Boundary layer grid generation simulation method, system, equipment and medium

Abstract

The invention provides a boundary layer grid generation simulation method, system and device and a medium, and relates to the technical field of simulation, and the method comprises the steps: inputting a geometric curved surface of a to-be-analyzed wall surface; controlling each node in the bottom surface of the first boundary layer grid to advance outwards along the respective advancing direction to generate the first boundary layer grid; determining the target triangular prisms which collide with each other, performing iterative reduction on the propulsion height of the target triangular prisms, and after each reduction, performing propulsion again at the reduced propulsion height to generate a new target triangular prism until the generated target triangular prism meets a first condition; when all the target triangular prisms meet a first condition, obtaining a first boundary layer grid which is divided to be qualified; when the total number of layers of the qualified boundary layer grids reaches a preset number of layers, obtaining a boundary layer grid division result; and based on the result, performing simulation analysis on the geometric curved surface to obtain a simulation analysis result. The invention aims to improve the boundary layer grid generation efficiency and quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of simulation, in particular to a boundary layer grid generation simulation method, system, device and medium. BACKGROUND

[0002] With the rapid development of the automobile industry, the safety, environmental protection and comfort of the automobile are increasingly required. In the development process, fluid mechanics and thermodynamics problems need to be considered more comprehensively. In the development process of the whole vehicle aerodynamic and thermal dynamic performance, the whole vehicle CFD analysis technology (Computational Fluid Dynamics) is undoubtedly one of the most important tools at present.

[0003] In the whole vehicle CFD analysis technology, the pre-processing process is the first step of the whole vehicle CFD analysis, and plays a crucial role in the whole analysis process. The most core step in the pre-processing process is to discretize the continuous flow field space into a limited grid cell, that is, the flow field body grid division. When CFD is solved, the continuous partial differential equation is discretized into a linear equation set based on the flow field body grid, and the flow field result is calculated. Obviously, the quality of the flow field body grid directly determines the accuracy, efficiency and stability of the CFD solving calculation.

[0004] The mixed grid generally used in the whole vehicle CFD analysis is mainly composed of near-field grid and far-field grid, wherein the fine treatment result of the near-field grid (boundary layer grid) has a decisive influence on the accuracy of subsequent calculation. This part of the grid mainly arranges anisotropic prismatic grid near the wall surface (near the surface of the whole vehicle) to accurately capture the key physical phenomena such as velocity gradient, shear stress and turbulent characteristics at the near-wall surface. If the division is improper, it is easy to cause wall function failure, separation flow prediction deviation or turbulent model distortion and other problems.

[0005] At present, for the division of the near-field grid (boundary layer grid), the boundary layer layer advancing algorithm (that is, the technology of constructing structured prismatic grid by layering outward from the surface of the object) is generally used to realize it. This method can accurately control the density and smoothness of the boundary layer grid, so as to ensure the accuracy and stability of the near-wall solution in subsequent solving. However, in the whole vehicle CFD analysis, due to the high complexity of the boundary of the whole vehicle (front engine compartment, chassis and other regions), it is difficult to ensure the quality of grid generation by using this method, and even the problem of grid generation failure may occur under extreme conditions. SUMMARY

[0006] Therefore, the present application provides a boundary layer grid generation simulation method, system, device and medium. It aims to solve or partially solve the problems in the background art.

[0007] The first aspect of the application provides a boundary layer grid generation simulation method, the method comprising:

[0008] inputting a geometric surface of a wall surface to be analyzed, the geometric surface being composed of discrete triangular patches;

[0009] controlling each node in a first boundary layer grid bottom surface to respectively advance outward by a first advancing height along a respective advancing direction, to generate a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a layer of boundary layer grid to be currently generated;

[0010] determining target triangular prisms that collide with each other in the first boundary layer grid;

[0011] iteratively reducing the advancing height of the target triangular prisms, and after each reduction, generating new target triangular prisms by re-advancing with the reduced advancing height, until the re-advanced target triangular prisms meet a first condition;

[0012] in a case where all the re-generated target triangular prisms meet the first condition, obtaining a qualified first boundary layer grid;

[0013] in a case where the total number of layers of the qualified boundary layer grid reaches a preset number of layers, determining a boundary layer grid division result corresponding to the geometric surface;

[0014] based on the boundary layer grid division result, performing simulation analysis on the geometric surface to obtain a corresponding simulation analysis result.

[0015] The second aspect of the application provides a boundary layer grid generation simulation system, the system comprising:

[0016] a model input module configured to input a geometric surface of a wall surface to be analyzed, the geometric surface being composed of discrete triangular patches;

[0017] an advancing control module configured to control each node in a first boundary layer grid bottom surface to respectively advance outward by a first advancing height along a respective advancing direction, to generate a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a layer of boundary layer grid to be currently generated;

[0018] a target triangular prism determination module configured to determine target triangular prisms that collide with each other in the first boundary layer grid;

[0019] an advancing iterative reduction module configured to iteratively reduce the advancing height of the target triangular prisms, and after each reduction, generate new target triangular prisms by re-advancing with the reduced advancing height, until the re-advanced target triangular prisms meet a first condition;

[0020] The boundary layer generation module is configured to, in a case where all the regenerated target triangular prisms satisfy the first condition, obtain a first boundary layer grid that passes the division;

[0021] The division result determination module is configured to, in a case where the total number of the boundary layer grids that pass the division reaches a preset number of layers, determine a boundary layer grid division result corresponding to the geometric curved surface.

[0022] The simulation analysis module is configured to, based on the boundary layer grid division result, perform simulation analysis on the geometric curved surface, and obtain a corresponding simulation analysis result.

[0023] The third aspect of the present application provides an electronic device, comprising a processor, a memory, and a computer program stored on the memory and running on the processor, wherein the computer program, when executed by the processor, implements the steps of the boundary layer grid generation simulation method according to the first aspect of the present application.

[0024] The fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program, when executed by a processor, implements the steps of the boundary layer grid generation simulation method according to the first aspect of the present application.

[0025] The boundary layer grid generation simulation method provided by the present application has the following advantages:

[0026] The boundary layer grid generation simulation method provided by the present application first inputs a geometric curved surface of a wall surface to be analyzed, and the geometric curved surface is composed of discrete triangular patches. Each node in the bottom surface of a first boundary layer grid is controlled to respectively advance outward by a first advancing height along a respective advancing direction, thereby generating a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a current boundary layer to be generated. Target triangular prisms that collide with each other in the first boundary layer grid are determined. The advancing height of the target triangular prisms is iteratively reduced, and after each reduction, a new target triangular prism is generated by re-advancing with the reduced advancing height, until the re-advanced target triangular prism satisfies a first condition. In a case where all the regenerated target triangular prisms satisfy the first condition, a first boundary layer grid that passes the division is obtained. In a case where the total number of the boundary layer grids that pass the division reaches a preset number of layers, a boundary layer grid division result corresponding to the geometric curved surface is determined. Based on the boundary layer grid division result, simulation analysis is performed on the geometric curved surface, and a corresponding simulation analysis result is obtained.

[0027] Therefore, the present application realizes the generation of a complete boundary layer grid through collision detection and corresponding optimization adjustment of the advancing height, which can significantly reduce manual intervention, effectively improve the grid generation efficiency and quality, and is suitable for fluid mechanics simulation of complex geometries (such as a whole vehicle model). Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart illustrating a boundary layer mesh generation simulation method according to one embodiment of this application;

[0030] Figure 2 This is a schematic diagram illustrating the advancing direction of nodes in a boundary layer mesh generation simulation method according to an embodiment of this application;

[0031] Figure 3 This is a schematic diagram illustrating the structure of a multi-layer boundary layer mesh in a boundary layer mesh generation simulation method according to an embodiment of this application;

[0032] Figure 4 This is a schematic diagram illustrating recursive segmentation in a boundary layer mesh generation simulation method according to an embodiment of this application;

[0033] Figure 5 This is another flowchart illustrating a boundary layer mesh generation simulation method according to one embodiment of this application;

[0034] Figure 6 This is a schematic diagram of a boundary layer mesh generation simulation system according to one embodiment of this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0036] refer to Figure 1 , Figure 1 This is a flowchart illustrating a boundary layer mesh generation simulation method according to one embodiment of this application. Figure 1 As shown, the method includes:

[0037] Step S1: Input the geometric surface of the wall to be analyzed, which is composed of discrete triangular patches.

