Full-quadrilateral mesh generation method based on predefined mode

Abstract

The invention discloses a full-quadrilateral grid generation method based on a predefined mode, and particularly relates to the technical field of grid generation, and the method comprises the specific steps: cleaning a geometric model, setting a predefined mode of grid generation according to the cleaned geometric model, and defining a grid generation rule, the method comprises the following steps: carrying out grid division according to a geometric model and a predefined mode, generating a full-quadrilateral grid based on an overlay network technology, carrying out cooperative scheduling on generation of the full-quadrilateral grid by utilizing a mode based on a distributed architecture platform, optimizing the full-quadrilateral grid by utilizing a post-processing technology, and carrying out quality inspection on the generated full-quadrilateral grid. According to the method, the generation efficiency of the full-quadrilateral grid is evaluated by combining the result of quality inspection of the full-quadrilateral grid and the calculation precision in the grid generation process, and a grid quality control strategy is provided according to the evaluation result, so that the contradiction between data consistency and processing speed when a large-scale complex geometric model is processed is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of mesh generation, and more specifically, to a method for generating a full quadrilateral mesh based on a predefined pattern. Background Art

[0002] A full quadrilateral mesh is a mesh composed of quadrilateral units and is used for regional discretization in numerical simulations. Compared with triangular meshes, full quadrilateral meshes can provide better accuracy and stability, but it is difficult to generate high-quality full quadrilateral meshes in complex geometries. Traditional mesh generation methods may cause the shapes of quadrilateral units to be severely distorted or irregular when dealing with complex geometries. Full quadrilateral mesh generation must efficiently handle large-scale computing and storage requirements while ensuring mesh quality. This problem is particularly prominent when dealing with large-scale complex geometric models. If parallel computing is used to improve computing efficiency, the data consistency of predefined patterns and mesh generation rules will become a new obstacle.

[0003] In order to solve the above defects, a technical solution is now proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a method for generating a full quadrilateral mesh based on a predefined pattern to solve the problems in the above-mentioned background technology.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: a method for generating a full quadrilateral mesh based on a predefined pattern, the specific steps of which include cleaning a geometric model, setting a predefined pattern for mesh generation according to the cleaned geometric model, and defining a mesh generation rule;

[0006] Meshing is performed according to the geometric model and predefined patterns, and full quadrilateral meshes are generated based on overlapping network technology. The generation of full quadrilateral meshes is coordinated and scheduled based on a distributed architecture platform.

[0007] Use post-processing technology to optimize the full quadrilateral mesh and perform quality checks on the generated full quadrilateral mesh;

[0008] The generation efficiency of the full quadrilateral mesh is evaluated by combining the results of the full quadrilateral mesh quality check and the calculation accuracy in the mesh generation process, and a mesh quality control strategy is proposed based on the evaluation results.

[0009] Preferably, the steps of cleaning the geometric model are:

[0010] Delete small features, fill small holes in geometric models, delete or merge small edges and faces in geometric models;

[0011] Simplify the model, merge multiple small coplanar faces in the geometric model into one large face, and simplify curves and surfaces;

[0012] Repair geometric defects, fill gaps and gaps in geometric models, and eliminate overlapping and intersecting parts in geometric models;

[0013] Optimize the geometric model, smooth the surface, eliminate wrinkles and irregular shapes, and make the surface normal vector direction consistent.

[0014] Preferably, the method for setting the predefined mode of grid generation is:

[0015] Set the basic mesh density for the cleaned geometric model and set the refined mesh density for the refined area in the geometric model;

[0016] Determine the orientation of the mesh based on the shape of the geometric model and the directionality of the physical application;

[0017] Select the meshing strategy based on the shape of the geometric model;

[0018] The rules and constraints for grid generation are set according to grid application requirements, including the maximum and minimum aspect ratios of grid cells, the internal angle range of grid cells, and smooth transition constraints between grid cells.

