulation result set containing spatial coordinates and thickness data. The thickness values ​​of the center point and boundary point of each segment are extracted from the simulation result set and compared with the initial thickness at the corresponding position in the original geometric model. The thickness change rate is calculated. If the thickness change rate exceeds the upper limit of the tolerance specified in the material standard, it is marked as a non-conforming area. The number of non-conforming areas is counted to determine whether the overall processing quality requirements are met. Based on the processing quality judgment results, if all areas meet the requirements, the processing parameters of each segment, including the magnitude of the stamping force, the position of application, and the execution order, are integrated to form aluminum sheet local thinning scheme data containing process number, process parameters, and expected thickness distribution map.

[0052] In one implementation, the virtual simulation environment is built using dedicated metal forming simulation software, which includes a material library, a tool library, and a boundary condition setting module. The explicit dynamics method integrates the motion equations over time using the central difference method, with the time step automatically determined based on the minimum element size and the material's sound velocity. Within each time step, nodal accelerations are calculated based on the current stress state, and nodal velocities and displacements are updated to obtain a new deformation configuration.

[0053] Specifically, extracting the thickness displacement component requires establishing a local coordinate system with the aluminum sheet surface normal as the z-axis. For each node, the cumulative displacement from the initial position to the current position is calculated by tracing its displacement history in the z-direction. The initial thickness value is read from the attribute data of the original geometric model, and the current thickness value is equal to the initial thickness minus the absolute value of the z-direction displacement. This calculation method considers the compressive deformation characteristics of the material and can accurately reflect the thickness change during the stamping process. The upper limit of the tolerance specified in the material standard is usually determined according to the application and grade of the aluminum sheet. For aerospace-grade aluminum alloy sheets, the thickness tolerance is generally controlled within ±5%. The formula for calculating the thickness change rate is the difference between the current thickness and the initial thickness divided by the initial thickness, then multiplied by 100% to obtain the percentage value.

[0054] For example, a graded evaluation mechanism is used to judge the processing quality. When the number of non-conforming areas is zero, it is judged as a superior product; when the number of non-conforming areas accounts for less than 10% of the total number of areas, it is judged as a qualified product; when it exceeds 10%, the process parameters need to be adjusted and the simulation needs to be repeated.

[0055] In one embodiment, the center point of the segmented region is determined by the geometric centroid method, while the boundary points are obtained by sampling at equal intervals along the region's contour line. The thickness value of each sampling point is calculated from the thickness data of surrounding nodes using an interpolation algorithm.

[0056] Preferably, the data for the partial thinning scheme of aluminum sheet is output in a structured document format, which includes three main parts: a process number table that records the processing sequence and time arrangement of each segment; a process parameter table that details the punching pressure, holding time, and pressing speed parameters of each process; and an expected thickness distribution map that visually displays the thickness distribution after processing in the form of a cloud map, with different colors representing different thickness ranges.

[0057] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method of monitoring a thickness gradient distribution of a section-pressed product, characterized by, The method comprises: extracting initial thickness data and overall structure parameters of a local area from a pre-established three-dimensional geometric model of a metal aluminum material, obtaining boundary conditions of a target area for thinning, segmenting and dividing the aluminum sheet material according to thickness gradient requirements to obtain a grid representation form; numerically simulating stamping forming mechanical behavior in the segmented stamping process according to the grid representation form by using a finite element analysis method, and determining thickness change distribution data of each segmented area after simulation; if the fluctuation of each segmented boundary area in the thickness change distribution data after simulation exceeds a preset threshold, adjusting the thinning processing parameters such as force size and direction to obtain an optimized parameter set; extracting a stress imbalance index corresponding to the minimum thickness fluctuation based on stress gradient analysis from the optimized parameter set, and using a genetic algorithm to globally search for a segmented stamping processing path planning of balanced stress distribution; for the processing path planning, determining an interference-free segmented thinning execution sequence by comparing the difference between the structure stability index of the overall influence evaluation data and the structure strength parameter of the initial geometric model; according to the determined thinning execution sequence, performing virtual verification simulation on the metal material, judging whether the thickness of each segmented area and its surrounding area remains within the original thickness value within the allowable deviation range, and obtaining final aluminum sheet local thinning scheme data.

2. The method of claim 1, wherein, The method comprises: obtaining vertex coordinate and normal vector data, identifying the spatial boundary contour of the target area for thinning by a convex hull algorithm, determining the vertex position according to the position relationship between the point and the polygon, and obtaining initial boundary condition data; the surface of the aluminum sheet is meshed by using a Delaunay triangulation method, the thickness gradient value of each grid node is calculated according to the distance from the center of the target area to the boundary, the thickness transition relationship between adjacent nodes is established by using a linear interpolation method, and a triangular mesh structure with thickness attribute is obtained; the thickness difference between adjacent nodes is calculated according to the thickness value of the node in the triangular mesh structure, and a thickness change threshold is set, a segmented boundary line is set at the position where the thickness difference exceeds the threshold, and a grid representation form is generated.

