Key area restoration parameter mapping method based on dynamic grid local re-subdivision

Abstract

The invention discloses a key area restoration parameter mapping method based on dynamic grid local re-subdivision, and belongs to the technical field of engineering simulation. The method comprises the following steps that octree and quadtree partitions of an old grid are searched for units of a re-subdivision area on a new grid; searching an old grid unit where the degree-of-freedom coordinate point is located in the unit array of the partition, and then interpolating the physical field through a primary function; and calculating the value of the freedom degree coordinate point of the new grid. Mapping operation is carried out in a key area near the moving part in a concentrated mode; performing a moving restoration operation on a re-subdivision area near the boundary of the moving part, and then performing search mapping; and a simplified treatment mode is adopted for the air domain. According to the method, the unit search range and the calculation overhead in the physical field parameter mapping process are reduced, the dynamic grid physical field mapping efficiency is improved, the accuracy and the consistency of physical field transmission in a key area are guaranteed, and efficient and accurate physical field mapping and comprehensive improvement of engineering simulation efficiency are achieved.

Description

Technical Field

[0001] This invention relates to a key region restoration parameter mapping method based on dynamic mesh local re-subdivision, belonging to the field of engineering simulation technology. Background Technology

[0002] In engineering simulation, especially in complex scenarios involving computational domain or boundary motion, dynamic mesh technology needs to adapt to the changing characteristics of boundaries and regions, thus creating a requirement for mapping physical field solutions from old meshes to new meshes. This mapping process incurs enormous computational overhead, significantly impacting simulation efficiency and becoming a key bottleneck restricting the effectiveness of engineering simulations.

[0003] To address the aforementioned issues, existing technologies primarily employ local mesh repartitioning and block mapping methods for optimization. However, mapping methods involve a large amount of upfront work; establishing geometric blocks and mapping relationships requires significant manual intervention and expertise, resulting in substantial and time-consuming preliminary work. Furthermore, they exhibit poor adaptability to complex topologies, making it difficult to achieve reasonable block operations for extremely complex geometries. Additionally, matching of mesh nodes between different blocks must be ensured at block boundaries; otherwise, incompatible meshes will be generated, requiring additional processing to eliminate computational errors caused by mesh incompatibility.

[0004] In specific engineering applications, such as finite element electromagnetic coupling simulations, the physical field values ​​of the air region after mesh re-division are usually smaller compared to mechanical components. In this case, the core focus of the mapping process should be on the air domain near the moving mechanical components to achieve efficient and accurate physical field transfer.

[0005] Existing technologies still face the aforementioned common technical shortcomings in this specific scenario, making it difficult to balance mapping efficiency and accuracy. Summary of the Invention

[0006] To address the problems existing in the background technology, the present invention provides a key region restoration parameter mapping method based on dynamic mesh local re-subdivision.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a key region restoration parameter mapping method based on dynamic mesh local re-subdivision, the method comprising the following steps: S1: In the local re-mesh, the mesh elements and nodes of the re-mesh region are arranged after the mesh and nodes of the non-re-mesh region; the elements and nodes of the non-re-mesh region remain unchanged, and the physical parameters of the element degrees of freedom are directly copied; the elements and mesh of the re-mesh region are regenerated, and the element physical parameters of the region are partitioned and interpolated. S101: For the cells in the repartitioned region on the new mesh, based on the coordinates corresponding to the degrees of freedom and using the maximum side length of the cell as the reference, find the octree and quadtree partitions of the old mesh. S102: In the found cell array of the partition, find the old grid cell where the degree of freedom coordinate point is located, and then interpolate the physical field through the cell basis function; S10201: The first level of quadtree grouping is the bounding box of the largest rectangular area grid, and the cell array of this level contains all grid cells; S10202: The bounding boxes of the first level of the quadtree are partitioned using the binary search method. The grid cells are classified according to the partition in which their own bounding boxes are located. The criterion for stopping the partitioning is a multiple of the cell side length. S10203: After the region mesh is re-divided, calculate the physical parameters of a node in the new element.

