Automatic generation method for finite element model of wing hexahedron element structure

Abstract

The invention discloses a method for automatically generating a finite element model of a wing hexahedron element structure, and belongs to the field of multi-physics field coupling numerical simulation. Comprising the steps that a radial basis interpolation model driven by a curve distance is adopted, topologically consistent wing root and wingtip quadrilateral grids are generated, geometric and topological mapping from two dimensions to three dimensions is achieved through control point alignment and spanwise over-limit interpolation, a wing is divided into 10 parameterized blocks, internal nodes are generated through tensor product interpolation, and the internal nodes are divided into 10 parameterized blocks. And combining repetitive nodes based on a spatial hash algorithm, and constructing a globally consistent hexahedral unit topology. According to the method, full hexahedron unit generation can be ensured, and the generation efficiency is remarkably improved compared with that of a manual interactive method.

Description

Technical Field

[0001] This invention relates to the field of multiphysics coupled numerical simulation, specifically to a method for automatically generating a finite element model of a wing hexahedral element structure. Background Technology

[0002] In wind tunnel tests of large-scale aircraft, the three-dimensional elastic deformation of the model under aerodynamic loads significantly interferes with the accuracy of measurement data. Therefore, it is crucial to accurately predict and correct the static deformation of the structure through aeroelastic coupling simulation. Such simulations require high-precision aerodynamic meshes and structural finite element meshes as inputs, with the structural mesh needing to fully characterize the three-dimensional deformation characteristics of the wing. Tetrahedral elements, as constant strain elements, require excessively large meshes for finite element analysis of wing structures due to their low shape function order. Hexahedral elements, on the other hand, offer higher interpolation accuracy. For wing structures dominated by bending and torsional deformation, the number of elements can be significantly reduced while maintaining the same accuracy requirements. Furthermore, proper control of the element aspect ratio can effectively improve the computational accuracy in areas with large deformation gradients. The wing mesh needs to be generated using parametric sweeping techniques to create a two-dimensional airfoil extension. The core of this technique lies in the topological consistency of the quadrilateral mesh in the airfoil profile and the robustness of the three-dimensional mapping relationship.

[0003] Currently, the generation of hexahedral meshes for three-dimensional airfoils mainly relies on two types of methods: (1) Manual interactive block division: The airfoil is manually divided into multiple hexahedral blocks, and meshes are generated separately and then spliced ​​together. Although this method can guarantee all hexahedral elements, it requires a lot of manual intervention and has poor adaptability to complex geometries (such as large swept wings and winglets). (2) Sweep method: The two-dimensional airfoil mesh is swept along the spanwise direction to generate a three-dimensional mesh. However, when the airfoil profile topology is inconsistent or there is torsion, the sweep method is prone to element distortion or even degeneration into prisms, and cannot guarantee all hexahedral elements. In addition, many existing methods often fail to achieve strict alignment of node connections between different airfoil profiles, resulting in discontinuities or overlaps in the three-dimensional mesh. Summary of the Invention

[0004] The purpose of this invention is to establish a unified topological framework for the wing root and wing tip sections, extend the two-dimensional airfoil quadrilateral mesh to three-dimensional space according to parametric rules, ensure the consistency of element connections in the spanwise, chordwise and normal directions, ensure the generation of all hexahedral elements, and significantly improve the generation efficiency compared to manual interactive methods.

[0005] This invention is achieved through the following technical solution:

[0006] A method for automatically generating a finite element model of an airfoil hexahedral element structure, which extends the two-dimensional airfoil quadrilateral mesh to three-dimensional space through parametric sweeping technology, including:

[0007] A radial basis interpolation model driven by curve distance is used to generate topologically consistent quadrilateral meshes at the wing root and wingtip.

[0008] By using control point alignment and spanwise over-limit interpolation, geometric and topological mapping from two dimensions to three dimensions is achieved;

[0009] The wing is divided into several parameterized blocks, and internal nodes are generated using tensor product interpolation.

[0010] Duplicate nodes are merged using a spatial hash algorithm to construct a globally consistent hexahedral unit topology.