[0038] In the embodiment, first, a three-dimensional model of the whole vehicle is built, and a geometric surface of a wall surface to be analyzed in the three-dimensional model of the whole vehicle is input to a fluid dynamics analysis tool. The geometric surface of the wall surface to be analyzed in the application is composed of discrete triangular facets, and each triangular facet is composed of three nodes.

[0039] Step S2: controlling each node in the bottom surface of the first boundary layer grid to respectively advance outward by the first advancing height along the respective advancing direction, to generate a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a current boundary layer grid to be generated.

[0040] In the embodiment, before the fluid dynamics analysis of the geometric surface of the wall surface to be analyzed, the geometric surface of the wall surface to be analyzed needs to be preprocessed, that is, the boundary layer grid of the geometric surface of the wall surface to be analyzed is divided. In the process of dividing the boundary layer grid of the geometric surface in the fluid dynamics analysis, a plurality of boundary layer grids are divided to obtain a final boundary layer grid division result, and each boundary layer grid in the plurality of boundary layer grids is generated by advancing outward by a certain height from the top surface of the boundary layer grid of the previous layer. The first boundary layer grid in the application is a current boundary layer grid to be generated, that is, a boundary layer grid that is currently being generated. The geometric surface of the wall surface to be analyzed in the application is usually a closed, manifold, and non-intersecting triangular watertight surface grid, and the geometric surface is the bottom surface of the first boundary layer grid.

[0041] In the embodiment, since each boundary layer grid is generated by advancing outward by a certain height from the top surface of the boundary layer grid of the previous layer, the bottom surface of the first boundary layer grid is actually the top surface of the boundary layer grid of the previous layer adjacent to the first boundary layer grid. At the same time, each boundary layer grid is generated by advancing outward by a certain height from each triangular facet in the bottom surface of the boundary layer grid, and finally the boundary layer grid of the layer is composed of a large number of triangular prisms, and the top surface and the bottom surface of the boundary layer grid of the layer are each composed of a large number of triangular facets, and each triangular facet is composed of three nodes. Therefore, the nodes in the bottom surface of the first boundary layer grid are actually the nodes in the top surface of the boundary layer grid of the previous layer adjacent to the first boundary layer grid.

[0042] In the embodiment, each node in the bottom surface of the first boundary layer grid is respectively pushed outwards along the respective pushing direction by a first pushing height corresponding to the first boundary layer grid to generate the first boundary layer grid composed of triangular prisms. The pushing height corresponding to the boundary layer grid varies with the specific layer of the boundary layer grid currently being generated. The pushing height of the boundary layer grid of the later layer sequence is larger, for example, the pushing height for generating the second layer boundary layer grid is larger than the pushing height for generating the first layer boundary layer grid.

[0043] Step S3: determining target triangular prisms that collide with each other in the first boundary layer grid.

[0044] In the embodiment, after the first boundary layer grid is generated by step S2, the first boundary layer grid is composed of a large number of triangular prisms generated by pushing. The triangular prism is composed of triangular facets in the bottom surface and triangular facets in the top surface, and the triangular facet is composed of three nodes, and the node has coordinate position information. Therefore, based on the coordinate position information of the nodes composing the triangular prism, the application performs collision detection on all triangular prisms in the first boundary layer grid composed of a large number of triangular prisms generated by pushing, determines which triangular prisms generated by pushing will collide with each other, that is, which triangular prisms will interfere with each other, and determines the triangular prisms that collide with each other in the first boundary layer grid as target triangular prisms. As long as one triangular prism in the first boundary layer grid collides with other triangular prisms in the first boundary layer grid, the triangular prism belongs to the target triangular prism, and the other triangular prisms also collide with the triangular prism, so the other triangular prisms also belong to the target triangular prism. For example, triangular prism a1 in the first boundary layer grid collides with triangular prism a2 in the first boundary layer grid, and also collides with triangular prism a3 in the first boundary layer grid, so triangular prisms a1, a2, and a3 are all determined to be target triangular prisms. For the target triangular prism, the application will correct its pushing height to avoid further collision.

[0045] Step S4: iteratively reducing the pushing height of the target triangular prism, and after each reduction, generating new target triangular prisms by pushing with the reduced pushing height until the target triangular prisms generated by pushing meet the first condition.

[0046] In the embodiment, after the target triangular prisms that collide with each other in the first boundary layer grid are determined by step S3, the pushing correction method for each target triangular prism is the same, and one target triangular prism is taken as an example for illustration.

[0047] Specifically, the current advancing height of the target triangular prism is determined, a certain advancing height is reduced from the current advancing height based on a preset advancing height reduction strategy, a new advancing height is obtained, and then the target triangular prism is returned to the initial state, that is, the state before the target triangular prism is generated by advancing. Based on the new advancing height, the three nodes in the first boundary layer grid bottom surface corresponding to the target triangular prism are controlled to advance outward by the new advancing height along the respective advancing direction. Then it is determined whether the new target triangular prism generated by re-advancing still collides, if not, it is determined that the new target triangular prism generated by re-advancing satisfies the first condition, and no further advancing correction is performed. If the collision still occurs, the current advancing height of the new target triangular prism is determined, a certain advancing height is continuously reduced from the current advancing height based on the preset advancing height reduction strategy, a second new advancing height is obtained, and then the target triangular prism is returned to the initial state, that is, the state before the target triangular prism is generated by advancing. Based on the second new advancing height, the three nodes in the first boundary layer grid bottom surface corresponding to the target triangular prism are controlled to advance outward by the second new advancing height along the respective advancing direction. Then it is determined whether the new target triangular prism generated by re-advancing still collides, if not, it is determined that the new target triangular prism generated by re-advancing satisfies the first condition, and no further advancing correction is performed. If the collision still occurs, the current advancing height is continuously reduced based on the preset advancing height reduction strategy, and a new target triangular prism is generated again with a new recommended height, until the new target triangular prism generated by re-advancing no longer collides, then it is determined that the new target triangular prism generated by re-advancing satisfies the first condition. Or after a preset number of times of correction of the advancing height of the target triangular prism, whether the target triangular prism still collides or not, it is determined that the final generated new triangular prism corresponding to the target triangular prism satisfies the first condition, and no further advancing correction is performed. The preset number of times of advancing height correction is set to avoid continuous correction of the recommended height in some abnormal cases. The preset number of times can be set according to the actual application scenario, which is not limited here.

[0048] In the present embodiment, one optional implementation of the preset advancing height reduction strategy is to reduce 50% from the advancing height of the last round to obtain a new advancing height, and generate a new target triangular prism corresponding to the target triangular prism again with the new advancing height, until the new target triangular prism generated no longer collides, then it is determined that the final generated new target triangular prism satisfies the first condition, and no further advancing correction is performed. Or after a preset number of times of correction, it is determined that the final generated new target triangular prism satisfies the first condition, and no further advancing correction is performed.

[0049] Step S5: In the case that all the regenerated target prisms satisfy the first condition, a first boundary layer grid that passes the division is obtained.

[0050] In this embodiment, the first boundary layer grid that passes the division is obtained after all the target prisms in the first boundary layer grid satisfy the first condition. The division of the boundary layer grid of each layer is performed by the same implementation as steps S2 to S5.

[0051] Step S6: In the case that the total number of the boundary layer grids that pass the division reaches a preset number of layers, it is determined that the boundary layer grid division result corresponding to the geometric surface is obtained.

[0052] In this embodiment, the total number of the boundary layer grids required for the complete boundary layer grid division result is predefined, which is the preset number of layers predefined by the present application. In the case that the total number of the boundary layer grids that pass the division reaches the preset number of layers, it is determined that the final boundary layer grid division result corresponding to the geometric surface of the wall surface to be analyzed is obtained. The preset number of layers can be set according to the actual application scenario, which is not limited here. Another optional implementation is that the total thickness of the boundary layer grid required for the complete boundary layer grid division result is predefined, and in the case that the thickness of each boundary layer grid that passes the division is accumulated to reach the total thickness, it is determined that the final boundary layer grid division result corresponding to the geometric surface of the wall surface to be analyzed is obtained.

[0053] Step S7: Based on the boundary layer grid division result, the geometric surface is simulated and analyzed to obtain a corresponding simulation and analysis result.