3. The method of claim 1, wherein: The method comprises: According to the node coordinates and thickness attribute in the grid representation, a constitutive relation matrix is constructed, the equivalent stress value of each grid element is calculated by using the von Mises yield criterion, and when the equivalent stress value exceeds the yield strength of the material, the plastic deformation stage is entered, and the strain hardening coefficient is obtained through the stress-strain relationship curve; a normal pressure load is applied to the contact interface between the stamping die and the aluminum sheet, and a stress field distribution matrix is established; the stress field distribution matrix is used as a load condition, and the node displacement value is solved by using the Newton-Raphson iteration method, the element strain tensor is calculated according to the displacement gradient, if the element distortion rate exceeds the threshold value, the element is subdivided and the new node is interpolated, and the spatial distribution data of the deformation field is obtained; according to the spatial distribution data of the deformation field, the volume strain is calculated by the volume ratio before and after the element deformation, and the thickness direction strain value is derived according to the Poisson's ratio under the assumption that the material is incompressible, and the thickness change distribution data is determined by multiplying the thickness strain value of all elements in each segmented region by the initial thickness and then counting.

4. A method of monitoring the distribution of thickness gradients in a segmented stamping according to claim 3, wherein, The normal pressure load is applied to the contact interface between the stamping die and the aluminum sheet, and the stress field distribution matrix is established, including: solving the node internal force value by using the static equilibrium equation, transferring the stress layer by layer along the load direction and calculating the attenuation amount to obtain the stress field distribution matrix.

5. The method of claim 3, wherein the thickness gradient distribution is monitored by a segmental press. 5 The stress field distribution matrix is used as a load condition, the thickness direction strain value is derived, the thickness strain value of all elements in each segmented region is multiplied by the initial thickness and then counted to determine the thickness change distribution data, including: the stress field distribution matrix is used as a load condition, the node displacement value is solved by using the Newton-Raphson iteration method, the element strain tensor is calculated according to the displacement gradient, the volume strain is calculated by the volume ratio before and after the element deformation, the thickness direction strain value is derived according to the Poisson's ratio under the assumption that the material is incompressible, and the thickness change distribution data is determined by multiplying the thickness strain value of all elements in each segmented region by the initial thickness and then counting.

6. The method of claim 1, wherein: If the fluctuation of each segmented boundary region in the simulated thickness change distribution data exceeds the preset threshold value, the thinning processing parameters such as the size and direction of the force are adjusted to obtain an optimized parameter set, including: From the thickness change distribution data, the thickness value sequence of each segmented boundary region is extracted, the absolute value of the thickness difference of adjacent nodes is calculated, the fluctuation degree value is obtained by the standard deviation, and the boundary region exceeding the threshold value is marked as a region to be optimized; the stamping force adjustment amount and the angle correction value are calculated according to the deviation of the fluctuation degree value and the target value, and a candidate parameter set is generated by using orthogonal experimental design; for each candidate parameter, the finite element simulation is re-executed, the thickness fluctuation prediction value is calculated, and it is judged whether the difference from the previous iteration is less than the convergence threshold value; if it is satisfied, the current parameter is recorded as the optimal parameter set, otherwise the iteration is continued until the optimized parameter set is determined.

7. The method of claim 1, wherein: The stress imbalance index corresponding to the minimum thickness fluctuation is extracted from the optimized parameter set based on stress gradient analysis, including: The stress distribution data of each segmented region is extracted from the optimized parameter set, the difference between the maximum principal stress and the minimum principal stress between adjacent regions is calculated as the stress gradient value, the stress concentration coefficient is determined by the stress gradient value and the average stress ratio, and the stress imbalance index is calculated according to the deviation.

8. The method of claim 1, wherein: The segmented stamping processing path planning of the balanced stress distribution searched globally by the genetic algorithm on the stress imbalance index comprises: constructing a chromosome code based on the stress imbalance index, each chromosome containing a processing sequence and a load value sequence, evaluating the stress balance degree by an adaptability function, and generating offspring by roulette selection and two-point crossover; decoding the offspring chromosome into a processing path, sequentially applying a load and calculating a stress superposition effect, correcting the subsequent load according to the previous residual stress, and randomly mutating the gene value if the yield strength is exceeded, to obtain a candidate path set; performing virtual stamping from the candidate path set, selecting the path with the minimum thickness fluctuation standard deviation as the optimal solution, and generating the segmented stamping processing path planning.

9. The method of claim 1, wherein: The processing path planning is determined by comparing the difference between the structural stability index of the overall influence evaluation data and the structural strength parameter of the initial geometric model, including: The overall stiffness matrix is constructed, and the minimum eigenvalue is calculated as the structural stability index by eigenvalue decomposition after applying the boundary conditions; the safety factor is calculated as the ratio of the structural stability index to the original yield strength, and the risk level segmentation set is divided; the stress propagation directed graph is constructed, the nodes are segmented areas, and the edge weight is the stress influence degree; the processing sequence with the minimum cumulative residual stress is determined by topological sorting to generate the non-interference segmented thinning execution sequence.

10. The method of claim 1, wherein: The virtual verification simulation of the metal material according to the determined thinning execution sequence judges whether the thickness of each segmented area and its surrounding area remains within the original thickness value of the allowable deviation range, and obtains the final aluminum sheet local thinning scheme data, including: According to the thinning execution sequence, the stamping load is sequentially applied, the material deformation state at each step is calculated by the explicit dynamics method, and the current thickness value is obtained by extracting the thickness direction displacement component; the thickness values of the center point and the boundary point of each segment are extracted, and the thickness change rate is calculated by comparing with the initial thickness to judge whether it exceeds the upper limit of the tolerance; if all requirements are met, the segmented processing parameters and the expected thickness distribution map are integrated to form the final aluminum sheet local thinning scheme data.

Images