[0011] In the above technical solution, the realization of two-dimensional to three-dimensional geometric and topological mapping includes:

[0012] Two-dimensional quadrilateral meshes are generated at the wing root and wingtip sections of the wing to establish a two-dimensional mesh topology.

[0013] The three-dimensional coordinates of each grid point are obtained by geometric transformation of the wing root, wing tip and airfoil profiles, and continuous spanwise grid lines are constructed by connecting control points and grid points;

[0014] A three-dimensional curved quadrilateral mesh surface is constructed by two spanning edges and two cross-sectional edges.

[0015] In the above technical solution, for wings with sweep angles or dihedral angles, the coordinates of the control points are mapped according to the aerodynamic shape;

[0016] The spanwise grid lines are connected by straight lines or parametric curves. Control points corresponding to the wing root, wing tip, and airfoil profile are selected as endpoints, and several intermediate control points are inserted according to the geometric change trend.

[0017] In the above technical solution, the parameterized block includes:

[0018] The airfoil is divided into multiple hexahedral blocks to construct a three-dimensional mesh block by using the characteristic mesh surfaces of the airfoil section, chordal surface, and span.

[0019] Perform trilinear interpolation on each 3D mesh block, define parameter coordinates, and generate a 3D mesh volume.

[0020] In the above technical solution, the hexahedral block is composed of six characteristic mesh surfaces, including the wing root section, the wingtip section, and the spanwise surface connecting the two.

[0021] In the above technical solution, the hexahedral block is divided into an inner block and an outer block. The outer block covers the outer aerodynamic surface of the wing, and the inner block fills the wing structure, which is composed of a skeleton surface and an internal cross-sectional area surface.

[0022] In the above technical solution, the node distribution corresponding to the trilinear interpolation is dynamically adjusted according to the wing geometry, including:

[0023] Curvature adaptation: In high curvature regions, the node distribution parameter is set to a smaller value to refine the mesh, while in flat regions such as the middle of the wing surface, it is set to a larger value to reduce the mesh density.

[0024] The grid point distribution is automatically adjusted by using an interpolation function to adjust the node distance based on the geometric characteristics of the airfoil's curvature radius and sweep angle.

[0025] In the above technical solution, the setting of the smaller value includes:

[0026] Select a typical airfoil profile and load condition;

[0027] In the chord direction, a series of meshes with gradually increasing node numbers are generated and structural static analysis is performed.

[0028] Record the key metrics for each grid density.

[0029] When the rate of change of the target result caused by two consecutive encryptions is lower than the preset convergence tolerance, the mesh density is judged to meet the requirements.

[0030] In the above technical solution, the construction of the interpolation function includes:

[0031] Discretize a series of sample points along the chord direction and calculate their corresponding geometric curvature to construct a weighting function:

[0032]

[0033] Where k max is the maximum curvature of the cross section, and 'a' is the coefficient of strongest cohesion.

[0034] By integrating and normalizing the weight function, a monotonically increasing mapping function is constructed to generate a smooth node distribution.

[0035]

[0036] Here, F(ξ) is a mapping from 0 to 1;

[0037] By uniformly selecting points in the parameter space and obtaining non-uniform parameter coordinates through inverse mapping,

[0038]

[0039] These ξ non-uniform k Substituting these values ​​into the geometric interpolation formula yields the node coordinates in physical space. Since F(ξ) grows faster in regions of high curvature, its inverse function will compress the uniform sequence here, resulting in denser nodes.

[0040] In the above technical solution, the merging of duplicate nodes includes:

[0041] Find and mark duplicate nodes, and record the correspondence between all duplicate nodes and the master node. Discretize the 3D coordinate quantization step size of each node, calculate its hash key value, and group all nodes according to their hash key values ​​into different buckets of the hash table. Each bucket stores a set of nodes with the same hash key. For nodes in the same bucket, perform precise coordinate comparison: if the coordinate difference between two nodes is less than the quantization step size in all three directions, it is determined to be a duplicate node, and its master-slave relationship is recorded.