[0054] In this embodiment, after the final boundary layer grid division result corresponding to the geometric surface of the wall surface to be analyzed is obtained by step S6, the outermost boundary layer grid node in the boundary layer grid division result is transitionally connected to the unstructured far-field core grid, and the Delaunay triangulation hybrid grid technology is adopted to ensure interface compatibility, so as to obtain the final division result. Based on the final division result, the geometric surface is subjected to subsequent fluid dynamics simulation and analysis to obtain a corresponding fluid dynamics simulation and analysis result. The wall surface to be analyzed can be the wall surface of each region of the vehicle that needs to be subjected to fluid dynamics analysis, such as the wall surface of the vehicle head, front engine compartment, chassis, etc.

[0055] The application provides a boundary layer grid generation simulation method. First, a geometric surface of a wall surface to be analyzed is input. The geometric surface is composed of discrete triangular facets. Each node in a first boundary layer grid bottom surface is controlled to respectively advance outward by a first advancing height along a respective advancing direction to generate a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a current boundary layer to be generated. Target triangular prisms that collide with each other in the first boundary layer grid are determined. The advancing height of the target triangular prisms is iteratively reduced, and after each reduction, the target triangular prisms are newly generated by advancing with the reduced advancing height until the newly generated target triangular prisms satisfy a first condition. In the case where all the newly generated target triangular prisms satisfy the first condition, a qualified first boundary layer grid is obtained. In the case where the total number of layers of the qualified boundary layer grid reaches a preset number of layers, a boundary layer grid division result corresponding to the geometric surface is determined. Based on the boundary layer grid division result, the geometric surface is simulated and analyzed to obtain a corresponding simulation analysis result.

[0056] Therefore, the application realizes the generation of a complete boundary layer grid through collision detection and corresponding optimization adjustment of the advancing height, which can significantly reduce manual intervention, effectively improve the grid generation efficiency and quality, and is suitable for fluid mechanics simulation of complex geometries (such as a whole vehicle model).

[0057] In combination with the above embodiments, in an implementation, the application also provides a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, the first advancing height of the first boundary layer grid is determined, including: determining a layer sequence of the current first boundary layer grid to be generated; in the case where the layer sequence is a first layer, determining a predefined first layer advancing height as the first advancing height of the first boundary layer grid; in the case where the layer sequence is a non-first layer, determining the first advancing height of the first boundary layer grid according to the first layer advancing height and a preset grid growth rate.

[0058] In the embodiment, the application predefines the first-layer boundary layer grid's advancing height, which is the first-layer advancing height, and predefines the grid growth rate, which is the preset grid growth rate. In the case that the layer sequence of the first boundary layer grid to be generated is the first layer, the first-layer advancing height is the first advancing height of the first boundary layer grid, that is, the advancing height of the first boundary layer grid to be generated. In the case that the layer sequence of the first boundary layer grid to be generated is not the first layer, the first advancing height of the first boundary layer grid to be generated is determined by the first-layer advancing height, the preset grid growth rate and the layer sequence of the first boundary layer grid to be generated in the form of a geometric progression or a double-exponential law. For example, the layer sequence of the first boundary layer grid to be generated is i, the law of the advancing height between layers of the boundary layer grid is a geometric progression, and the preset grid growth rate is β. Then, the first advancing height y i of the first boundary layer grid with the layer sequence i is y i-1 ,i = 2, 3, 4...N, and the value of the first-layer advancing height y1 is a known quantity.

[0059] In combination with the above embodiment, in an implementation, the application also provides a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, in the case that the first boundary layer grid is a first-layer boundary layer grid, the first-layer boundary layer grid's advancing height meeting the demand of the wall function of the turbulent flow model in simulation is determined by a preset algorithm, and the preset algorithm is expressed as: wherein y + represents the dimensionless wall distance, μ represents the fluid dynamic viscosity, ρ represents the fluid density, u τ represents the friction velocity, τ w represents the wall shear stress, and y1 represents the first-layer boundary layer grid's advancing height.

[0060] In the embodiment, in order to ensure that the grid center of the first-layer boundary layer grid is located in the logarithmic law layer, the application adopts the following implementation to set the first-layer advancing height of the first-layer boundary layer grid. Specifically, in the case that the first boundary layer grid is a first-layer boundary layer grid, the first-layer boundary layer grid's first-layer advancing height meeting the demand of the wall function of the turbulent flow model in simulation is determined by a preset algorithm, and the preset algorithm is expressed as: wherein y + represents the dimensionless wall distance, μ represents the fluid dynamic viscosity, ρ represents the fluid density, u τ represents the friction velocity, τ w represents the wall shear stress, and y1 represents the first-layer boundary layer grid's advancing height.

[0061] In combination with the above embodiments, in an implementation, the embodiments of the present application further provide a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, the advancing direction of a node is determined, including: determining each target triangular patch with the node as a common vertex; determining the area and normal vector of each target triangular patch according to the node position of the target triangular patch; and calculating the vertex normal of the node by a first algorithm according to the area and normal vector of each target triangular patch, wherein the first algorithm is expressed as: wherein, is the vertex normal of node j in the bottom surface of the i-th boundary layer grid, i,T is the area of triangular patch T with node j in the bottom surface of the i-th boundary layer grid as a common vertex, is the normal vector of triangular patch T with node j in the bottom surface of the i-th boundary layer grid as a common vertex, and triangular patch T belongs to the triangular patches with node j in the bottom surface of the i-th boundary layer grid as a common vertex; and the vertex normal of the node is determined as the advancing direction of the node.

[0062] In the present embodiment, the implementation of determining the advancing direction of each node in the first boundary layer grid bottom surface is the same, and one node is taken as an example for description. First, each triangular patch with the node as a common vertex is determined, and these triangular patches are referred to as target triangular patches corresponding to the node. One node will have three triangular patches with the node as a common vertex, as shown in FIG. 1. Figure 2 Figure 2 In FIG. 1, node j has three triangular patches with the node j as a common vertex, Figure 2 i,T1 , A i,T2 , A i,T3 respectively represent the areas of the three triangular patches respectively; Figure 2 respectively represent the normal vectors of the three triangular patches respectively, and the normal vector of a triangular patch refers to a vector perpendicular to the triangular patch; Figure 2 ​​​​represents the vertex normal of node j in the bottom surface of the i-th layer boundary layer grid. Then, based on the position information of each node constituting the target triangular face, the area and normal vector of the target triangular face are calculated to obtain the area and normal vector of the target triangular face. Through the same implementation, the area and normal vector of each triangular face with the node as a common vertex can be determined. After obtaining the area and normal vector of each triangular face with the node as a common vertex, the area and normal vector of each target triangular face are substituted into the first algorithm for calculation to obtain the vertex normal of the node. Finally, the vertex normal of the node obtained by calculation is determined as the advancing direction of the node. Through the same implementation, each node in the first boundary layer grid bottom surface can calculate and obtain its corresponding advancing direction, and the advancing direction of each node determined by the method is different. As shown in Figure 3 Figure 3 The structure of the multi-layer boundary layer grid of the present application is schematically shown. The bottom surface of the i-th layer boundary layer grid is actually the top surface of the (i-1)-th layer boundary layer grid, and node j in the bottom surface of the i-th layer boundary layer grid is obtained by continuously advancing node j on the wall surface.

[0063] In the present embodiment, the expression of the first algorithm is: wherein, is the vertex normal of node j in the bottom surface of the i-th layer boundary layer grid, A i,T is the area of triangular face T with node j in the bottom surface of the i-th layer boundary layer grid as a common vertex, is the normal vector of triangular face T with node j in the bottom surface of the i-th layer boundary layer grid as a common vertex; triangular face T belongs to the triangular faces with node j in the bottom surface of the i-th layer boundary layer grid as a common vertex.

[0064] In combination with the above embodiments, in one implementation, the present embodiment further provides a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, the method further comprises: correcting the first advancing height of the node in the first boundary layer grid bottom surface by a advancing height correction algorithm to obtain a corrected first advancing height; and the expression of the advancing height correction algorithm is: wherein, y i,j is the advancing height of node j in the bottom surface of the i-th layer boundary layer grid, y′ i,j is the corrected advancing height of node j in the bottom surface of the i-th layer boundary layer grid, is the maximum principal curvature of node j in the bottom surface of the i-th layer boundary layer grid, ​is the maximum value of the maximum principal curvature of all nodes in the bottom surface of the i-th layer of boundary layer grid; the advancing direction of the node in the bottom surface of the first boundary layer grid is corrected by an advancing direction correction algorithm to obtain a corrected advancing direction; the expression of the advancing direction correction algorithm is: wherein, is the advancing direction of the node j in the bottom surface of the i-th layer of boundary layer grid, is the corrected advancing direction of the node j in the bottom surface of the i-th layer of boundary layer grid, and a is a correction coefficient, i,j is the average value of the maximum principal curvature and the minimum principal curvature of the node j in the bottom surface of the i-th layer of boundary layer grid is the principal direction tangent vector corresponding to the maximum principal curvature of the node j in the bottom surface of the i-th layer of boundary layer grid; the first boundary layer grid composed of three prisms is generated by controlling each node in the bottom surface of the first boundary layer grid to respectively advance outward by the first advancing height along the respective advancing direction, including: the first boundary layer grid composed of three prisms is generated by controlling each node in the bottom surface of the first boundary layer grid to respectively advance outward by the respective corrected first advancing height along the respective corrected advancing direction.

