A method and system for evaluating the structural strength of an energy storage device compartment

Abstract

The present application relates to the field of finite element analysis, and more particularly to a method and system for evaluating the structural strength of an energy storage device cabin, the method comprising: obtaining a finite element mesh of the energy storage device cabin, extracting corner points as sampling points to form a discrete point set; calculating the initial normal vector of each sampling point to obtain the structural curvature degree of each sampling point; obtaining the local element stiffness matrix of the mesh element, calculating the diagonal deviation degree to determine the degradation reference value, and combining the structural curvature degree to calculate the matrix deterioration degree; solving the global algebraic equation set to obtain the global relative force residual value, combining the matrix deterioration degree to calculate the confidence of the evaluation; adaptively adjusting the finite element mesh according to the confidence and the matrix deterioration degree and solving, and outputting the evaluation result. The present application accurately locates the poor mesh element and performs redrawing, effectively eliminates the false positive stress warning caused by mesh distortion, and improves the accuracy and reliability of the evaluation without causing a surge in computing power.

Description

Technical Field

[0001] This invention relates to the field of finite element analysis. In particular, it relates to a method and system for assessing the structural strength of an energy storage device compartment. Background Technology

[0002] As the core carrier of battery modules and electrical systems, the energy storage compartment must withstand complex dynamic and static loads during service. To ensure its structural safety, the industry typically uses the finite element analysis (FEA) method for mechanical simulation calculations. This method mainly involves discretizing the complex compartment structure into multiple finite element mesh elements, extracting the element node information to establish the local stiffness equations for each element, and then assembling them into a global algebraic equation system for numerical solution to evaluate the mechanical response and structural strength of the energy storage compartment.

[0003] For example, Chinese patent publication CN116757026B discloses a method for ultimate strength analysis of plate frame structures based on isogeometric analysis. This method discloses a mechanical simulation data processing flow that uses numerical discretization to assemble local stiffness matrices and solve for the global ultimate strength. However, in practical engineering applications, energy storage device cabins often have extremely complex topological morphologies, such as numerous heat dissipation openings, fixed lugs, welded frames, and other local structures on the cabin surface. Existing general finite element numerical discretization preprocessing methods typically rely on the mapping and segmentation of regular geometric blocks. When faced with the aforementioned non-uniform, anisotropic, and truly abrupt geometric boundaries, the boundaries of regular mesh elements cannot adaptively and perfectly fit.

[0004] In order to forcibly enclose and fit these complex discontinuous boundaries, existing preprocessing systems are prone to excessive stretching, compression or distortion of local mesh elements, resulting in a large amount of mesh distortion at abrupt structural changes (such as the edges of openings or the acute angles of ear pieces).

[0005] Distortions in geometric features and loss of key topological features lead to severe mesh distortion, which in turn causes the local stiffness matrix of the corresponding finite element to deteriorate. When elements containing such ill-conditioned matrices are assembled into the global algebraic equations, it will directly cause the error in the numerical solution process to accumulate sharply, which will eventually lead to the overall calculation being difficult to converge or outputting severely distorted stress assessment results. Summary of the Invention

[0006] To address the problem that traditional finite element methods suffer from mesh distortion due to mechanical partitioning and forced fitting when dealing with complex topological boundaries of energy storage compartments, which in turn leads to the accumulation of numerical errors in subsequent algebraic solutions and distortion of strength assessments, this invention provides solutions in the following aspects.

[0007] In the first aspect, a method for assessing the structural strength of an energy storage device compartment includes: The finite element mesh of the energy storage device compartment is obtained, and all the mesh elements in the finite element mesh are obtained based on the finite element mesh. The corner points of each mesh element are used as sampling points to form a discrete point set. The initial normal vectors of each sampling point in the discrete point set and its adjacent local neighborhood sampling points are obtained to obtain the structural curvature of the sampling points; The local element stiffness matrix of each grid cell is obtained based on the three-dimensional coordinates of each sampling point in each grid cell, and the diagonal deviation of each grid cell is obtained based on the local element stiffness matrix of each grid cell, and the degradation reference value of each grid cell is determined. The matrix degradation degree of each grid cell is obtained by using the degradation reference value of each grid cell and the structural curvature of all sampling points in each grid cell; Based on the global relative force residual value obtained from the global algebraic equation system constructed from the stiffness matrices of each local element, and the matrix deterioration degree of each grid element, the confidence level of the structural strength assessment of the energy storage device cabin is calculated. When the confidence level of the structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, the maximum local stress value of all grid cells in the finite element mesh is obtained, and the structural strength of the energy storage device cabin is evaluated based on the maximum local stress value of all grid cells.