[0019] Preferably, the process of generating a full quadrilateral mesh using overlapping network technology is:

[0020] Decompose the geometric model into several subdomains, each of which can be meshed independently and the overlapping parts of the subdomains can cover the entire geometric model;

[0021] Each subdomain generates a fully quadrilateral mesh individually according to the predefined pattern and mesh generation rules;

[0022] The overlapping part between each subdomain is defined as the overlapping area, and the interpolation boundary within the overlapping area is set;

[0023] The grids of all subdomains are combined to form a complete overlapping network, and the interpolation algorithm is used to perform data transition in the overlapping area.

[0024] Preferably, the method of using the interpolation algorithm to perform data transition is:

[0025] Calibrate subdomain A and subdomain B, the overlapping area of ​​subdomain A and subdomain B is A∩B, and all grid nodes within the overlapping area A∩B are used as interpolation points;

[0026] Assign an interpolation weight to each interpolation point. The interpolation point is between the grid node (j, k) of A and the grid node (m, n) of B. The interpolation weight of the grid node (j, k) and the grid node (m, n) is W. jk and Wmn ;

[0027] According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point has Q = W jk *Q A(j,k) +W mn *Q B(m,n) , where Q A(j,k) is the calculated value of the grid node (j, k) of A, Q B(m,n) is the calculated value of the grid node (m,n) of B.

[0028] Preferably, the logic of collaboratively scheduling the generation of the full quadrilateral mesh through the distributed architecture is:

[0029] In the calibration distributed platform, the server processes the geometric model cleaning, local refinement, mesh generation, and quality inspection subtasks in order of a, b, c, and d, respectively. The number of times each subtask is processed is N. a 、N b 、N c 、N d , set Sub = {N a ,N b ,N c ,N d}, the total number of subtask processing is N u , the servers of the distributed architecture platform are numbered as e, the total number of servers is G, and e = {1, 2, 3…G}, where G is a positive integer, the average maximum aspect ratio of the grid units generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid units generated by each server in the same predefined mode is Ar2, and the average half-angle value of the grid units generated by each server in the same predefined mode is Hs, then the scheduling difference coefficient is In the formula, α and β are and The proportional coefficient, and α and β are both positive numbers. The maximum number of times the server processes subtasks in the distributed architecture platform is max(Sub), and the minimum number is min(Sub);

[0030] The scheduling threshold St is preset. When the calculated scheduling difference coefficient Sd is greater than or equal to the scheduling threshold St, the data consistency of distributed processing is stable, the efficiency of grid generation is stable, and collaborative management is not performed;

[0031] When the obtained scheduling difference coefficient Sd is less than the scheduling threshold St, the data consistency of distributed processing fluctuates, and the load generated by the grid is unbalanced and inefficient, prompting technicians to perform cross-server collaborative management.

[0032] Preferably, the method for post-processing the full quadrilateral mesh is:

[0033] In-Situ post-processing technology is used to perform real-time analysis on the intermediate data of the mesh generation process, check the mesh quality indicators, check the intermediate results of the mesh generation process according to predefined quality standards, identify the mesh units that do not meet the quality standards, record the location and type of the problem, feedback the quality inspection results, and dynamically adjust and optimize the generation process.

[0034] Preferably, the generation efficiency of the full quadrilateral mesh is evaluated, and the logic of the mesh quality control strategy is proposed based on the evaluation results:

[0035] The non-standard mesh unit identified in the calibration quality check is Gt, t is the number of the mesh unit, the number of times the mesh generation process is adjusted and optimized by post-processing technology is Op, the internal angle difference between adjacent quadrilateral units is ΔΘ, the distortion of adjacent units is ΔS, the gradient of the size of the mesh unit is Gl, the computational resource consumption of mesh generation is Cp, the mesh generation time is Tg, the consistency efficiency coefficient is Ec = |St-Sd|, and the generation efficiency of the full quadrilateral mesh is evaluated. The calculation expression is Where, Ev is the evaluation coefficient;

[0036] The evaluation threshold Et is preset. When the calculated evaluation coefficient Ev is greater than or equal to the evaluation threshold Et, the quality of the generated full quadrilateral mesh is low. The quality control strategy of the informed mesh is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization rate of computing resources.

[0037] When the calculated evaluation coefficient Ev is less than the evaluation threshold Et, the grid quality control strategy is not notified.