[0065] In the present embodiment, after the first advancing height of the first boundary layer grid is determined, each node in the bottom surface of the first boundary layer grid advances by the first advancing height, that is, the advancing height of each node in the bottom surface of the first boundary layer grid is the same. In order to avoid the intersection between the layers of the multi-layer boundary layer grid, the advancing direction and the advancing height are adjusted in the strong bending area. The principle is that for the concave area in the complex geometric area, the advancing direction is outwardly extended to a certain extent, while in the convex area, it is appropriately inwardly collected, and the advancing height is reduced in the high curvature area.

[0066] The specific implementation of further adjusting the first advancing height of the first boundary layer grid and the advancing direction of each node in the bottom surface of the first boundary layer grid is: the first advancing height of the node in the bottom surface of the first boundary layer grid is corrected by an advancing height correction algorithm to obtain a corrected first advancing height. The expression of the advancing height correction algorithm is: wherein, i,j is the advancing height of the node j in the bottom surface of the i-th layer of boundary layer grid, i,j is the corrected advancing height of the node j in the bottom surface of the i-th layer of boundary layer grid, is the maximum principal curvature of the node j in the bottom surface of the i-th layer of boundary layer grid, is the maximum value of the maximum principal curvatures at all nodes in the bottom surface of the i-th boundary layer grid. wherein, is the advancing direction of node j in the bottom surface of the i-th boundary layer grid, is the corrected advancing direction of node j in the bottom surface of the i-th boundary layer grid, and a is a correction coefficient, i,j is the average of the maximum principal curvatures and the minimum principal curvatures of node j in the bottom surface of the i-th boundary layer grid is the principal direction tangent vector corresponding to the maximum principal curvatures of node j in the bottom surface of the i-th boundary layer grid. After the correction of the first advancing height of the first boundary layer grid and the correction of the advancing direction of each node in the bottom surface of the first boundary layer grid are completed, in the step S2, the advancing generation of the first boundary layer grid is performed in another embodiment, that is, each node in the bottom surface of the first boundary layer grid is respectively advanced by the corrected first advancing height along the respective corrected advancing direction to generate the first boundary layer grid composed of three prisms.

[0067] In combination with the above embodiments, in an embodiment, the present application further provides a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, the step S3 can include the steps S31 to S34:

[0068] The step S31: discretizing the side surface of the three prism in the first boundary layer grid into a triangle to obtain a face sheet set corresponding to the three prism, wherein the face sheet set includes a preset number of triangular face sheets.

[0069] In this embodiment, in order to reduce the computing resources required for collision detection, the present application will perform recursive segmentation for each triangular prism, and further collision detection based on the recursive segmentation results. Among them, the recursive segmentation technology is a technology of dividing the space structure into smaller sub-space regions recursively, each time following the same rules to form a hierarchical structure, such as BSP tree (Binary Space Partitioning Tree), octree for three-dimensional space partitioning, quadtree for two-dimensional plane partitioning, etc. The specific recursive segmentation and collision detection are as follows: the composition of each triangular prism in the first boundary layer grid includes 2 triangular patches of top and bottom surfaces, and 3 quadrilaterals of 3 side surfaces. When performing collision detection between triangular prisms in the first boundary layer grid, the side quadrilaterals of each triangular prism are first discretized into 2 triangles, so that each triangular prism is composed of a patch set of 8 triangular patches, that is, a triangular prism is composed of 8 triangular patches. The patch set of each triangular prism is 8 triangular patches. The collision intersection of two triangular prisms can be equivalent to the collision intersection of the patch sets of the two triangular prisms.

[0070] Step S32: recursively segmenting the patch set by a hierarchical space segmentation algorithm to obtain a corresponding recursive segmentation hierarchical structure.

[0071] In this embodiment, after obtaining the respective patch set of each triangular prism by division in step S31, the patch set of each triangular prism is respectively processed by recursive segmentation by a hierarchical space segmentation algorithm to obtain a corresponding recursive segmentation hierarchical structure for each triangular prism.

[0072] Step S33: based on the recursive segmentation hierarchical structure of each patch set in the first boundary layer grid, performing collision detection calculation between triangular prisms to obtain a corresponding calculation result.

[0073] In this embodiment, after obtaining the recursive segmentation hierarchical structure of each patch set, the recursive segmentation hierarchical structure between the patch sets is subjected to collision detection calculation to obtain a corresponding calculation result, so as to determine whether collision occurs between the triangular prisms in the first boundary layer grid.

[0074] Step S34: determining the target triangular prisms that collide with each other in the first boundary layer grid according to the calculation result.

[0075] In this embodiment, based on the obtained calculation result, it is determined whether collision occurs between the triangular prisms in the first boundary layer grid, and the triangular prisms that collide with each other are determined as target triangular prisms, and the target triangular prisms will be further iteratively reduced and corrected in the advancing height.

[0076] In combination with the above embodiments, in an implementation, the embodiments of the present application further provide a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, in the case that the hierarchical spatial partitioning algorithm is an octree algorithm, step S32 can include steps S321 to S324:

[0077] Step S321: taking an axis-aligned bounding box completely containing the patch set as a root node of an octree.

[0078] In the present embodiment, the recursive partitioning processing mode of each triangular prism is the same, and an example of a triangular prism is described herein. In the case that the hierarchical spatial partitioning algorithm is an octree algorithm, first, an axis-aligned bounding box completely containing the patch set of the triangular prism is constructed, and the axis-aligned bounding box is taken as a root node of an octree. The axis-aligned bounding box is different for different shapes of the triangular prism, and the specific shape is a cuboid or a cube.

[0079] Step S322: recursively partitioning the axis-aligned bounding box by the octree algorithm, and determining whether the partitioning nodes obtained in the recursive partitioning process satisfy a preset partitioning rule.

[0080] In the present embodiment, after the root node corresponding to the patch set of the triangular prism is determined by step S321, the axis-aligned bounding box is continuously recursively partitioned by the octree algorithm, and after 8 new partitioning nodes are obtained in each recursive partitioning, it is determined whether each of the 8 new partitioning nodes satisfies a preset partitioning rule. The preset partitioning rule can be that the number of triangular patches in the partitioning node is lower than a set threshold, or the size of the partitioning node is less than or equal to a pre-set minimum threshold, wherein the set threshold can be set according to the actual application scenario and is not specifically limited herein, such as 1, 2, etc.; the minimum threshold can be set according to the actual application scenario and is not specifically limited herein.

[0081] Step S323: in the case that the obtained partitioning node satisfies the preset partitioning rule, stopping the current partitioning, and determining that the partitioning node satisfying the preset partitioning rule is a leaf partitioning node.

[0082] In the present embodiment, the partitioning node satisfying the preset partitioning rule obtained by partitioning will not be further partitioned, and at this time, the partitioning node satisfying the preset partitioning rule is a tailmost leaf partitioning node. For the partitioning node obtained by partitioning that does not satisfy the preset partitioning rule, further recursive partitioning will be performed on the partitioning node until the partitioning node obtained by partitioning satisfies the preset partitioning rule. The partitioning nodes located between the root node and the leaf partitioning node obtained in the recursive partitioning process are sub-partitioning nodes in the entire partitioning result. For example, Figure 4 Figure 4 ​The middle node 1 is a root node, and the first round of recursive partitioning is performed on the root node to obtain eight partition nodes 11 to 18. It should be understood that Figure 4 The actual root node and partition nodes are three-dimensional axis-aligned bounding boxes (i.e., cuboids or cubes). The partition nodes 12 to 17 all satisfy the preset partitioning rule, and thus no further recursive partitioning is performed on the partition nodes 12 to 17. At this time, the partition nodes 12 to 17 are all leaf partition nodes at the end of the corresponding branches. The partition node 11 and the partition node 18 do not satisfy the preset partitioning rule, and thus further recursive partitioning is performed on the partition node 11 and the partition node 18. At this time, the partition node 11 is further partitioned into eight new partition nodes 111 to 118, and the partition node 18 is further partitioned into eight new partition nodes 181 to 188. The partition nodes 111 to 118 and the partition nodes 181 to 188 all satisfy the preset partitioning rule, and thus no further recursive partitioning is performed on these partition nodes. At this time, the partition nodes 111 to 118 and the partition nodes 181 to 188 are all leaf partition nodes at the end of the corresponding branches. Since the partition node 11 and the partition node 18 do not satisfy the preset partitioning rule, they are sub-partition nodes in the entire partitioning result. It should be understood that Figure 4 The simple diagram is only drawn for the purpose of explaining the specific recursive partitioning process of the present application. In the actual recursive partitioning process, there are more rounds of recursive partitioning, and more sub-partition nodes and leaf partition nodes are involved.