[0008] Preferably, the initial normal vectors of each sampling point in the discrete point set and its adjacent local neighborhood sampling points include: For any sampling point, a nearest neighbor topology search is performed based on the topological connection relationship of each grid cell in the finite element mesh to obtain at least four sampling points that are topologically adjacent to the sampling point through the grid edge, forming a local neighborhood point set. Each sampling point in the local neighborhood point set is a local neighborhood sampling point. In the set of local neighborhood points, first, second, third, and fourth local neighborhood sampling points that are not collinear are selected, and a first set of planar vectors is constructed using these sampling points with the first and second local neighborhood sampling points, and a second set of planar vectors is constructed using these sampling points with the third and fourth local neighborhood sampling points, respectively. Calculate the vector product of the two sets of plane vectors to obtain two spatial sub-normal vectors, determine the consistency of the directions of the two spatial sub-normal vectors, and algebraically add the two spatial sub-normal vectors while ensuring that the directions are consistent; The initial normal vector of the sampling point is obtained by normalizing the spatial normal vector after algebraic addition by dividing it by its own L2 norm magnitude.

[0009] Preferably, the step of determining the direction consistency of the two spatial sub-normal vectors and performing algebraic addition on the premise of ensuring direction consistency includes: Calculate the dot product of the two spatial subnormal vectors; If the result of the dot product is less than zero, then one of the spatial sub-normal vectors is reversed, and then the two spatial sub-normal vectors are algebraically added together.

[0010] Preferably, the degradation benchmark value includes: For any of the local element stiffness matrices, calculate the diagonal deviation of each row in the local element stiffness matrix; The process of calculating the diagonal deviation of each row is as follows: calculate the sum of the absolute values ​​of all non-diagonal elements in the row, and divide the sum of absolute values ​​by the absolute value of the main diagonal element in the row to obtain the diagonal deviation of the row. Obtain the maximum value of the diagonal deviation of all rows of the local element stiffness matrix, and determine the maximum value as the corresponding degradation reference value of the mesh element.

[0011] Preferably, when the confidence level of the structural strength assessment of the energy storage device cabin is less than a preset confidence level threshold, the mean matrix deterioration of all grid cells is obtained. Based on the mean matrix deterioration, inferior cells in all grid cells are identified. The grid step size of the identified inferior cells and their adjacent topologically adjacent cells is updated to obtain the updated grid cells. Based on the global algebraic equations constructed from the stiffness matrices of each local cell corresponding to the updated grid cells, the updated global relative force residual value is obtained by solving the equations. Based on the updated global relative force residual value and the matrix deterioration of the updated grid cells, the result is considered to be less than or equal to the mean matrix deterioration. The confidence level of the updated energy storage device cabin structural strength assessment is obtained by using the matrix deterioration degree corresponding to the grid cell with the mean matrix deterioration degree. If the confidence level of the updated energy storage device cabin structural strength assessment is greater than or equal to a preset confidence level threshold, the structural strength of the energy storage device cabin is evaluated. If the confidence level of the updated energy storage device cabin structural strength assessment is less than the preset confidence level threshold, the process continues to obtain and update all inferior cells in the updated grid cells, iterating until the confidence level of the obtained energy storage device cabin structural strength assessment is greater than or equal to the preset confidence level threshold, at which point the iteration stops.

[0012] Preferably, when the confidence level of the structural strength assessment of the energy storage device cabin is less than a preset confidence level threshold, the mesh step size of all mesh elements is updated to obtain all updated mesh elements; based on the global algebraic equation system constructed from the stiffness matrices of each local element corresponding to all updated mesh elements, the updated global relative force residual value is obtained by solving; based on the matrix deterioration degree of all updated mesh elements and the updated global relative force residual value, the confidence level of the updated structural strength assessment of the energy storage device cabin is obtained; if the confidence level of the updated structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, the structural strength of the energy storage device cabin is evaluated; if the confidence level of the updated structural strength assessment of the energy storage device cabin is less than the preset confidence level threshold, all updated mesh elements are obtained and updated, and this process is iterated until the confidence level of the obtained structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, at which point the iteration stops.

[0013] Preferably, the mesh step size of all inferior cells and their topologically adjacent cells is halved during the update.

[0014] Preferably, for all mesh elements in the current full-cabin finite element mesh, their original mesh step size is uniformly reduced by half and updated.

[0015] Preferably, the step of obtaining the maximum local stress value of all mesh elements in the finite element mesh, and evaluating the structural strength of the energy storage device cabin based on the maximum local stress value of all mesh elements, includes: When the confidence level is greater than or equal to the preset confidence level threshold, the maximum local stress value obtained by solving the finite element mesh is extracted; The maximum local stress value is compared with the inherent allowable stress corresponding to the material of the energy storage device compartment. If the maximum local stress value is less than or equal to the inherent allowable stress, then the assessment conclusion that the energy storage device cabin structure is qualified is output. If the maximum local stress value is greater than the inherent allowable stress, then the assessment conclusion that the local structural strength of the energy storage device compartment is insufficient is output.

[0016] Secondly, a structural strength assessment system for an energy storage device compartment includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the aforementioned structural strength assessment method for an energy storage device compartment is implemented.