[0042] Iterate through the local node list of all blocks, and check for each node whether it has a subordinate record in the conflict map. If it is a subordinate node, skip the node. If it is a master node or an independent node that is not marked as a duplicate, assign it a unique global number and write it into the global node list in order.

[0043] Iterate through the local element list of all blocks, and replace the eight local node numbers of each hexahedral element with the corresponding global numbers through the global mapping table; update the global element list to achieve seamless mesh connection.

[0044] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0045] Taking a typical airfoil as an example, this invention generates a hexahedral mesh strictly composed of all hexahedral elements, with element quality meeting aerospace structural analysis standards. Node deduplication significantly reduces redundant data, and the mesh topology is strictly consistent, supporting high-precision aeroelastic coupling simulation.

[0046] This invention solves the mesh degradation and efficiency bottleneck of the traditional sweeping method through parametric mapping and automated data processing, providing an efficient and reliable mesh generation tool for the digital design of aircraft. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0048] Figure 1 This is a schematic diagram of the overall process.

[0049] Figure 2 This is a schematic diagram of the control points of the flange root section.

[0050] Figure 3 This is a schematic diagram of the control points of the wingtip section.

[0051] Figure 4 This is a schematic diagram of the mesh edge of the wing root section.

[0052] Figure 5 This is a schematic diagram of the wingtip cross-section grid edge.

[0053] Figure 6 This is a schematic diagram of the wing root topology partitioning.

[0054] Figure 7 This is a schematic diagram of wingtip topology.

[0055] Figure 8 This is a diagram of three-dimensional block partitioning.

[0056] Figure 9 This is a schematic diagram of the wing mesh.

[0057] The attached diagram shows the markings and corresponding component names:

[0058] 1 is wing root control point one, 2 is wing root control point two, 3 is wing root control point three, 4 is wing root control point four, 5 is wing root control point five, 6 is wing root control point six, 7 is wing root control point seven, 8 is wing root control point eight, 9 is wing root control point nine, 10 is wing root control point ten, 11 is wingtip control point eleven, 12 is wingtip control point twelve, 13 is wingtip control point thirteen, 14 is wingtip control point fourteen, 15 is wingtip control point fifteen, 16 is wingtip control point sixteen, 17 is wingtip control point seventeen, 18 is wingtip control point sixteen, 18 is wingtip control point sixteen, 19 is wingtip control point sixteen, 10 is wingtip control point ten, 11 is wingtip control point eleven, 12 is wingtip control point twelve, 13 is wingtip control point thirteen, 14 is wingtip control point fourteen, 15 is wingtip control point fifteen, 16 is wingtip control point sixteen, 17 is wingtip control point seventeen, 18 is wingtip control point sixteen, 19 is wingtip control point sixteen, 10 is wingtip control point sixteen, 11 is wingtip control point sixteen, 12 is wingtip control point sixteen, 19 is wingtip control point sixteen, 10 is wingtip control point sixteen, 11 is wingtip control point sixteen, 19 is wingtip control point sixteen, 10 ... Wingtip control point 18, 19 is wingtip control point 19, 20 is wingtip control point 20, 21 is wing root grid edge 1, 22 is wing root grid edge 2, 23 is wing root grid edge 3, 24 is wing root grid edge 4, 25 is wing root grid edge 5, 26 is wing root grid edge 6, 27 is wing root grid edge 7, 28 is wing root grid edge 8, 29 is wing root grid edge 9, 30 is wing root grid edge 10, 31 is wing root grid edge 11, 32 is wing root grid edge 12, 33 is wing root grid edge 13, 34 is wing root grid edge 14. 35 is the wing root grid edge 15, 36 is the wingtip grid edge 16, 37 is the wingtip grid edge 17, 38 is the wingtip grid edge 18, 39 is the wingtip grid edge 19, 40 is the wingtip grid edge 20, 41 is the wingtip grid edge 21, 42 is the wingtip grid edge 22, 43 is the wingtip grid edge 23, 44 is the wingtip grid edge 24, 45 is the wingtip grid edge 25, 46 is the wingtip grid edge 26, 47 is the wingtip grid edge 27, 48 is the wingtip grid edge 28, 49 is the wingtip grid edge 29, and 50 is the wingtip grid edge... 30. 51 is wing root grid surface one, 52 is wing root grid surface two, 53 is wing root grid surface three, 54 is wing root grid surface four, 55 is wing root grid surface five, 56 is wing root grid surface six, 57 is wingtip grid surface seven, 58 is wingtip grid surface eight, 59 is wingtip grid surface nine, 60 is wingtip grid surface ten, 61 is wingtip grid surface eleven, 62 is wingtip grid surface twelve, 63 is grid block one, 64 is grid block two, 65 is grid block three, 66 is grid block four, 67 is grid block five, and 68 is grid block six. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention. Example 1