[0008] S1020301: Determine the quadtree partition where the node is located based on the node's coordinates. The determination starts from the second level of the quadtree and proceeds level by level until the partition of the last level is found. S1020302: In the single-element group of this partition, determine whether the node is contained within the cell for each cell; S1020303: After finding the element containing the node, use the element's degree of freedom field values ​​and the element's basis functions to calculate the value of the node within the element.

[0009] S103: The values ​​of the coordinate points of the new mesh degrees of freedom are calculated through interpolation mapping, and the interpolation process is concentrated in the key area near the moving parts.

[0010] S2: Perform restoration processing on key areas near moving parts to correct the field distribution near the boundary of the moving area.

[0011] S201: Focus mapping operations on key areas near moving parts; S202: When performing mapping, first perform a move restoration operation on the re-divided region near the boundary of the moving part, and then perform a lookup mapping. S20201: Restore the node coordinates to their original positions, so that the critical region returns to the region in the initial mesh where there are valid physical field values; S20202: Using the restored node coordinates, find the element containing the point in the initial mesh. After finding the corresponding element, use the degree of freedom field value of the element and the basis function of the element to perform interpolation calculation to obtain the physical field value of the node in the key region. S20203: For the 0 value region far from the motion area, set the reference value according to the specific simulation scenario.

[0012] S203: For air regions far from moving parts, a simplified treatment method is adopted.

[0013] Compared with the prior art, the beneficial effects of the present invention are: This invention effectively reduces the cell search range and computational overhead in the physical field parameter mapping process by employing a strategy of locally re-subdividing mesh cells and nodes in an ordered arrangement, combined with an octree / quadtree partitioned interpolation mapping mechanism, thus significantly improving the efficiency of physical field mapping for moving meshes. For key regions near moving parts, coordinate movement restoration and precise interpolation mapping are implemented, eliminating zero-value anomalies in the field distribution near the boundary and ensuring the accuracy and consistency of physical field transmission in key regions. Simultaneously, by employing a simplified processing strategy for the air domain far from the moving region, computational resource allocation is optimized while maintaining the mapping accuracy of the core region. This overcomes the technical shortcomings of traditional methods, such as high manual intervention, weak adaptability to complex topologies, and cumbersome handling of non-coordinated mesh errors, achieving a comprehensive improvement in efficient and accurate physical field mapping and engineering simulation performance. Attached Figure Description

[0014] Figure 1 Here are schematic diagrams of the quadtree layering of the original grid, where: (a) is a schematic diagram of the first layer of the quadtree, and (b) is a schematic diagram of the second layer of the quadtree; Figure 2 This is a schematic diagram showing the position of element 1 in the original element mesh corresponding to node 2 after the mesh movement; Figure 3 The diagram shows the key region restoration process of the repartitioned region, where: (a) is the initial region diagram, (b) is the 0-value area left after the moved region is moved, (c) is the key region restoration diagram, and (d) is the key region value repair diagram. Figure 4 This is a schematic diagram of a two-dimensional planar model of Example 1; Figure 5 These are schematic diagrams of the mesh before and after re-subdivision of the two-dimensional planar model in Example 1, where: (a) is a schematic diagram of the initial mesh, and (b) is a schematic diagram of the re-subdivided mesh; Figure 6 This is the field distribution process before and after the re-subdivision and mapping of the two-dimensional planar model in Example 1, where: (a) is a schematic diagram of the initial field distribution, and (b) is a schematic diagram of the mapped field distribution. Detailed Implementation