[0017] The present invention has the following effects: 1. Existing finite element method (FEM) preprocessing techniques typically evaluate mesh quality based solely on a single geometric shape, failing to accurately reflect the impact of mesh distortion on the governing equations. This invention calculates matrix degradation and combines the structural curvature, characterizing the severity of local physical topographic changes, with a degradation benchmark value characterizing the degree of degradation in the governing equations. This reflects the extent to which the local stiffness matrix loses its diagonal dominance due to mesh distortion, improving the accuracy of error identification. 2. To address the issue of a surge in computational load that can occur when global mesh refinement is applied to energy storage compartments, this invention calculates confidence levels and establishes an adaptive mesh adjustment mechanism based on these confidence levels. The system only updates the mesh step size by halving the mesh size for mesh cells with matrix degradation values ​​higher than the mean and their topologically adjacent cells, while retaining the original mesh size for other flat regions. Attached Figure Description

[0018] Figure 1 This is a flowchart of steps S1-S6 in a structural strength assessment method for an energy storage device cabin according to an embodiment of the present invention. Figure 2 This is a structural block diagram of a structural strength assessment system for an energy storage device compartment according to an embodiment of the present invention; Figure 3 This is a schematic diagram of sampling points and their local neighborhood sampling points in a structural strength assessment method for an energy storage device cabin according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0020] Reference Figure 1 A method for assessing the structural strength of an energy storage device compartment includes steps S1-S6, as detailed below: S1: Obtain the finite element mesh of the energy storage device compartment, and extract the corner points of each mesh element in the finite element mesh as sampling points to form a discrete point set; It should be noted that the energy storage device compartment, as the core mechanical component carrying the battery module and electrical system, has a complex local topological morphology with densely distributed heat dissipation openings, fixing lugs, and thin-plate skeletons on its surface. Since these critical abrupt change areas are prone to severe stress concentration during vehicle vibration, hoisting, or external collisions, this invention first requires digital discrete sampling of the entire surface of the compartment in three-dimensional space. This converts the continuous physical geometry into deterministic three-dimensional spatial coordinate data, providing a data foundation for subsequent curvature calculations and algebraic evaluations.

[0021] In one embodiment, the specific process for obtaining the finite element mesh of the energy storage device compartment is as follows: First, the system imports the 3D geometric model of the energy storage device compartment. Using a spatial bounding box isometric mesh generation algorithm, a set of regular background meshes with a preset spatial step size is generated within the 3D closed domain containing the energy storage device compartment. Then, using the conventional boundary node projection and snapping mechanism in finite element preprocessing, the regular mesh nodes near the boundaries of the 3D geometric model of the energy storage device compartment are forcibly projected and snapped onto the actual physical surface of the energy storage device compartment, thereby generating the finite element mesh used for this evaluation. The preset spatial step size can be limited by the implementers according to the specific implementation situation; for example, a preset spatial step size of 50mm.

[0022] It should be noted that the conventional boundary node projection and snap-in mechanism in finite element preprocessing is as follows: the algorithm automatically identifies regular mesh nodes that are close to the real boundary of the 3D geometric model of the energy storage device compartment, calculates the shortest projection distance from these nodes to the real physical surface, and then forcibly moves and snaps the spatial coordinates of these nodes onto the real physical surface of the energy storage device compartment (including local structures such as the edges of heat dissipation openings, fixing lugs, and thin plate skeletons). After the above projection and snap-in processing, the 3D spatial coordinates of the boundary corners of the finite element mesh perfectly fit the real geometric abrupt surface, possessing the structural characteristics that characterize the curvature of the real physical morphology.

[0023] Then, the system extracts the corner points of each mesh element in the finite element mesh and uses these corner points directly as sampling points. All extracted sampling points together constitute a global three-dimensional spatial discrete point set.

[0024] S2: Obtain the initial normal vectors of each sampling point in the discrete point set and its local neighborhood sampling points, and calculate the structural curvature of each sampling point; It should be noted that in the physical structure of the energy storage device compartment, key components such as heat dissipation vents and mounting brackets are abruptly discontinuous geometric regions. Due to the enormous weight of the densely packed battery modules inside the compartment, these abrupt topological boundaries experience severe stress concentration under conditions such as vibrations during vehicle transport, hoisting, or external collisions. Traditional finite element discretization algorithms typically employ mechanical equidistant partitioning when handling this scenario, resulting in mesh element nodes that cannot perfectly align with the physical boundaries of the heat dissipation vent arcs or the sharp corners of the mounting brackets. Forcibly stretching the nodes to align with the boundaries would inevitably cause severe mesh distortion during the preprocessing mesh generation stage. Therefore, by calculating the structural curvature at each sampling point, the severity of abrupt geometric boundary changes in each region is quantified. This provides a reference benchmark for subsequent adaptive adjustment of the finite element mesh, avoiding mesh distortion caused by blind partitioning.