[0060] Combined with appendix Figure 1 ~Appendix Figure 9 The present invention will be further described in detail with a specific example of finite element mesh generation for a hexahedral wing structure, NACA2312. This example is descriptive and not limiting.

[0061] Step 1: Establish the two-dimensional mesh topology of the wing root and wing tip sections.

[0062] Two-dimensional quadrilateral meshes are generated at the wing root and wingtip sections of the wing, and the topology is generated using a two-dimensional airfoil mesh generation method, specifically including:

[0063] Input data standardization: Discrete point data at the wing root and wingtip must meet the following requirements: same starting point (trailing edge of upper wing surface), same arrangement order (counter-clockwise), and same number of discrete points. .

[0064] Parametric modeling alignment: Perform curve distance parametric modeling as described in claim 3 on the wing root and wing tip sections respectively to ensure the interpolation basis function (Wendland's C2) and the tight support radius of the two sections. The rules for defining control points are consistent.

[0065] Control points are aligned with grid edges: the ten control corner points (six surface points and four internal points) of the wing root and wing tip sections are numbered one-to-one, and the distribution strategy of the fifteen grid edges (six contour edges, five intersecting edges, and four internal edges) is strictly synchronized.

[0066] Step 2: Topology control points correspond to spanwise grid lines.

[0067] Connect the corresponding control points of the wing root and wingtip sections to generate spanwise grid lines.

[0068] The wing root grid edge 1131 is formed by connecting the wing root control point 11 and the wing tip control point 1111 with a straight line;

[0069] The wing root grid edge 12.32 is formed by connecting the wing root control point 22 and the wing tip control point 12.12 with a straight line.

[0070] The wing root grid edge 1333 is formed by connecting the wing root control point 33 and the wing tip control point 1313 with a straight line;

[0071] The wing root grid edge 14.34 is formed by connecting the wing root control point 4.4 and the wing tip control point 14.14 with a straight line.

[0072] The wing root grid edge 15.35 is formed by connecting the wing root control point 5.5 and the wing tip control point 15.15 with a straight line;

[0073] The wingtip grid edge 1636 is formed by connecting the wing root control point 66 and the wingtip control point 1616 with a straight line;

[0074] The wingtip grid edge 17.37 is formed by connecting the wing root control point 7.7 and the wingtip control point 17.17 with a straight line;

[0075] The wingtip grid edge 1838 is formed by connecting the wing root control point 88 and the wingtip control point 1818 with a straight line;

[0076] The wingtip grid edge 1939 is formed by connecting the wing root control point 99 and the wingtip control point 1919 with a straight line;

[0077] The wingtip grid edge 2040 is formed by connecting the wing root control point 10 and the wingtip control point 2020 with a straight line.

[0078] Step 3: Constructing the quadrilateral mesh.

[0079] A quadrilateral mesh is generated by four edges (two longitudinal edges + two cross-sectional edges), resulting in a total of fifteen faces:

[0080] External Surfaces: Six surfaces cover the leading edge, trailing edge, and upper and lower wing surfaces. The ten surfaces at the wing root are numbered grid surfaces 1-6, and the ten surfaces at the wingtip are numbered grid surfaces 7-12. The external wing surfaces are numbered 13-18, among which...