[0015] 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 only some embodiments of the invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0016] A key region restoration parameter mapping method based on local re-meshment of dynamic meshes is proposed, primarily serving engineering simulation fields requiring handling large deformations, dynamic boundaries, aerospace applications, and multiphysics coupling analysis. The method includes the following steps: S1: In the local re-mesh, the mesh elements and nodes of the re-mesh region are arranged after the mesh and nodes of the non-re-mesh region; the elements and nodes of the non-re-mesh region remain unchanged, and the physical parameters of the element degrees of freedom are directly copied; the elements and mesh of the re-mesh region are regenerated, and the element physical parameters of the region are partitioned and interpolated. S101: For the cells in the repartitioned region on the new mesh, based on the coordinates corresponding to the degrees of freedom and using the maximum side length of the cell as the reference, the octree and quadtree partitions of the old mesh are found by quickly searching for cells through partitioning. S102: In the cell array of the found partition, find the old grid cell where the degree of freedom coordinate point is located, and then interpolate the physical field through the basis function of the cell to avoid unreasonable partitioning under complex structure and problems in the preparation of interpolation function in the early stage, thereby reducing the complexity of interpolation mapping. S10201: As Figure 1 As shown in (a), the largest rectangular bounding box of the region grid is used as the first layer of the quadtree grouping, and the cell array of this layer contains all grid cells. S10202: The bounding boxes of the first level of the quadtree are partitioned using a binary search method. Mesh cells are classified according to the partition in which their own bounding boxes belong, such as... Figure 1 As shown in (b), cell 1 in the figure belongs to the cell array of partition 4. The criterion for stopping the partition is a multiple of the cell side length. S10203: After the region mesh is re-divided, calculate the new element (assuming there is an element such as...). Figure 2 The physical parameters of a node are shown in Unit 2.

[0017] S1020301: Determine the quadtree partition where the node is located based on the node's coordinates. The determination starts from the second level of the quadtree and proceeds level by level until the partition of the last level is found. S1020302: In the single-element group of this partition, determine whether the node is contained within the cell for each cell; S1020303: After finding the element containing the node, use the element's degree of freedom field values ​​and the element's basis functions to calculate the value of the node within the element.

[0018] S103: The values ​​of the coordinate points of the new mesh degrees of freedom are calculated through interpolation mapping, and the interpolation process is mainly concentrated in the key areas near the moving parts.

[0019] S2: Perform restoration processing on key areas near moving parts to correct the field distribution near the boundary of the moving area.

[0020] S201: Focus mapping operations on key areas near moving parts; S202: When performing mapping, first perform a move restoration operation on the re-divided area near the boundary of the moving part, and then perform a lookup mapping to reduce the mapping error in the nearby area. The distribution of the initial region is as follows Figure 3 As shown in (a), after the motion area moves upward, as Figure 3 As shown in (b), in the moving mesh, the meshes of other regions and the moving region remain unchanged, and the degrees of freedom are also unchanged, so the physical field can be directly copied. However, the mesh of the re-partitioned region has changed. When using the partitioned interpolation mapping method, it is found that no elements can be found at the original position of the moving region. This is equivalent to the existence of a 0-value region below the moved region. The moving region is generally an important component such as an armature, and the appearance of a 0-value region on its boundary will have an adverse effect on the results. Therefore, this invention uses a method of restoring the key region to repair the field distribution near the boundary of the moving region. Step S202 includes the following steps: S20201: Before performing mapping cell lookup on the re-mesh nodes in the key area, the node coordinates are first restored to their original positions, such as... Figure 3 As shown in (c), the critical region is brought back to the region where there are valid physical field values ​​in the initial grid; S20202: Using the restored node coordinates, find the element containing the point in the initial mesh (old mesh). After finding the corresponding element, use the degree of freedom field value of the element and the basis function of the element to perform interpolation calculation to obtain the physical field value of the node in the key region (that is, the field value of the key region is mapped from the field at the same position below the motion region in the initial mesh). S20203: For the 0-value region far from the motion area, set reference values ​​according to the specific simulation scenario to improve the physical field mapping results of the entire region.

[0021] S203: For air regions far from moving parts, a simplified processing method is adopted to improve mapping efficiency.

[0022] Example 1: The effect of repartitioning is demonstrated using a two-dimensional planar model. For example... Figure 4 As shown, the model includes a rectangular repartitioned region, a rectangular moving region contained within it, and other surrounding regions.