[0025] Specifically, the method for obtaining the initial normal vectors of each sampling point and its local neighborhood sampling points in a discrete point set includes the following steps: For any sampling point, a nearest neighbor topological search is performed based on the topological adjacency list of nodes and mesh elements in the finite element mesh. The search proceeds along the topological edges of the mesh element boundaries to obtain at least four sampling points that form a first-order topological adjacency relationship with the current sampling point, thus constituting a local neighborhood point set. Each sampling point in the local neighborhood point set is a local neighborhood sampling point. For example... Figure 3 The diagram shown is a schematic of a sampling point and its local neighborhood sampling points. For sampling point a, its local neighborhood sampling points are sampling point b, sampling point c, sampling point d, sampling point e, sampling point f, sampling point g, sampling point h, and sampling point i. Sampling point b, sampling point c, sampling point d, sampling point e, sampling point f, sampling point g, sampling point h, and sampling point i constitute the local neighborhood point set of sampling point a.

[0026] After obtaining the local neighborhood point set, the local neighborhood point set is searched in ascending order of Euclidean distance from the sampling point to select the first and second local neighborhood sampling points that are not collinear with the sampling point, as well as the third and fourth local neighborhood sampling points that are not collinear with the sampling point. Then, the sampling point is used to construct a first set of planar vectors with the first and second local neighborhood sampling points, and the sampling point is used to construct a second set of planar vectors with the third and fourth local neighborhood sampling points.

[0027] It is important to note that when discretizing the geometric data of the energy storage compartment, there are drastic geometric abrupt changes and non-uniform anisotropy in the local topography, such as heat dissipation openings and fixing lugs. If conventional methods are used to select only the two nearest points to construct a dot product to calculate the normal vector, if either of these two points happens to fall on a geometric singularity of the explosion-proof opening or weld edge, the calculated normal vector will be excessively skewed to one side due to local rigidity abrupt changes, failing to accurately represent the true orientation of the continuous geometric surface at that sampling point. Therefore, this invention selects four non-collinear local neighborhood sampling points, subtracts their coordinates to decompose them into two independent sets of planar vectors, aiming to offset the single-set normal component deviation error caused by uneven discrete sampling or abrupt sharp angles through subsequent addition and averaging.

[0028] Then, using the three-dimensional space cross product formula, the vector product of the two sets of planar vectors is calculated to obtain two spatial sub-normal vectors. To avoid the problem that the L2 norm of the algebraic sum of the two spatial sub-normal vectors would be 0 when their directions are opposite, thus causing subsequent normalization processing to fail, the system checks the consistency of the directions of the two spatial sub-normal vectors before algebraic addition. The specific judgment and processing process is as follows: calculate the dot product of the two spatial sub-normal vectors; if the result of the dot product is less than zero (i.e., the direction angle is greater than 90 degrees, and the directions are inconsistent), then one of the spatial sub-normal vectors is reversed (multiplied by -1), and then the two spatial sub-normal vectors are algebraically added; if the result of the dot product is greater than or equal to zero, then the two spatial sub-normal vectors are directly algebraically added.

[0029] Finally, the spatial normal vector, after algebraic summation, is normalized by dividing it by its own L2 norm modulus to obtain the initial normal vector of the sampling point.

[0030] After obtaining the initial normal vectors at each point, the methods for obtaining the curvature of the structure include: For any sampling point, calculate the dot product between the initial normal vector of the sampling point and the initial normal vectors of each local neighbor sampling point in the corresponding local neighbor point set, and obtain the dot product result; and calculate the structural curvature of the sampling point based on the dot product result.

[0031] In a specific embodiment, the specific mathematical expression for calculating the structural curvature based on the dot product result is as follows: ; In the formula, This indicates the curvature of the structure at that sampling point; This represents the total number of local neighborhood sampling points contained in the local neighborhood point set; This represents the initial normal vector of the sampling point; Represents the first local neighborhood point set. The initial normal vector of each local neighborhood sampling point; This is the result of the dot product of the two.

[0032] In finite element geometric analysis, the dot product of two unit normal vectors reflects the cosine variation trend of the angle between them in local space. The term in the formula ( It can accurately and quantitatively map the degree of normal torsion of the surface of the energy storage device in a local microscopic neighborhood.

[0033] The structural curvature obtained from the above calculations This characterizes the degree of curvature, abrupt changes, or discontinuities on the surface of the energy storage device at the sampling point. Specifically, the flatter the local neighborhood surface around the sampling point, the more consistent the directions between adjacent initial normal vectors, the larger their dot product, and the higher the calculated structural curvature. The smaller the value, the safer and less prone to distortion when performing finite element mesh generation at this location; conversely, when the sampling point is located at the edge of the heat dissipation hole arc, the angle of the fixed lug, or a sudden dead angle of the frame welding, or when the local mesh is twisted and deformed, the normal vector will produce a significant directional deflection locally, and its dot product result will be smaller, thus affecting the calculated structural curvature. The corresponding increase in the value indicates that the risk of feature loss and grid distortion during grid discretization is extremely high.