[0083] Step S324: In the case where all partitioning in the recursive partitioning process is stopped, a recursive partitioning hierarchy corresponding to the patch set is obtained.

[0084] In the present embodiment, in the case where all recursive partitioning in the recursive partitioning process is stopped, a recursive partitioning hierarchy corresponding to the patch set of the triangular prism is obtained.

[0085] In the embodiment, after the recursive partition hierarchy corresponding to the face sheet set of each triangular prism is obtained by using octree partition, the collision detection of the triangular prisms is: detecting whether the root nodes of the recursive partition hierarchies of two triangular prisms exist overlapping regions in space. If not, it is directly determined that there is no collision. If yes, it is recursively entered into the next layer of sub-partition nodes of the root nodes to detect whether they intersect, until the leaf sub-partition nodes are reached. That is, it is determined whether the next layer of sub-partition nodes of one of the two root nodes will intersect with the next layer of sub-partition nodes of the other root node. There are 8 sub-partition nodes in the next layer of sub-partition nodes of the root node. If one of the next layer of sub-partition nodes of one root node will exist overlapping regions in space with one of the next layer of sub-partition nodes of the other root node, it is recursively entered into the next layer of sub-partition nodes of the two sub-partition nodes, until the leaf sub-partition nodes are reached. If intersection occurs at the leaf sub-partition nodes, the accurate triangle face sheet and triangle face sheet intersection detection is performed on each triangle face sheet in the intersecting leaf sub-partition nodes. If intersection is found, it is determined that the two triangular prisms corresponding to the two intersecting triangle face sheets exist collision.

[0086] In combination with the above embodiments, in an implementation, the embodiments of the present application further provide a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, step S5 can include steps S51 to S53:

[0087] Step S51: in the case where all the regenerated target triangular prisms satisfy the first condition, determining the relationship between the grid line normal angle of a node in the first boundary layer grid bottom surface and the orthogonality threshold value, wherein the grid line normal angle of the node is the included angle between the advancing direction of the node and the vertex normal at the corresponding node in the geometric surface.

[0088] In the embodiment, in order to ensure that the grid line generated by the node advancing is approximately orthogonal to the wall surface, thereby improving the generation effect of the boundary layer grid, the present application provides another implementation for step S5: in the case where all the regenerated target triangular prisms satisfy the first condition, determining the relationship between the grid line normal angle of each node in the first boundary layer grid bottom surface and the orthogonality threshold value. The orthogonality threshold value is preferably 15°, and it should be understood that this is only a preferred value, which can also be taken as other values. The grid line normal angle of a node is the included angle between the advancing direction of the node and the vertex normal at the corresponding node in the geometric surface, wherein the corresponding node of a node in the geometric surface refers to the source node of the node in the geometric surface, and the node is obtained by advancing the source node of the node in the geometric surface through one layer of boundary layer grid.

[0089] Step S52: in the case where the grid line normal angle of the node is greater than the orthogonality threshold value, adjusting the position vector of the node to reduce the grid line normal angle of the node.

[0090] In the embodiment, in the case that the grid line normal angle of a node in the bottom surface of the first boundary layer grid is less than or equal to the normality threshold value, the grid line generated by the advancing of the node is approximately normal to the wall surface, and no adjustment is made to the advancing direction of the node. In the case that the grid line normal angle of a node in the bottom surface of the first boundary layer grid is greater than the normality threshold value, the position vector of the node is adjusted to reduce the grid line normal angle of the node, so that the grid line generated by the advancing of the node is approximately normal to the wall surface through the adjustment control.

[0091] Step S53: updating the advancing direction of the node according to the final position vector of the node after the adjustment, and controlling the node to re-advance to generate a new triangular prism along the updated advancing direction to obtain the first boundary layer grid that passes the division.

[0092] In the embodiment, after the adjustment of the position vector of the node with the grid line normal angle greater than the normality threshold value through step S52, the advancing direction of the node is re-determined based on the final position vector of the node after the adjustment, and then the node is controlled to re-advance to generate a new triangular prism along the re-determined advancing direction. The implementation of re-determining the advancing direction of the node is the same as the implementation of determining the advancing direction of the node, which will not be described here. Through the same implementation, the position vector of each node that needs to be adjusted can be adjusted. Meanwhile, through the same implementation, the advancing direction of each node after the position vector adjustment can be re-determined respectively, and each node can re-advance to generate a new triangular prism based on the new position vector of the node. After the re-advancing generation of all related triangular prisms, the final first boundary layer grid that passes the division is obtained. The related triangular prisms refer to any triangular prisms related to the nodes whose position vectors are adjusted. For example, the triangular prism T1 is generated by the nodes t11, t12 and t13, and when the node t11 is the node whose position vector is adjusted, the triangular prism T1 is the related triangular prism, and then the new triangular prism T1 is re-generated based on the advancing direction of the node t11, the advancing direction of the node t12 and the re-determined advancing direction of the node t13.

[0093] In combination with the above embodiments, in an implementation, the embodiment of the application further provides a boundary layer grid generation simulation method. In the boundary layer grid generation simulation method, step S52 can include: in the case that the grid line normal angle of a node is greater than the normality threshold value, substituting the position vector of the node into a pre-constructed local constraint algorithm, and adjusting the position vector of the node through a gradient descent algorithm to minimize the local constraint of the node.

[0094] In the case that the adjustment of the position vector of the node satisfies a second condition, the current position vector is determined as the final position vector of the node; and the expression of the local constraint algorithm is: wherein λ Orthogonal is an orthogonality weight factor, λ Smooth is a smoothness weight factor, is the position vector of node j in the bottom surface of the i-th boundary layer grid, is the position vector of node j in the geometric curved surface, is the vertex normal of node j in the geometric curved surface, is the position vector of the adjacent node k of node j in the bottom surface of the i-th boundary layer grid, and N(j) represents the number of adjacent nodes; and the expression of the gradient descent algorithm is: wherein θ is a step size, is the position vector of node j in the bottom surface of the i-th boundary layer grid after each adjustment of the gradient descent.

[0095] In the embodiment, the expression of the local constraint algorithm defined in the application is: wherein λ Orthogonal is an orthogonality weight factor, λ Smooth is a smoothness weight factor, is the position vector of node j in the bottom surface of the i-th boundary layer grid, is the position vector of node j in the geometric curved surface, is the vertex normal of node j in the geometric curved surface, is the position vector of the adjacent node k of node j in the bottom surface of the i-th boundary layer grid, and N(j) represents the number of adjacent nodes. In the local constraint algorithm, the orthogonality constraint term mainly considers the orthogonality of the grid line and the wall surface, and the weight ratio is larger, and the smoothness constraint term mainly considers the difference between adjacent nodes.

[0096] In the embodiment, the expression of the gradient descent algorithm defined in the application is: wherein θ is a step size, is the position vector of node j in the bottom surface of the i-th boundary layer grid after each adjustment of the gradient descent.

[0097] In the embodiment, in the case that the grid line orthogonality angle of the node in the first boundary layer grid bottom surface is greater than the orthogonality threshold value, the position vector of the node is substituted into the pre-constructed local constraint algorithm, and the position vector of the node is adjusted by the gradient descent algorithm to minimize the local constraint of the node. In the case that the adjustment of the position vector of the node satisfies a second condition, the current adjusted position vector of the node is determined as the final position vector of the node. The second condition is that the adjustment amount reaches a preset threshold value, or the number of iterations of the adjustment reaches a preset number of steps; the preset threshold value can be pre-set according to an actual application scenario, and the preset number of steps can also be pre-set according to an actual application scenario, which are not specifically limited here.