[0038] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0039] This application combines a distributed interactive platform with post-processing technology verification to verify and optimize the generation of full quadrilateral meshes based on predefined patterns, improve the utilization of computing resources, coordinate the generation of meshes, speed up the efficiency of mesh generation while ensuring data consistency, and use overlapping network technology to improve the accuracy and speed of mesh generation, thus resolving the contradiction between data consistency and processing speed when processing large-scale complex geometric models. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1: Please refer to Figure 1 As shown, the present invention is a method for generating a full quadrilateral mesh based on a predefined pattern, and the specific steps include cleaning a geometric model, setting a predefined pattern for mesh generation according to the cleaned geometric model, and defining a mesh generation rule;

[0044] Meshing is performed according to the geometric model and predefined patterns, and full quadrilateral meshes are generated based on overlapping network technology. The generation of full quadrilateral meshes is coordinated and scheduled based on a distributed architecture platform.

[0045] Use post-processing technology to optimize the full quadrilateral mesh and perform quality checks on the generated full quadrilateral mesh;

[0046] The generation efficiency of the full quadrilateral mesh is evaluated by combining the results of the full quadrilateral mesh quality check and the calculation accuracy in the mesh generation process, and a mesh quality control strategy is proposed based on the evaluation results.

[0047] A full quadrilateral mesh is a mesh composed of quadrilateral elements, which is used for regional discretization in numerical simulations. Compared with triangular meshes, full quadrilateral meshes can provide better accuracy and stability in finite element analysis and fluid dynamics calculations. In typical applications of numerical calculations, whether it is deformation or stress analysis, full quadrilateral meshes can provide more accurate and stable computing performance than triangular meshes. At the same time, in the same geometric area, quadrilateral meshes can use fewer elements to achieve the same resolution, thereby reducing computing costs. The uniformity and shape of quadrilateral elements are more regular than those of triangular elements, and they are more suitable for high-order numerical methods.

[0048] However, there are still some difficulties in generating full quadrilateral meshes in complex geometric structures and high-precision numerical simulations. It is difficult to generate high-quality full quadrilateral meshes for complex geometric shapes. Traditional mesh generation methods may cause serious distortion or irregularity of the shape of quadrilateral elements when dealing with complex geometries.

[0049] High-quality meshes require that the shape of the quadrilateral units is close to a square and has a reasonable aspect ratio. Shape optimization is required during mesh generation to avoid distorted and deformed units. The mesh needs to be smooth and uniform to avoid sudden changes in unit size and shape. Smoothness and uniformity are crucial to the stability and accuracy of numerical simulations. In fluid dynamics calculations, the boundary layer region requires detailed meshing to capture the viscous effects of the fluid. Generating high-quality all-quadrilateral meshes to meet the capture requirements is very challenging.

[0050] Complex geometries usually contain multiple topological features. Automatically generating high-quality all-quad meshes requires processing complex topological relationships. It is an important challenge to develop an automated mesh generation algorithm based on predefined patterns that can adaptively generate high-quality all-quad meshes under different geometric and physical conditions.

[0051] The full-quadric mesh generation algorithm needs to efficiently handle large-scale computing and storage requirements while ensuring mesh quality. In order to improve generation speed and processing power, the mesh generation algorithm needs to support parallel processing. While making full use of multi-core processors and distributed computing resources, it is also necessary to ensure data consistency of predefined patterns and mesh generation rules.

[0052] Complex geometric shapes include curved surfaces, corners, narrow channels, and topological features such as holes, intersections, and gaps.

[0053] Clean up the geometric model, set the predefined mode of mesh generation according to the cleaned geometric model, and define the mesh generation rules;

[0054] The steps to clean up the geometry model are:

[0055] Delete small features, fill small holes in geometric models, delete or merge small edges and faces in geometric models;

[0056] Simplify the model, merge multiple small coplanar faces in the geometric model into one large face, and simplify curves and surfaces;

[0057] Repair geometric defects, fill gaps and gaps in geometric models, and eliminate overlapping and intersecting parts in geometric models;

[0058] Optimize the geometric model, smooth the surface, eliminate wrinkles and irregular shapes, and make the surface normal vector direction consistent.