[0034] S3: Obtain the local element stiffness matrix corresponding to each grid element, calculate the diagonal deviation of the local element stiffness matrix of each grid element to determine the degradation reference value, and combine the degradation reference value with the structural curvature of each sampling point corresponding to each grid element to calculate the matrix degradation degree of each grid element. For each mesh element in the finite element mesh, the system obtains the local element stiffness matrix based on the three-dimensional coordinates of its existing sampling points, combined with preset material constitutive properties and local geometric section properties, using conventional numerical integration techniques in finite element analysis (such as Gaussian numerical integration). In finite element analysis, the local element stiffness matrix is ​​a reference algebraic matrix characterizing the ability of a single finite element mesh element to resist deformation under stress; it represents a linear matrix operator that maps the element nodal displacement vector to the element nodal force vector.

[0035] In this embodiment, a thin-shell mesh element is used as an example for discretization. Each mesh element contains 4 corner points (i.e., 4 sampling points), and each sampling point has 6 degrees of freedom. The corresponding local element stiffness matrix is ​​a 24th-order symmetric square matrix. As mentioned above, when the finite element mesh deals with non-uniform abrupt boundaries such as heat dissipation openings, the nodes undergo forced projection and adsorption, which can easily lead to distortion of the mesh element's geometry (such as excessive aspect ratio or interior angle distortion). This geometric distortion directly causes the corresponding local element stiffness matrix to lose its diagonal dominance, resulting in algebraic solution errors. To avoid the huge computational load brought about by traditional matrix eigenvalue decomposition, this embodiment innovatively uses the Gerschgorin Circle theorem to calculate the diagonal deviation of the local element stiffness matrix, thereby achieving low-cost quantification of the algebraic degradation degree of any mesh element.

[0036] Specifically, the degradation benchmark value is obtained by calculating the diagonal deviation of each row in any local element stiffness matrix. The calculation process for the diagonal deviation of each row is as follows: calculate the sum of the absolute values ​​of all off-diagonal elements in the row, and divide the sum of absolute values ​​by the absolute value of the main diagonal element in the row to obtain the diagonal deviation of the row.

[0037] In one embodiment, the stiffness matrix of any local element is... The calculation method for row diagonal deviation is as follows: ; In the formula, The local element stiffness matrix represents the first... Diagonal deviation of the row; The local element stiffness matrix represents the first... Line 1 The off-diagonal elements of the column are scalars; The local element stiffness matrix represents the first... Scalars of elements on the main diagonal of a row.

[0038] in, This represents the grid cell at the 1st The self-stiffness of the bodily structure in each degree of freedom; and the sum of the absolute values ​​of the non-diagonal elements. This indicates the influence of external disturbance forces on the internal body's self-stiffness. The larger the value, the more severe the loss of diagonal dominance of the local element stiffness matrix in that specific degree of freedom dimension, and the more ill-conditioned the algebraic state becomes.

[0039] Then, the system obtains the maximum value of the diagonal deviation of all rows of the local element stiffness matrix and determines it as the degradation reference value of the mesh element.

[0040] It is important to note that in the structural strength assessment of energy storage compartments, due to the strong directionality of operating loads and local geometry, the degradation degree of distorted mesh elements often exhibits a highly uneven distribution across their 24 generalized degrees of freedom. If this invention employs conventional averaging methods, the values ​​of degrees of freedom with less degradation would mask and dilute the calculation errors of the most severely disturbed rows, leading to an overall underestimation of ill-conditioned behavior. Therefore, this invention extracts the maximum value of the diagonal deviation as the degradation baseline value for that mesh element.

[0041] After determining the degradation baseline value, the methods for obtaining the matrix degradation degree include: For any given grid cell, obtain the structural curvature of all sampling points corresponding to that grid cell; Secondly, calculate the mean value of the structural curvature of all sampling points corresponding to the grid cell; Finally, the mean is multiplied by the degradation baseline value corresponding to the grid cell to obtain the matrix degradation degree of the grid cell.

[0042] It should be noted that the mean value of the structural curvature reflects the degree of geometric abrupt change on the surface of the energy storage device cabin in a local area, while the degradation benchmark value reflects the degree of deterioration caused by the geometric distortion of the mesh. This invention multiplies the two to obtain the matrix degradation degree, which can accurately identify inferior mesh cells that are both on the edge of topological abrupt change and have lost the diagonal advantage of the matrix. This avoids the computing power explosion caused by blindly refining the mesh globally and improves the computing power allocation efficiency of finite element evaluation.

[0043] S4: Based on the global algebraic equations constructed from the stiffness matrices of each local element, the global relative force residual value is obtained by solving the equations. Combined with the matrix deterioration degree of each grid element, the confidence level of the structural strength assessment of the energy storage device cabin is calculated. It should be noted that after solving all the stiffness matrix equations of the energy storage compartment and outputting the preliminary stress distribution results, the general finite element method inevitably generates some geometrically distorted mesh elements when mechanically dividing complex abrupt boundaries such as heat dissipation openings and fixed lugs. This leads to an increase in the matrix deterioration value corresponding to these mesh elements. This makes it difficult for technicians to accurately determine whether the stress value at the edge of the heat dissipation opening or at the weld of the frame is due to actual physical structural failure of the energy storage compartment or due to the accumulation of front-end mesh distortion and algebraic errors when faced with local high stress peaks. Therefore, this embodiment combines the matrix deterioration distribution of all mesh elements in the energy storage compartment with the global relative force residual value output by the final solution to comprehensively calculate the confidence level of the structural strength assessment.