[0081] The wing root grid edge 121, the wing tip grid edge 1636, and the spanwise grid edge 31 and grid edge 32 constitute grid surface 13;

[0082] The wing root grid edge 22 and the wing tip grid edge 1737, along with the spanwise grid edges 32 and 33, constitute grid surface 14;

[0083] The wing root grid edge 323, the wing tip grid edge 1838, and the spanwise grid edges 33 and 34 constitute grid surface 15;

[0084] The wing root grid edge 424, the wing tip grid edge 1939, and the spanwise grid edges 34 and 35 constitute grid surface 16;

[0085] The wing root grid edge 525 and the wing tip grid edge 2040, along with the spanwise grid edges 35 and 36, constitute grid surface 17;

[0086] The wing root grid edge 626, the wing tip grid edge 2141, and the spanwise grid edges 36 and 37 constitute grid surface 18.

[0087] Internal skeleton surfaces: Nine surfaces constitute the internal support structure of the wing.

[0088] The internal skeleton is numbered 19-27, among which,

[0089] The root grid edge 7.27, the wingtip grid edge 2.2.42, the spanwise grid edge 3.2, and the grid edge 3.7 constitute grid surface 19;

[0090] The root grid edge 828, the wingtip grid edge 2343, and the spanwise grid edge 38 and grid edge 33 constitute the grid surface 20;

[0091] The root grid edge 929, the wingtip grid edge 2444, and the spanwise grid edge 39 and grid edge 34 constitute grid surface 21;

[0092] The root grid edge 10.30, the wingtip grid edge 25.45, the spanwise grid edge 40, and the grid edge 35 constitute the grid surface 22;

[0093] The root grid edge 11.31, the wingtip grid edge 26.46, the spanwise grid edge 32, and the grid edge 35 constitute grid surface 23;

[0094] The wing root grid edge 12.32, the wing tip grid edge 27.47, and the spanwise grid edge 38 and grid edge 37 constitute grid surface 24.

[0095] The root grid edge 13.33, the wingtip grid edge 28.48, the spanwise grid edge 38, and the grid edge 39 constitute grid surface 25;

[0096] The root grid edge 14.34, the wingtip grid edge 29.49, the spanwise grid edge 39, and the grid edge 40 constitute grid surface 26;

[0097] The wing root grid edge 15.35, the wing tip grid edge 30.50, the spanwise grid edge 37, and the grid edge 40 constitute the grid surface 27.

[0098] Step 4: Divide the three-dimensional blocks.

[0099] The wing is divided into six hexahedrons, each block consisting of six faces.

[0100] Shell blocks: The four blocks correspond to the leading edge, trailing edge, and upper and lower wing surfaces, respectively, including:

[0101] The trailing edge grid block 63 is composed of wing root grid surface 1 51, wing tip grid surface 7 57, grid surface 13, grid surface 17, grid surface 18, and grid surface 2 and 3.

[0102] The upper wing surface is composed of mesh block 2 64, which consists of wing root mesh surface 2 52, wingtip mesh surface 8 58, mesh surface 14, mesh surface 2 4, mesh surface 19, and mesh surface 20.

[0103] The leading edge is composed of grid block 365, which consists of grid surface 353 at the wing root, grid surface 959 at the wingtip, grid surface 20, grid surface 21, grid surface 25, and grid surface 15.

[0104] The lower wing surface is composed of grid block 466, which consists of wing root grid surface 454, wingtip grid surface 1060, grid surface 16, grid surface 26, grid surface 21, and grid surface 22.

[0105] Internal Blocks: Two blocks fill the wing interior, bounded by the frame surface and the internal edges of the cross-section. Each block contains two interface regions and a spanwise surface connecting them, including:

[0106] The internal grid block 57 is composed of wing root grid surface 55, wingtip grid surface 11 61, grid surface 19, grid surface 23, grid surface 27, and grid surface 22.

[0107] The internal grid block 68 is composed of wing root grid surface 656, wing tip grid surface 1262, grid surface 24, grid surface 26, grid surface 27, and grid surface 25.

[0108] Step 5: Generate mesh using 3D over-limit interpolation (TFI3D).