[0059] The predefined modes for mesh generation are set as follows:

[0060] Set the basic mesh density for the cleaned geometric model and set the refined mesh density for the refined area in the geometric model;

[0061] Determine the orientation of the mesh based on the shape of the geometric model and the directionality of the physical application;

[0062] Select the meshing strategy based on the shape of the geometric model;

[0063] The rules and constraints for grid generation are set according to grid application requirements, including the maximum and minimum aspect ratios of grid cells, the internal angle range of grid cells, and smooth transition constraints between grid cells.

[0064] It should be noted that the refined area includes the sharp corners, boundary layers, and geometric transition zones of the geometric model. Commonly used partitioning strategies include mapping method, sweeping method, and regional segmentation method. For fluid mechanics problems, the direction of the grid is consistent with the flow direction. For structural mechanics problems, the direction of the grid is consistent with the main stress direction.

[0065] It should be pointed out that the setting of the predefined grid mode is based on the analysis requirements and shape characteristics of the grid, and is constrained and set by professional and technical personnel in this field.

[0066] Meshing is performed according to the geometric model and predefined patterns, and full quadrilateral meshes are generated based on overlapping network technology;

[0067] Overlapping mesh technology is a mesh generation method that allows the use of multiple independent mesh regions in computational fluid dynamics simulations. These regions can be partially or completely overlapped. Each mesh region can be meshed and optimized independently to accommodate complex geometries or flow characteristics. Overlapping mesh technology manages data exchange and boundary conditions between different meshes by defining interpolation regions and hole regions, thereby achieving effective simulation of the entire computational domain.

[0068] The process of generating a full quadrilateral mesh using overlapping network technology is:

[0069] Decompose the geometric model into several subdomains, each of which can be meshed independently and the overlapping parts of the subdomains can cover the entire geometric model;

[0070] Each subdomain generates a fully quadrilateral mesh individually according to the predefined pattern and mesh generation rules;

[0071] The overlapping part between each subdomain is defined as the overlapping area, and the interpolation boundary within the overlapping area is set;

[0072] The grids of all subdomains are combined to form a complete overlapping network, and the interpolation algorithm is used to perform data transition in the overlapping area.

[0073] The interpolation boundary is the overlapping area between different grid areas in the overlapping grid technology, which is used for data exchange and transition. By interpolating on the interpolation boundary, the consistency and smooth transition of the solution between the overlapping grid areas can be ensured, thereby improving the accuracy and stability of the overall calculation;

[0074] The method of using interpolation algorithm for data transition is:

[0075] Calibrate subdomain A and subdomain B, the overlapping area of ​​subdomain A and subdomain B is A∩B, and all grid nodes within the overlapping area A∩B are used as interpolation points;

[0076] Assign an interpolation weight to each interpolation point. The interpolation point is between the grid node (j, k) of A and the grid node (m, n) of B. The interpolation weight of the grid node (j, k) and the grid node (m, n) is W. jk and W mn ;

[0077] According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point has Q = W jk *Q A(j,k) +W mn *Q B(m,n) , where Q A(j,k) is the calculated value of the grid node (j, k of A, Q B(m,n) is the calculated value of the grid node (m,n) of B.

[0078] It should be noted that the interpolation weight of the interpolation point is determined by the relative position of the interpolation point in the overlapping area.

[0079] Overlapping mesh technology uses multiple independent mesh regions to cover complex geometric shapes. Each region can be divided using a quadrilateral mesh suitable for its specific shape, making the geometric shape expression more accurate and effectively reducing geometric approximation errors.

[0080] In areas with complex flow characteristics or requiring high computational accuracy, the mesh can be locally refined without affecting the meshing of other areas, helping to obtain more accurate computational results in key areas while maintaining the efficiency and manageability of the overall mesh;

[0081] Overlapping meshing technology can decompose complex geometries into multiple smaller, more manageable mesh regions, simplifying the mesh generation process and reducing the time required to generate high-quality meshes;

[0082] Overlapping grid technology provides greater flexibility to define grid partitioning strategies, and can select different grid types and densities according to different flow characteristics or geometric requirements, which is conducive to obtaining better computational performance and results based on predefined patterns.