[0044] In one embodiment, the confidence level calculation process includes: First, the system iteratively solves a global algebraic equation system based on the stiffness matrices of each local element using existing finite element method (FE) techniques, and extracts the global relative force residuals output during the solution process. The processes of constructing and solving the global algebraic equation system are conventional techniques in this field, and the global relative force residuals are dimensionless scalar data inherently generated by existing techniques when determining the convergence of finite element nonlinearity or iterative calculations.

[0045] Then, the confidence level of the structural strength assessment of the energy storage device cabin is calculated based on the matrix deterioration degree corresponding to the grid cell. The specific calculation formula is as follows: ; In the formula, This indicates the confidence level of the calculated structural strength assessment of the energy storage device cabin; This represents the scalar value of the global relative force residual obtained. Indicates the first The matrix degradation degree corresponding to each grid cell; This represents the total number of mesh elements contained in the finite element mesh of the entire energy storage device compartment.

[0046] Confidence This represents the accuracy and distortion resistance of the stress and deformation response results of the energy storage chamber, and its value range is located in... Between. When The closer the value is to 1, the higher the convergence accuracy of the global algebraic equation system (i.e., the higher the convergence accuracy). When the algebraic stiffness of each mesh element is extremely small (i.e., the mean matrix deterioration is extremely small), the output local maximum stress data closely approximates the actual physical state and has extremely high reliability; conversely, if... If the value drops sharply and approaches 0, it indicates that the ill-conditioned cumulative error of the matrix caused by the front-end mesh distortion has seriously affected the true solution. At this time, the calculated local high stress data is very likely to be a singular error and should not be blindly accepted.

[0047] It should be noted that the global relative force residual value As a dimensionless convergence adjoint, it is used to characterize the approximation error at the algebraic level, while the mean term in the latter part is used to introduce a penalty constraint for mesh degradation. When computational distortion is caused by severe mesh distortion, although the algebraic equations in the residuals... The above shows weak convergence ( (Smaller), but local units can worsen the matrix performance. The sharp increase in the number of terms leads to the penalty term in the denominator playing a dominant amplifying role, forcing the confidence level to rise. This significantly reduces the error, thus enabling precise capture of latent distortion errors.

[0048] S5: Compare the confidence level with the preset confidence level threshold. If the confidence level is less than the confidence level threshold, update the finite element mesh and iteratively return to the structural strength assessment method based on the updated finite element mesh until the confidence level of the obtained energy storage device cabin structural strength assessment is greater than or equal to the confidence level threshold, so as to trigger the execution of step S6. In actual engineering simulations of energy storage device cabins, high stress peaks are typically concentrated in geometrically discontinuous areas such as the edges of heat dissipation openings, the angles of fixed lugs, or the welded joints of thin-plate frames. However, these areas are also often where the mesh is most prone to geometric distortions such as twisting and stretching. When local mesh elements are distorted, their corresponding local element stiffness matrices will severely lose their diagonal dominance, leading to undesirable responses in the algebraic equations and causing singularity errors to accumulate in these areas during numerical solutions. Therefore, this invention uses confidence levels to determine whether the current finite element mesh is contaminated. If the confidence level falls below a preset acceptable threshold, it indicates that the current finite element mesh is contaminated and the current output result is unreliable.

[0049] In one embodiment, when the confidence level of the structural strength assessment of the energy storage device cabin is less than a preset confidence level threshold, the system first calculates the average matrix deterioration degree of each grid cell; secondly, grid cells with a matrix deterioration degree higher than the average value are screened as substandard cells; then, the grid step size of all substandard cells and their topologically adjacent cells is halved and updated. The preset confidence level threshold is 0.85.

[0050] It should be noted that, in order to ensure the mathematical continuity of the finite element analysis, when re-dividing based on the updated mesh step size, the system will automatically insert a gradient quadrilateral transition mesh element between the locally non-uniform dense mesh with half the step size and the regular large-size sparse mesh that has not been re-divided; the shape function of the transition mesh element is used to perform displacement coordination interpolation, thereby completely eliminating mesh hanging nodes caused by local refinement and ensuring the continuity of the stiffness matrix of the global algebraic equation system during assembly.

[0051] In another embodiment, when the confidence level of the structural strength assessment of the energy storage device compartment is less than the preset confidence level threshold, the system performs a global unified mesh update, that is, it directly performs a full halving update on all mesh elements in the current full compartment finite element mesh (for example, reducing the original 50mm mesh space step size to 25mm), and re-divides the entire energy storage device compartment based on the fully updated mesh step size.