[0098] The simulation method for generating a boundary layer grid provided in the present application is as shown in the accompanying drawings. Figure 5 As shown in the accompanying drawings, first, a geometric curved surface of a wall surface to be analyzed is input, and the geometric curved surface is composed of discrete triangular patches. Then, some key control parameters in the entire boundary layer grid generation process are defined, such as a grid growth rate, a total number of layers, a total thickness, and an orthogonality threshold value. Then, the advancing directions of the nodes are calculated based on the positions of the nodes in the bottom surface of the first boundary layer grid to be generated, and after the advancing directions of the nodes are obtained, the first advancing height of the first boundary layer grid to be generated is calculated according to the preset grid growth rate. At this time, the height is a unified height, and the nodes in the bottom surface of the first boundary layer grid are advanced at the height.

[0099] After the advancing directions and the advancing heights of the nodes in the bottom surface of the first boundary layer grid are obtained, the advancing directions and the advancing heights of the nodes are further corrected to a certain extent based on the curvature of the region where the nodes are located (the correction is the correction of the advancing height and the advancing direction by the advancing height correction algorithm and the advancing direction correction algorithm, respectively). Then, the nodes are advanced based on the corrected advancing directions and the advancing heights to generate the first boundary layer grid composed of a large number of triangular prisms. At this time, the triangular prisms generated are not the final generation result of the first boundary layer grid, and further collision detection of the triangular prisms in the first boundary layer grid is required. For the triangular prisms that will not collide with other triangular prisms, no further adjustment is made to the triangular prisms, and for the triangular prisms that will collide with other triangular prisms, the advancing height of the triangular prisms is reduced, and then a new triangular prism is generated with the reduced advancing height, until no collision occurs or until the preset number of times of correction is performed, and the reduction adjustment of the advancing height of the triangular prism is ended.

[0100] After each of the three prisms in the first boundary layer grid that will collide with other three prisms completes the reduction adjustment, the final generation result of the first boundary layer grid is obtained. At this time, it is further determined whether the grid line orthogonality angle of each node in the bottom surface of the first boundary layer grid is greater than the orthogonality threshold. For the node with the grid line orthogonality angle greater than the orthogonality threshold, the position vector of the node is corrected based on the gradient descent algorithm to solve the local constraint equation. If the adjustment of the position vector of the node satisfies the second condition (the second condition is the second condition in the above embodiment, which will not be described again here), it is determined that the current adjusted position vector of the node is the final position vector of the node, and then the three prisms related to the node are regenerated based on the final position vector of the node. In the regeneration process, the advancement direction of each related three prism is changed only by the influence of the final position vector of the node, and the advancement height of each related three prism does not change. If the adjustment of the position vector of the node does not satisfy the second condition, the position vector of the node is continuously corrected based on the gradient descent algorithm and the local constraint equation. If there is no grid line orthogonality angle greater than the orthogonality threshold in all nodes in the bottom surface of the first boundary layer grid, it is further determined whether the number of generated boundary layer grids reaches the total number of pre-set layers, or whether the thickness of all generated boundary layer grids reaches the total thickness of pre-set thickness. If the total number of pre-set layers is reached, or the total thickness of pre-set thickness is reached, it is determined that the division of the boundary layer grid is completed, at this time the complete boundary layer grid division result is transitionally connected with the unstructured far-field core grid, and the Delaunay triangulation hybrid grid technology is used to ensure the interface compatibility. If there is a node greater than the orthogonality threshold,

[0101] The boundary layer grid generation simulation method provided in the application uses octree space division and collision detection to mark the three prisms that collide after each layer of grid is advanced, dynamically adjusts the advancement height according to the local curvature and collision, eliminates the collision through iteration and re-advancement, and optimizes the grid quality through curvature correction, orthogonality correction and smoothness correction technology. Thus, the problem of grid overlap, illegal unit or distortion caused by curvature mutation or space occlusion when the traditional boundary layer advancement algorithm processes complex geometry (such as grooves and slits) can be effectively solved. The method can effectively reduce manual intervention and computational burden by adaptively adjusting the advancement height, and avoid the problem of waste of computational resources caused by manual intervention or global reduction after collision of the traditional boundary layer advancement algorithm. At the same time, the grid quality in the high curvature area can be effectively improved through curvature correction, orthogonality correction and smoothness correction, and the problem of grid distortion in the high curvature area caused by the fixed advancement step of the traditional algorithm that cannot adapt to the local curvature change can be avoided. The method can support complete boundary layer generation of complex geometry, improve the fluid simulation accuracy, and avoid the problem of decreased fluid simulation accuracy caused by stopping advancement after collision of the traditional algorithm.

[0102] Based on the same inventive concept, an embodiment of the present application provides a boundary layer grid generation simulation system, as shown in the figure, the boundary layer grid generation simulation system 600 comprises: Figure 6 A model input module 601 is configured to input a geometric surface of a wall surface to be analyzed, wherein the geometric surface is composed of discrete triangular patches.

[0103] A propulsion control module 602 is configured to control each node in a first boundary layer grid base surface to be respectively pushed outwards along a respective propulsion direction by a first propulsion height, so as to generate a first boundary layer grid composed of triangular prisms, wherein the first boundary layer grid is a current boundary layer grid to be generated.

[0104] A target triangular prism determination module 603 is configured to determine target triangular prisms that collide with each other in the first boundary layer grid.

[0105] A propulsion iteration reduction module 604 is configured to iteratively reduce the propulsion height of the target triangular prisms, and after each reduction, generate new target triangular prisms by re-propelling with the reduced propulsion height, until the re-propelled target triangular prisms meet a first condition.

[0106] A boundary layer generation module 605 is configured to obtain a qualified first boundary layer grid when all the re-generated target triangular prisms meet the first condition.

[0107] A division result determination module 606 is configured to determine a boundary layer grid division result corresponding to the geometric surface when the total number of qualified boundary layer grids reaches a preset number of layers.

[0108] A simulation analysis module 607 is configured to perform simulation analysis on the geometric surface based on the boundary layer grid division result, and obtain a corresponding simulation analysis result.

[0109] Optionally, the boundary layer grid generation simulation system 600 further comprises a first propulsion height determination module configured to determine a first propulsion height of the first boundary layer grid, and the first propulsion height determination module comprises:

[0110] A layer sequence determination module is configured to determine a layer sequence of the current first boundary layer grid to be generated.

[0111] A propulsion height first determination module is configured to determine a predefined first layer propulsion height as the first propulsion height of the first boundary layer grid when the layer sequence is a first layer.

[0112] A propulsion height second determination module is configured to determine a predefined second layer propulsion height as the first propulsion height of the first boundary layer grid when the layer sequence is a second layer.

[0113] The second determination module is configured to determine the first advancing height of the first boundary layer grid according to the advancing height of the first layer and a preset grid growth rate when the layer sequence is not the first layer.

[0114] Optionally, the boundary layer grid generation simulation system 600 further comprises a determination module for determining the advancing direction of the node.

[0115] The target triangular patch determination module is configured to determine each target triangular patch with the node as a common vertex.

[0116] The area and normal vector determination module is configured to determine the area and normal vector of the target triangular patch according to the node position of the target triangular patch.

[0117] The vertex normal determination module is configured to obtain the vertex normal of the node by calculating through a first algorithm according to the area and normal vector of each target triangular patch.

[0118] The first algorithm is expressed as: wherein, is the vertex normal at node j in the bottom surface of the i-th boundary layer grid, A i,T is the area of the triangular patch T with node j in the bottom surface of the i-th boundary layer grid as a common vertex, is the normal vector of the triangular patch T with node j in the bottom surface of the i-th boundary layer grid as a common vertex, the triangular patch T belongs to the triangular patches with node j in the bottom surface of the i-th boundary layer grid as a common vertex.

[0119] The advancing direction determination submodule is configured to determine the vertex normal of the node as the advancing direction of the node.

[0120] Optionally, the boundary layer grid generation simulation system 600 further comprises:

[0121] The first correction module is configured to correct the first advancing height of the node in the bottom surface of the first boundary layer grid through a advancing height correction algorithm to obtain a corrected first advancing height.

[0122] The advancing height correction algorithm is expressed as: wherein, y i,j is the advancing height of node j in the bottom surface of the i-th boundary layer grid, y′ i,j is the corrected advancing height of node j in the bottom surface of the i-th boundary layer grid, is the maximum principal curvature of node j in the bottom surface of the i-th boundary layer grid, is the maximum value of the maximum principal curvatures of all nodes in the bottom surface of the i-th boundary layer grid.