[0083] The generation of full quadrilateral meshes is coordinated and scheduled using a distributed architecture platform;

[0084] Through the distributed architecture platform, the mesh generation task is decomposed into four subtasks: geometric model cleaning, local refinement, mesh generation and quality inspection, which are carried out simultaneously, and combined with multiple servers for parallel computing;

[0085] Computing resources are dynamically allocated according to server load conditions, and load balancing is performed in a multi-server environment.

[0086] Use post-processing technology to optimize the full quadrilateral mesh and perform quality checks on the generated full quadrilateral mesh;

[0087] The logic of coordinating the generation of full quadrilateral meshes through a distributed architecture is:

[0088] In the calibration distributed platform, the server processes the geometric model cleaning, local refinement, mesh generation, and quality inspection subtasks in order of a, b, c, and d, respectively. The number of times each subtask is processed is N. a 、N b 、N c 、N d , set Sub = {N a ,N b ,N c ,N d}, the total number of subtask processing is N u , the servers of the distributed architecture platform are numbered as e, the total number of servers is G, and e = {1, 2, 3…G}, where G is a positive integer, the average maximum aspect ratio of the grid units generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid units generated by each server in the same predefined mode is Ar2, and the average half-angle value of the grid units generated by each server in the same predefined mode is Hs, then the scheduling difference coefficient is In the formula, α and β are and The proportional coefficient, and α and β are both positive numbers. The maximum number of times the server processes subtasks in the distributed architecture platform is max(Sub), and the minimum number is min(Sub).

[0089] The scheduling threshold St is preset. When the calculated scheduling difference coefficient Sd is greater than or equal to the scheduling threshold St, the data consistency of distributed processing is stable, the efficiency of grid generation is stable, and collaborative management is not performed;

[0090] When the resulting scheduling difference coefficient Sd is less than the scheduling threshold St, the data consistency of distributed processing fluctuates, the load generated by the grid is unbalanced, and the efficiency is low, prompting technicians to perform cross-server collaborative management.

[0091] In-Situ post-processing technology is a technology that directly processes and analyzes the generated data in real time during the calculation process. Compared with traditional post-processing methods, In-Situ post-processing technology does not need to save the data to disk for post-processing after the calculation is completed, but performs real-time processing while the data is generated. In-Situ post-processing technology embeds the visualization calculation code into the simulation calculation program, bypasses the I / O bottleneck, and shares memory data with the simulation calculation program, thereby realizing real-time data processing and visualization;

[0092] The method for post-processing a full quad mesh is:

[0093] In-Situ post-processing technology is used to perform real-time analysis on the intermediate data of the mesh generation process, check the mesh quality indicators, check the intermediate results of the mesh generation process according to predefined quality standards, identify the mesh units that do not meet the quality standards, record the location and type of the problem, feedback the quality inspection results, and dynamically adjust and optimize the generation process.

[0094] It should be noted that mesh quality indicators include unit shape, size, aspect ratio, angle, etc. Methods for dynamically adjusting and optimizing the generation process include adjusting mesh density, improving mesh division algorithms, repairing irregular units, etc.

[0095] Combined with the results of the full quadrilateral mesh quality check and the calculation accuracy in the mesh generation process, the generation efficiency of the full quadrilateral mesh is evaluated, and a mesh quality control strategy is proposed based on the evaluation results;

[0096] The non-standard mesh unit identified in the calibration quality check is Gt, t is the number of the mesh unit, the number of times the mesh generation process is adjusted and optimized by post-processing technology is Op, the internal angle difference between adjacent quadrilateral units is ΔΘ, the distortion of adjacent units is ΔS, the gradient of the size of the mesh unit is Gl, the computational resource consumption of mesh generation is Cp, the mesh generation time is Tg, the consistency efficiency coefficient is Ec = |St-Sd|, and the generation efficiency of the full quadrilateral mesh is evaluated. The calculation expression is In the formula, Ev is the evaluation coefficient, and the evaluation threshold Et is preset. When the calculated evaluation coefficient Ev is greater than or equal to the evaluation threshold Et, the generation quality of the full quadrilateral mesh is low. The quality control strategy of the reported mesh is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization rate of computing resources;

[0097] It should be noted that the distortion is an indicator that describes the degree to which the four vertices of a quadrilateral are coplanar, that is, whether the quadrilateral is twisted into a non-planar shape.