[0109] Perform trilinear interpolation on each 3D block to generate the coordinates of its internal nodes:

[0110] Parametric coordinate definition: Points inside the block are defined by... Parameterization, where:

[0111] Along the chord from the trailing edge to the leading edge

[0112] : Along the normal direction from the airfoil surface to the interior

[0113] Along the span from wing root to wingtip

[0114] Tensor product interpolation formula:

[0115]

[0116] In the formula Given the coordinates of the eight corner points of the block, the linear basis function is defined as follows:

[0117] Node distribution control: The number of grid points in the chord, normal, and span directions of each block are set as follows: , , The total number of nodes is 20 × 15 × 10 = 3000.

[0118] Step Six: Constructing the list of nodes and cells.

[0119] In the specific implementation of generating a 3D airfoil hexahedral mesh, taking the NACA0012 airfoil (chord length c=1.5, span 6.0m) as an example, the node deduplication and element construction process is achieved through the following steps: First, the main wing section and winglet generated by parametric sweeping contain 1,024 and 576 local nodes respectively, with coordinate ranges of respectively. , and , The block elements are arranged according to the rule of "bottom face counterclockwise → top face counterclockwise". Then, a spatial hash algorithm is used to discretize the node coordinates according to the quantization step size. After consolidation, the total number of global nodes is reduced from 1,600 (1,024 + 576) to 1,352 (a reduction of 15.5%), and the mapping table is updated to ensure consistency of element node references. Finally, the global element list is reconstructed into 2,520 hexahedral elements, with a Jacobian determinant mean of 0.84 and an alignment error of less than 0.9 for shared face nodes of adjacent blocks, meeting the finite element solver compatibility requirements.

[0120] Step 7: Find and mark duplicate nodes.

[0121] During the processing of repetitive nodes in the 3D model, the system uses the wing chord length c as a reference and employs a spatial hash algorithm to process the 3D coordinates of each block node. according to The quantization step size is discretized to generate hash key values. As a spatial index, after grouping all nodes into corresponding buckets using a hash table based on their key values, a second precise check is performed on each node within each bucket: when the block... Local nodes (0.123456c, 0.785432c, 0.214569c) and the block of When the coordinate differences (0.123457c, 0.785433c, 0.214568c) in the x, y, and z directions are all less than Δ, the system determines that the two are duplicate nodes and sets them accordingly. The global ID is mapped to the master node. Simultaneously, the corresponding relationships are recorded in the conflict mapping table. This process, through a double verification mechanism, maintains geometric accuracy while reducing the computational complexity of matching tens of millions of nodes from... Down to This enables efficient topology reconstruction.

[0122] Step 8: Update the node list.

[0123] During the global node integration phase, the system traverses the local node data of all blocks and filters master nodes and independent nodes based on the conflict mapping table. The global mapping table synchronously records the correspondence between local nodes of the block and their global numbers, such as... , To ensure that the cell topology passes during subsequent mesh assembly. It can be correctly associated with global node 1. This mechanism compresses scattered local node data into a non-redundant global list, and... The complexity is reduced to a level of one billion nodes, enabling unified grouping and cross-block index mapping.

[0124] Step 9: Update the cell list.

[0125] During the global unit integration phase, the system traverses the local unit data of each block and completes node replacement through a global mapping table. The system simultaneously performs shared surface verification; if adjacent blocks... The node sequence after unit mapping is But it is similar to Shared surface nodes of the cell Coordinate offset occurs (e.g.) Coordinates 0.500012 and 0.500000 Deviation exceeds If the problem is identified as "grid fault - block", then the anomaly backtracking mechanism is triggered, and the problem is marked as "grid fault - block". "Interface", and associated with the sweep parameter group. The stretching coefficient is abnormal. The final output global element list retains the original block affiliation information, ensuring that a large number of hexahedral elements achieve seamless topological connections across block boundaries while preserving the block modeling features.

[0126] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for automatically generating a finite element model of a wing's hexahedral element structure, characterized in that... Extending the 2D airfoil quadrilateral mesh to 3D space using parametric sweeping techniques includes: A radial basis interpolation model driven by curve distance is used to generate topologically consistent quadrilateral meshes at the wing root and wingtip. By using control point alignment and spanwise over-limit interpolation, geometric and topological mapping from two dimensions to three dimensions is achieved; The wing is divided into several parameterized blocks, and internal nodes are generated using tensor product interpolation. Based on the spatial hashing algorithm, a globally consistent hexahedral cell topology is constructed.