[0052] After completing the mesh re-division in any of the above embodiments, since the mesh geometry and element node coordinates have changed, the system controls the program to loop back to step S1 to re-execute. The system re-extracts the corner points of each mesh element based on the updated high-density mesh to construct a new global discrete point set, and sequentially triggers steps S2, S3, and S4. It then re-executes the nearest neighbor topology search on the new discrete point set to recalculate the initial normal vector, structural curvature, local stiffness matrix diagonal deviation, and matrix degradation, and then re-solves the assembled global algebraic equations to update the confidence level. Only when the re-derived confidence level is greater than or equal to the confidence level threshold can the numerical solution under the currently updated mesh be confirmed as completely valid, the loop control chain be exited, and step S6 be triggered. Furthermore, to prevent infinite mesh refinement due to geometric model singularities leading to an infinite loop, this embodiment also presets a maximum iteration threshold (e.g., 5 times). When the number of iterative re-divisions reaches the maximum iteration threshold and the confidence level still does not reach the confidence level threshold, the system forcibly terminates the iteration, outputs the current stress distribution, and provides a risk warning.

[0053] S6: If the confidence level is greater than or equal to the confidence level threshold, the structural strength assessment result of the energy storage device cabin will be output based on the stress response data obtained by solving the current finite element mesh.

[0054] In one embodiment, the process of outputting the structural strength assessment results of the energy storage device compartment includes: When the confidence level is greater than or equal to the preset confidence level threshold, the system recognizes the current stress distribution calculation results and automatically extracts the maximum local stress value obtained from solving the finite element mesh.

[0055] The maximum local stress value obtained from the finite element mesh is numerically compared with the inherent allowable stress corresponding to the material of the energy storage compartment. The inherent allowable stress corresponding to the material of the energy storage compartment is a fixed material mechanical constant obtained based on the metal grade actually used in the energy storage compartment.

[0056] After comparison, the corresponding evaluation conclusion is output: If the maximum local stress value is not greater than the inherent allowable stress, it means that even at the point of geometric change where the stress is most concentrated, the material is still safe, and the system automatically outputs the assessment conclusion that "the energy storage device cabin structure is qualified". If the maximum local stress value is greater than the inherent allowable stress, it indicates that there is a risk of physical damage, and the system will automatically output the assessment conclusion that "the local structural strength of the energy storage device compartment is insufficient".

[0057] This invention also provides a structural strength assessment system for an energy storage device compartment. For example... Figure 2As shown, the system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement a structural strength assessment method for an energy storage device compartment according to the first aspect of the present invention. The system also includes other components well-known to those skilled in the art, such as a communication bus and a communication interface, the setup and functions of which are known in the art and will not be described further here.

[0058] It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this invention, and these modifications and improvements are all within the scope of protection of this invention.

Claims

1. A method for assessing the structural strength of an energy storage device compartment, characterized in that, include: The finite element mesh of the energy storage device compartment is obtained, and all the mesh elements in the finite element mesh are obtained based on the finite element mesh. The corner points of each mesh element are used as sampling points to form a discrete point set. The initial normal vectors of each sampling point in the discrete point set and its adjacent local neighborhood sampling points are obtained to obtain the structural curvature of the sampling points; The local element stiffness matrix of each grid cell is obtained based on the three-dimensional coordinates of each sampling point in each grid cell, and the diagonal deviation of each grid cell is obtained based on the local element stiffness matrix of each grid cell, and the degradation reference value of each grid cell is determined. The matrix degradation degree of each grid cell is obtained by using the degradation reference value of each grid cell and the structural curvature of all sampling points in each grid cell; Based on the global relative force residual value obtained from the global algebraic equation system constructed from the stiffness matrices of each local element, and the matrix deterioration degree of each grid element, the confidence level of the structural strength assessment of the energy storage device cabin is calculated. When the confidence level of the structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, the maximum local stress value of all grid cells in the finite element mesh is obtained, and the structural strength of the energy storage device cabin is evaluated based on the maximum local stress value of all grid cells.

2. The structural strength assessment method for an energy storage device compartment according to claim 1, characterized in that, The initial normal vectors of each sampling point in the discrete point set and its adjacent local neighborhood sampling points include: For any sampling point, a nearest neighbor topology search is performed based on the topological connection relationship of each grid cell in the finite element mesh to obtain at least four sampling points that are topologically adjacent to the sampling point through the grid edge, forming a local neighborhood point set. Each sampling point in the local neighborhood point set is a local neighborhood sampling point. In the set of local neighborhood points, first, second, third, and fourth local neighborhood sampling points that are not collinear are selected, and a first set of planar vectors is constructed using these sampling points with the first and second local neighborhood sampling points, and a second set of planar vectors is constructed using these sampling points with the third and fourth local neighborhood sampling points, respectively. Calculate the vector product of the two sets of plane vectors to obtain two spatial sub-normal vectors, determine the consistency of the directions of the two spatial sub-normal vectors, and algebraically add the two spatial sub-normal vectors while ensuring that the directions are consistent; The initial normal vector of the sampling point is obtained by normalizing the spatial normal vector after algebraic addition by dividing it by its own L2 norm magnitude.