[0123] a second correction module, configured to correct the advancing direction of each node in the first boundary layer grid bottom surface by using an advancing direction correction algorithm, and obtain a corrected advancing direction;

[0124] The expression of the advancing direction correction algorithm is as follows: wherein, is the advancing direction of node j in the i-th boundary layer grid bottom surface, is the corrected advancing direction of node j in the i-th boundary layer grid bottom surface, a is a correction coefficient, H i,j is the average of the maximum principal curvature and the minimum principal curvature of node j in the i-th boundary layer grid bottom surface is the principal direction tangent vector corresponding to the maximum principal curvature of node j in the i-th boundary layer grid bottom surface;

[0125] The advancing control module 602 is further configured to control each node in the first boundary layer grid bottom surface to advance outward by a respective corrected first advancing height along a respective corrected advancing direction, and generate a first boundary layer grid composed of three-prism.

[0126] Optionally, the target three-prism determination module 603 comprises:

[0127] The discrete processing module is configured to discretize the side surface of the three-prism in the first boundary layer grid into a triangle, and obtain a face sheet set corresponding to the three-prism, wherein the face sheet set comprises a preset number of triangular face sheets.

[0128] The recursive segmentation module is configured to perform recursive segmentation on the face sheet set by using a hierarchical spatial segmentation algorithm, and obtain a corresponding recursive segmentation hierarchical structure.

[0129] The collision detection module is configured to perform collision detection calculation between three-prisms based on the recursive segmentation hierarchical structure of each face sheet set in the first boundary layer grid, and obtain a corresponding calculation result.

[0130] The target three-prism determination sub-module is configured to determine target three-prisms that collide with each other in the first boundary layer grid according to the calculation result.

[0131] Optionally, the recursive segmentation module comprises:

[0132] The root node determination module is configured to, in the case that the hierarchical spatial segmentation algorithm is an octree algorithm, take an axis-aligned bounding box that completely contains the face sheet set as a root node of the octree.

[0133] The first recursive partitioning module is configured to perform recursive partitioning on the axis-aligned bounding box by using an octree algorithm, and determine whether a partitioning node obtained in the recursive partitioning process satisfies a preset partitioning rule;

[0134] The second recursive partitioning module is configured to, in a case where the obtained partitioning node satisfies the preset partitioning rule, stop the current partitioning, and determine the partitioning node satisfying the preset partitioning rule as a leaf partitioning node.

[0135] The third recursive partitioning module is configured to, in a case where all the partitioning in the recursive partitioning process is stopped, obtain a recursive partitioning hierarchical structure corresponding to the patch set.

[0136] Optionally, the boundary layer generation module 605 comprises:

[0137] The comparison module is configured to, in a case where all the regenerated target tri-prisms satisfy the first condition, determine a relationship between a grid line normal angle of a node in the first boundary layer grid bottom surface and the orthogonality threshold, wherein the grid line normal angle of the node is an included angle between a pushing direction of the node and a vertex normal at a corresponding node in the geometric curved surface.

[0138] The adjustment module is configured to, in a case where the grid line normal angle of the node is greater than the orthogonality threshold, adjust a position vector of the node to reduce the grid line normal angle of the node.

[0139] The pushing direction updating and boundary layer generation module is configured to update the pushing direction of the node by using the final position vector of the node after the adjustment, and control the node to be re-pushed along the updated pushing direction to generate a new tri-prism, so as to obtain a first boundary layer grid that passes the division.

[0140] Optionally, the adjustment module comprises:

[0141] The constraint calculation module is configured to, in a case where the grid line normal angle of the node is greater than the orthogonality threshold, substitute the position vector of the node into a pre-constructed local constraint algorithm, and adjust the position vector of the node by using a gradient descent algorithm to minimize the local constraint of the node.

[0142] The adjustment submodule is configured to, in a case where the adjustment of the position vector of the node satisfies a second condition, determine the current position vector as the final position vector of the node.

[0143] An expression of the local constraint algorithm is as follows:

[0144] wherein λ Orthogonal is an orthogonality weight factor, λ Smooth is a smoothness weight factor, is a position vector of a node j in a bottom surface of an i-th boundary layer grid, is a position vector of a node j in a geometric curved surface, is a vertex normal of a node j in a geometric curved surface, is a position vector of a neighboring node k of the node j in the bottom surface of the i-th boundary layer grid, and N(j) represents a number of the neighboring nodes;

[0145] An expression of the gradient descent algorithm is as follows: wherein, θ is a step length, is a position vector of the node j in the bottom surface of the i-th boundary layer grid after each gradient descent adjustment.

[0146] Optionally, the boundary layer grid generation simulation system 600 further comprises a first boundary layer height determination module configured to, in a case where the first boundary layer grid is a first boundary layer grid, determine a first boundary layer grid advancing height of the first boundary layer grid that meets a wall function requirement of a turbulence model during simulation by using a preset algorithm, and an expression of the preset algorithm is as follows: wherein, y + represents a dimensionless wall distance, μ represents a fluid dynamic viscosity, ρ represents a fluid density, u τ represents a friction velocity, τ w represents a wall shear stress, and y1 represents the first boundary layer grid advancing height.

[0147] Based on the same inventive concept, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and running on the processor, and the computer program is executed by the processor to implement the steps in the boundary layer grid generation simulation method according to the first aspect of the present application.

[0148] Based on the same inventive concept, an embodiment of the present application provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in the boundary layer grid generation simulation method according to the first aspect of the present application.

[0149] For the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the related parts refer to the part of the method embodiment.

[0150] It should be noted that, for the method embodiments, the operations performed are described in sequence for simplicity and clarity, but those skilled in the art should understand that, according to the method embodiments of the present application, certain steps can be performed in other sequences or at the same time, and the described sequence is not used to limit the application. In addition, those skilled in the art should understand that the embodiments described in the specification are all preferred embodiments, and the operations involved are not necessarily essential to the application.

[0151] Each of the embodiments in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same and similar parts between the embodiments can be referred to each other.

[0152] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the embodiments of the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0153] The embodiments of the present application are described with reference to flowcharts and / or block diagrams according to the method, terminal device (system), and computer program product of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the computer or other programmable data processing terminal device produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one block or multiple blocks.

[0154] These computer program instructions can also be stored in a computer readable storage medium that can guide the computer or other programmable data processing terminal device to work in a specific way, so that the instructions stored in the computer readable storage medium produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one block or multiple blocks.

[0155] These computer program instructions can also be loaded into a computer or other programmable data processing terminal device, so that a series of operational steps are performed on the computer or other programmable terminal device to generate a computer-implemented process, thus the instructions executed on the computer or other programmable terminal device provide a process for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or the functions specified in the block Figure 1 one block or multiple blocks.

[0156] Although the preferred embodiments of the present application have been described, those skilled in the art who understand the basic inventive concept after getting to know the present application can make additional changes and modifications to the embodiments. Therefore, the appended claims are intended to cover the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0157] Finally, it should also be noted that, in this document, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or terminal device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or terminal device including the element.

[0158] The above describes in detail the boundary layer grid generation simulation method, system, device and medium provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above embodiment descriptions are only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range will be changed; and the above description of the present application should not be understood as a limitation.

Claims

1. A boundary layer mesh generation simulation method, characterized in that, The method includes: Input the geometric surface of the wall to be analyzed, which is composed of discrete triangular patches; Each node in the bottom surface of the first boundary layer mesh is controlled to advance outward by the first advancement height along its respective advancement direction, thereby generating a first boundary layer mesh composed of triangular prisms, wherein the first boundary layer mesh is the current boundary layer mesh to be generated; Identify the target triangular prisms that collide with each other in the first boundary layer mesh; The advancement height of the target triangular prism is iteratively reduced, and after each reduction, a new target triangular prism is generated by advancing it again with the reduced advancement height, until the target triangular prism generated by advancing it again satisfies the first condition. If all the regenerated target triangular prisms satisfy the first condition, a qualified first boundary layer mesh is obtained; If the total number of qualified boundary layer meshes reaches the preset number of layers, the boundary layer meshing result corresponding to the geometric surface is determined. Based on the boundary layer mesh division results, the geometric surface is simulated and analyzed to obtain the corresponding simulation analysis results.