2. The method for automatically generating a finite element model of a wing hexahedral unit structure according to claim 1, characterized in that... The implementation of the two-dimensional to three-dimensional geometric and topological mapping includes: Two-dimensional quadrilateral meshes are generated at the wing root and wingtip sections of the wing to establish a two-dimensional mesh topology. The three-dimensional coordinates of each grid point are obtained by geometric transformation of the wing root, wing tip and airfoil profiles, and continuous spanwise grid lines are constructed by connecting control points and grid points; A three-dimensional curved quadrilateral mesh surface is constructed by two spanning edges and two cross-sectional edges.

3. The method for automatically generating a finite element model of a wing hexahedral unit structure according to claim 2, characterized in that: For wings with sweep angles or dihedral angles, the coordinates of the control points are mapped according to the aerodynamic shape; The spanwise grid lines are connected by straight lines or parametric curves. Control points corresponding to the wing root, wing tip, and airfoil profile are selected as endpoints, and several intermediate control points are inserted according to the geometric change trend.

4. The method for automatically generating a finite element model of a wing hexahedral element structure according to claim 1, characterized in that... The parameterized block includes: The airfoil is divided into multiple hexahedral blocks to construct a three-dimensional mesh block by using the characteristic mesh surfaces of the airfoil section, chordal surface, and span. Perform trilinear interpolation on each 3D mesh block, define parameter coordinates, and generate a 3D mesh volume.

5. The method for automatically generating a finite element model of a wing hexahedral element structure according to claim 4, characterized in that: The hexahedral block is composed of six characteristic mesh surfaces, including the wing root section, the wingtip section, and the spanwise surface connecting the two.

6. The method for automatically generating a finite element model of a wing hexahedral element structure according to claim 4, characterized in that: The hexahedral block is divided into an inner block and an outer block. The outer block covers the outer aerodynamic surface of the wing, and the inner block fills the wing structure, which is composed of a skeleton surface and an internal cross-sectional area surface.

7. The method for automatically generating a finite element model of a wing hexahedral element structure according to claim 4, characterized in that... The node distribution corresponding to the trilinear interpolation is dynamically adjusted according to the wing geometry, including: Curvature adaptation: In high curvature regions, the node distribution parameter is set to a smaller value to refine the mesh, while in flat regions such as the middle of the wing surface, it is set to a larger value to reduce the mesh density. The grid point distribution is automatically adjusted by using an interpolation function to adjust the node distance based on the geometric characteristics of the airfoil's curvature radius and sweep angle.

8. The method for automatically generating a finite element model of a wing hexahedral element structure according to claim 7, characterized in that... The setting of the smaller value includes: Select a typical airfoil profile and load condition; In the chord direction, a series of meshes with gradually increasing node numbers are generated and structural static analysis is performed. Record the key metrics for each grid density. When the rate of change of the target result caused by two consecutive encryptions is lower than the preset convergence tolerance, the mesh density is judged to meet the requirements.

9. The method for automatically generating a finite element model of a wing hexahedral unit structure according to claim 7, characterized in that... The construction of the interpolation function includes: Discretize a series of sample points along the chord direction and calculate their corresponding geometric curvature to construct a weighting function; By integrating and normalizing the weight function, a monotonically increasing mapping function is constructed. By taking points uniformly in the parameter space, non-uniform parameter coordinates can be obtained by inverting the mapping.

10. The method for automatically generating a finite element model of a wing hexahedral unit structure according to claim 1, characterized in that... The merged duplicate nodes include: Find and mark duplicate nodes, and record the correspondence between all duplicate nodes and the primary node; Iterate through the list of local nodes of all blocks and check whether each node has a subordinate record in the conflict mapping table; Iterate through the list of local elements of all blocks, and replace the eight local node numbers of each hexahedral element with the corresponding global numbers using a global mapping table.

Images