3. The structural strength assessment method for an energy storage device compartment according to claim 2, characterized in that, The step of determining the consistency of the directions of the two spatial sub-normal vectors, and performing algebraic addition of the two spatial sub-normal vectors while ensuring that the directions are consistent, includes: Calculate the dot product of the two spatial subnormal vectors; If the result of the dot product is less than zero, then one of the spatial sub-normal vectors is reversed, and then the two spatial sub-normal vectors are algebraically added together.

4. The structural strength assessment method for an energy storage device compartment according to claim 1, characterized in that, The degradation baseline value includes: For any of the local element stiffness matrices, calculate the diagonal deviation of each row in the local element stiffness matrix; The process of calculating the diagonal deviation of each row is as follows: calculate the sum of the absolute values ​​of all non-diagonal elements in the row, and divide the sum of absolute values ​​by the absolute value of the main diagonal element in the row to obtain the diagonal deviation of the row. Obtain the maximum value of the diagonal deviation of all rows of the local element stiffness matrix, and determine the maximum value as the corresponding degradation reference value of the mesh element.

5. The structural strength assessment method for an energy storage device compartment according to claim 1, characterized in that, When the confidence level of the structural strength assessment of the energy storage device cabin is less than the preset confidence level threshold, the mean matrix deterioration of all grid cells is obtained. Based on the mean matrix deterioration, the inferior cells in all grid cells are identified. The grid step size of the identified inferior cells and their adjacent topologically adjacent cells is updated to obtain the updated grid cells. Based on the global algebraic equation system constructed from the stiffness matrices of each local cell corresponding to the updated grid cells, the updated global relative force residual value is obtained by solving the equation system. Based on the updated global relative force residual value, the matrix degradation degree of the updated grid cells, and the matrix degradation degree corresponding to grid cells with a matrix degradation degree less than or equal to the mean matrix degradation degree, the confidence level of the updated energy storage device cabin structural strength assessment is obtained. If the confidence level of the updated energy storage device cabin structural strength assessment is greater than or equal to a preset confidence level threshold, the structural strength of the energy storage device cabin is evaluated. If the confidence level of the updated energy storage device cabin structural strength assessment is less than the preset confidence level threshold, then the inferior cells in all updated grid cells are obtained and updated, and this process is iterated until the confidence level of the obtained energy storage device cabin structural strength assessment is greater than or equal to the preset confidence level threshold, at which point the iteration stops.

6. The structural strength assessment method for an energy storage device compartment according to claim 1, characterized in that, When the confidence level of the structural strength assessment of the energy storage device cabin is less than a preset confidence level threshold, the mesh step size of all mesh elements is updated to obtain all updated mesh elements. Based on the global algebraic equation system constructed from the stiffness matrices of each local element corresponding to all updated mesh elements, the updated global relative force residual value is obtained by solving the system. Based on the matrix deterioration degree of all updated mesh elements and the updated global relative force residual value, the confidence level of the updated structural strength assessment of the energy storage device cabin is obtained. If the confidence level of the updated structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, the structural strength of the energy storage device cabin is evaluated. If the confidence level of the updated structural strength assessment of the energy storage device cabin is less than the preset confidence level threshold, all updated mesh elements are obtained and updated, and this process is iterated until the confidence level of the obtained structural strength assessment of the energy storage device cabin is greater than or equal to the preset confidence level threshold, at which point the iteration stops.

7. The structural strength assessment method for an energy storage device compartment according to claim 5, characterized in that, The mesh step size of all inferior cells and their topologically adjacent cells is halved during the update.

8. The structural strength assessment method for an energy storage device compartment according to claim 6, characterized in that, For all mesh elements in the current full-cabin finite element mesh, their original mesh step size is uniformly reduced by half and updated.

9. The structural strength assessment method for an energy storage device compartment according to claim 1, characterized in that, The process of obtaining the maximum local stress value of all mesh elements in the finite element mesh and evaluating the structural strength of the energy storage device cabin based on the maximum local stress value of all mesh elements includes: When the confidence level is greater than or equal to the preset confidence level threshold, the maximum local stress value obtained by solving the finite element mesh is extracted; The maximum local stress value is compared with the inherent allowable stress corresponding to the material of the energy storage device compartment. If the maximum local stress value is less than or equal to the inherent allowable stress, then the assessment conclusion that the energy storage device cabin structure is qualified is output. If the maximum local stress value is greater than the inherent allowable stress, then the assessment conclusion that the local structural strength of the energy storage device compartment is insufficient is output.

10. A structural strength assessment system for an energy storage device compartment, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement a structural strength assessment method for an energy storage device compartment according to any one of claims 1-9.

Images