2. The boundary layer mesh generation simulation method according to claim 1, characterized in that, Determining the first advance height of the first boundary layer mesh includes: Determine the layer sequence of the first boundary layer mesh to be generated; When the layer sequence is the first layer, the predefined first layer advance height is determined as the first advance height of the first boundary layer grid; When the layer sequence is not the first layer, the first advancement height of the first boundary layer grid is determined based on the advancement height of the first layer and the preset grid growth rate.

3. The boundary layer mesh generation simulation method according to claim 2, characterized in that, Determine the direction of node advancement, including: Determine each target triangle facet with the node as a common vertex; Based on the node positions of the target triangular facet, determine the area and normal vector of the target triangular facet; Based on the area and normal vector of each target triangle facet, the vertex normal of the node is calculated using the first algorithm. The first algorithm expression is: in, Let A be the vertex normal at node j on the bottom surface of the i-th boundary layer mesh. i,T Let T be the area of ​​the triangular facet T that shares a common vertex with node j in the bottom face of the i-th boundary layer mesh. Let T be the normal vector of the triangular face T that has node j as a common vertex in the bottom surface of the i-th boundary layer mesh. Triangular face T belongs to the triangular face that has node j as a common vertex in the bottom surface of the i-th boundary layer mesh. The vertex normal of the node is determined as the direction of advancement of the node.

4. The boundary layer mesh generation simulation method according to claim 3, characterized in that, The method further includes: The first advance height of the nodes in the bottom surface of the first boundary layer mesh is corrected by the advance height correction algorithm to obtain the corrected first advance height; The expression for the propulsion height correction algorithm is: Among them, y i,j Let y' be the advance height at node j on the bottom surface of the i-th boundary layer mesh. i,j The corrected thrust height is the height at node j on the bottom surface of the i-th boundary layer mesh. Let j be the maximum principal curvature at node j on the bottom surface of the i-th boundary layer mesh. The maximum value of the principal curvature at all nodes on the bottom surface of the i-th boundary layer mesh; The advancing direction of the nodes in the bottom surface of the first boundary layer mesh is corrected by the advancing direction correction algorithm to obtain the corrected advancing direction; The expression for the propulsion direction correction algorithm is: in, The direction of advancement is at node j on the bottom surface of the i-th boundary layer mesh. H represents the corrected propagation direction at node j on the bottom surface of the i-th boundary layer mesh, where α is the correction coefficient. i,j The mean of the maximum and minimum principal curvatures at node j on the bottom surface of the i-th boundary layer mesh. is the principal tangent vector of the principal direction of the maximum principal curvature at node j on the bottom surface of the i-th boundary layer mesh; The control method involves each node in the bottom surface of the first boundary layer mesh advancing outwards by the first advancement height along its respective advancement direction, generating a first boundary layer mesh composed of triangular prisms, including: Each node in the bottom surface of the first boundary layer mesh is controlled to advance outward along its own modified advance direction to its own modified first advance height, thereby generating a first boundary layer mesh composed of triangular prisms.

5. The boundary layer mesh generation simulation method according to claim 1, characterized in that, Determining the target triangular prisms colliding with each other in the first boundary layer mesh includes: Discretize the lateral face of the triangular prism in the first boundary layer mesh into triangles to obtain a set of facets corresponding to the triangular prism, wherein the set of facets includes a preset number of triangular facets; The set of facets is recursively segmented using a hierarchical spatial segmentation algorithm to obtain the corresponding recursive segmentation hierarchy structure. Based on the recursive segmentation hierarchy of each face set in the first boundary layer mesh, collision detection calculations are performed between triangular prisms to obtain the corresponding calculation results; Based on the calculation results, the target triangular prisms that collide with each other in the first boundary layer mesh are determined.

6. The boundary layer mesh generation simulation method according to claim 5, characterized in that, When the hierarchical spatial partitioning algorithm is an octree algorithm, the step of recursively partitioning the face set using the hierarchical spatial partitioning algorithm to obtain the corresponding recursive partitioning hierarchy structure includes: The root node of the octree is the axis-aligned bounding box that completely contains the set of patches. The axis-aligned bounding box is recursively segmented using an octree algorithm, and it is determined whether the segmentation nodes obtained during the recursive segmentation process meet the preset segmentation rules. If the obtained segmentation node satisfies the preset segmentation rule, stop the current segmentation and determine the segmentation node that satisfies the preset segmentation rule as a leaf segmentation node; If all segmentations stop during the recursive segmentation process, a recursive segmentation hierarchy structure corresponding to the set of patches is obtained.

7. The boundary layer mesh generation simulation method according to claim 1, characterized in that, If all regenerated target triangular prisms satisfy the first condition, a qualified first boundary layer mesh is obtained, including: If all regenerated target triangular prisms satisfy the first condition, determine the relationship between the orthogonality angle of the grid lines of the nodes in the bottom surface of the first boundary layer grid and the orthogonality threshold, wherein the orthogonality angle of the grid lines of the node is the angle between the advancing direction of the node and the normal of the vertex at the corresponding node in the geometric surface; If the orthogonality angle of the grid lines of a node is greater than the orthogonality threshold, the position vector of the node is adjusted to reduce the orthogonality angle of the grid lines of the node. The node's final position vector after adjustment is used to update the node's advancing direction, and the node is controlled to advance again along the updated advancing direction to generate a new triangular prism, so as to obtain a qualified first boundary layer mesh.

8. The boundary layer mesh generation simulation method according to claim 7, characterized in that, If the orthogonality angle of a node's grid lines is greater than an orthogonality threshold, the position vector of the node is adjusted to reduce the orthogonality angle of the node's grid lines, including: When the orthogonality angle of the grid lines of a node is greater than the orthogonality threshold, the position vector of the node is substituted into a pre-built local constraint algorithm, and the position vector of the node is adjusted by the gradient descent algorithm to minimize the local constraints of the node. If the adjustment of the node's position vector satisfies the second condition, the current position vector is determined to be the node's final position vector; The expression for the local constraint algorithm is: Where, λ Orthogonal λ is the orthogonality weighting factor. Smooth As a smoothness weighting factor, Let be the position vector of node j in the bottom surface of the i-th boundary layer mesh. Let j be the position vector of node j in the geometric surface. Let be the vertex normal of node j in the geometric surface. N(j) is the position vector of the neighboring node k of node j in the bottom surface of the i-th boundary layer mesh, and N(j) represents the number of neighboring nodes. The expression for the gradient descent algorithm is: Where θ is the step size. Let be the position vector of node j on the bottom surface of the i-th boundary layer mesh after each gradient descent adjustment.

9. The boundary layer mesh generation simulation method according to claim 1, characterized in that, When the first boundary layer mesh is a first-layer boundary layer mesh, the advancement height of the first-layer boundary layer mesh that satisfies the wall function requirements of the turbulence model during simulation is determined by a preset algorithm. The preset algorithm expression is: Among them, y + Represents the dimensionless wall distance, μ represents the hydrodynamic viscosity, ρ represents the fluid density, and u τ τ represents the frictional velocity. w y1 represents the wall shear stress, and y1 represents the advance height of the first boundary layer mesh.

10. A boundary layer mesh generation simulation system, characterized in that, The system includes: The model input module is used to input the geometric surface of the wall to be analyzed, which is composed of discrete triangular patches; The propulsion control module is used to control each node in the bottom surface of the first boundary layer mesh to propel outward along its respective propulsion direction to the first propulsion height, thereby generating a first boundary layer mesh composed of triangular prisms, wherein the first boundary layer mesh is the current layer of boundary layer mesh to be generated; The target triangular prism determination module is used to determine the target triangular prisms that collide with each other in the first boundary layer mesh; The iterative reduction module is used to iteratively reduce the advancement height of the target triangular prism, and after each reduction, it advances again with the reduced advancement height to generate a new target triangular prism until the re-generated target triangular prism satisfies the first condition. The boundary layer generation module is used to obtain a qualified first boundary layer mesh when all regenerated target triangular prisms meet the first condition. The partitioning result determination module is used to determine the boundary layer mesh partitioning result corresponding to the geometric surface when the total number of qualified boundary layer meshes reaches a preset number of layers. The simulation analysis module is used to perform simulation analysis on the geometric surface based on the boundary layer mesh division results, and obtain the corresponding simulation analysis results.

11. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and running on the processor, wherein the computer program, when executed by the processor, implements the steps of a boundary layer mesh generation simulation method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a boundary layer mesh generation simulation method as described in any one of claims 1 to 9.

Images