[0023] First, mesh the model. The initial mesh is as follows: Figure 5As shown in (a), the number of mesh nodes is 9323 and the number of elements is 18304; the middle moving region moves upward, and the re-mesh is as follows. Figure 5 As shown in (b), the number of grid nodes is 9299 and the number of cells is 18256.

[0024] The field distribution of the initial grid is as follows Figure 6 As shown in (a), the field values ​​are mainly distributed in the central moving region, while the values ​​in the surrounding re-partitioned regions are very small. The field distribution of the mapped mesh is as follows. Figure 6 As shown in (b), the meshes of other regions and the moving region remain unchanged, and the field values ​​of the degrees of freedom are directly copied accordingly; the field values ​​of the re-divided regions near the moving region are mapped by partitioned interpolation. For the empty regions left after the moving region, small values ​​are directly assigned. The final mapped field ensures the field value distribution near important components.

[0025] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0026] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for key region recovery parameter mapping based on local re-meshing, characterized in that: The method comprises the following steps: S1: in the local re-meshing, the grid elements and nodes of the re-meshing area are arranged after the grids and nodes of the non-re-meshing area; the elements and nodes of the non-re-meshing area remain unchanged, and the physical parameters on the element degrees of freedom are directly copied; the elements and grids of the re-meshing area are regenerated, and the element physical parameters of the area are partitioned and interpolated and mapped; S2: the key area near the moving part is processed to correct the field distribution near the moving area boundary.

2. The key region restoration parameter mapping method based on local re-mesh subdivision of claim 1, wherein: The S1 comprises the following steps: S101: for the elements of the re-meshing area on the new grid, the octree and quadtree partitions of the old grid are searched according to the coordinates corresponding to the degrees of freedom and taking the maximum edge length of the element as the reference; S102: the old grid element where the degree of freedom coordinate point is located is searched in the element array of the found partition, and then the physical field is interpolated through the element base function; S103: the value of the new grid degree of freedom coordinate point is calculated through the interpolation mapping, and the interpolation process is concentrated in the key area near the moving part.

3. The key region restoration parameter mapping method based on local re-mesh subdivision of claim 2, wherein: The S102 comprises the following steps: S10201: the maximum rectangular bounding box of the area grid is taken as the first layer of the quadtree grouping, and the element array of the layer contains all the grid elements; S10202: the bounding box of the first layer of the quadtree is partitioned by using the dichotomy method, the grid elements are classified according to the partition where the bounding box of the element is located, and the stopping criterion of the partition is the multiple of the element edge length; S10203: after the area grid is re-meshed, the physical parameters of a node of the new element are calculated.

4. The key region restoration parameter mapping method based on local re-mesh subdivision of claim 3, wherein: The S10203 comprises the following steps: S1020301: the quadtree partition where the node is located is judged according to the coordinates of the node, the judgment is performed layer by layer from the second layer of the quadtree, and the partition of the last layer is found; S1020302: whether the node is contained in the element is judged in the single element group of the partition; S1020303: after the element containing the node is found, the value of the node in the element is calculated by using the degree of freedom field value of the element and the element base function.

5. The key region restoration parameter mapping method based on local re-mesh subdivision of claim 1, wherein: The S2 comprises the following steps: S201: the mapping operation is concentrated in the key area near the moving part; S202: when the mapping is performed, the moving recovery operation is first performed on the re-meshing area near the boundary of the moving part, and then the finding and mapping are performed; S203: for the air domain far away from the moving part, a simplified processing mode is adopted.

6. The key region restoration parameter mapping method based on local re-mesh subdivision of claim 5, wherein: The S202 comprises the following steps: S20201: the node coordinates are recovered to the original positions, so that the key area returns to the area range in the initial grid where the effective physical field values exist; S20202: the element containing the point is found in the initial grid by using the recovered node coordinates, and the physical field value of the node in the key area is obtained by using the interpolation calculation of the degree of freedom field value of the element and the element base function after the corresponding element is found; S20203: for the 0-value area far away from the moving area, a reference value is set according to the specific simulation scene.

Images