[0098] When the calculated evaluation coefficient Ev is less than the evaluation threshold Et, the grid quality control strategy is not notified.

[0099] This application combines a distributed interactive platform with post-processing technology verification to verify and optimize the generation of full quadrilateral meshes based on predefined patterns, improve the utilization of computing resources, coordinate the generation of meshes, speed up the efficiency of mesh generation while ensuring data consistency, and use overlapping network technology to improve the accuracy and speed of mesh generation, thus resolving the contradiction between data consistency and processing speed when processing large-scale complex geometric models.

[0100] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0101] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of computer program goods. The computer program goods include one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0102] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0103] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0104] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0105] If the functions are implemented in the form of software functional units and sold or used as independent goods, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of software goods, and the computer software goods are stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0106] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for generating a full quadrilateral mesh based on a predefined pattern, characterized in that: The specific steps include cleaning the geometric model, setting a predefined mode for mesh generation according to the cleaned geometric model, and defining mesh generation rules; Meshing is performed according to the geometric model and predefined patterns, and full quadrilateral meshes are generated based on overlapping network technology. The generation of full quadrilateral meshes is coordinated and scheduled based on a distributed architecture platform. Use post-processing technology to optimize the full quadrilateral mesh and perform quality checks on the generated full quadrilateral mesh; The generation efficiency of the full quadrilateral mesh is evaluated based on the results of the full quadrilateral mesh quality check and the calculation accuracy in the mesh generation process, and a mesh quality control strategy is proposed based on the evaluation results.

2. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 1, characterized in that: The steps to clean up the geometry model are: Delete small features, fill small holes in geometric models, delete or merge small edges and faces in geometric models; Simplify the model, merge multiple small coplanar faces in the geometric model into one large face, and simplify curves and surfaces; Repair geometric defects, fill gaps and gaps in geometric models, and eliminate overlapping and intersecting parts in geometric models; Optimize the geometric model, smooth the surface, eliminate wrinkles and irregular shapes, and make the surface normal vector direction consistent.

3. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 2, characterized in that: The predefined modes for mesh generation are set as follows: Set the basic mesh density for the cleaned geometric model and set the refined mesh density for the refined area in the geometric model; Determine the orientation of the mesh based on the shape of the geometric model and the directionality of the physical application; Select the meshing strategy based on the shape of the geometric model; The rules and constraints for grid generation are set according to grid application requirements, including the maximum and minimum aspect ratios of grid cells, the internal angle range of grid cells, and smooth transition constraints between grid cells.

4. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 3, characterized in that: The process of generating a full quadrilateral mesh using overlapping network technology is: Decompose the geometric model into several subdomains, each of which can be meshed independently and the overlapping parts of the subdomains can cover the entire geometric model; Each subdomain generates a fully quadrilateral mesh individually according to the predefined pattern and mesh generation rules; The overlapping part between each subdomain is defined as the overlapping area, and the interpolation boundary within the overlapping area is set; The grids of all subdomains are combined to form a complete overlapping network, and the interpolation algorithm is used to perform data transition in the overlapping area.

5. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 4, characterized in that: The method of using interpolation algorithm for data transition is: Calibrate subdomain A and subdomain B, the overlapping area of ​​subdomain A and subdomain B is A∩B, and all grid nodes within the overlapping area A∩B are used as interpolation points; Assign an interpolation weight to each interpolation point. The interpolation point is between the grid node (j, k) of A and the grid node (m, n) of B. The interpolation weight of the grid node (j, k) and the grid node (m, n) is W. jk and W mn ; According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point has Q = W jk *Q A(j，k) +W mn *Q B(m,n) , where Q A(j，k) is the calculated value of the grid node (j, k) of A, Q B(m,n) is the calculated value of the grid node (m, n) of B.

6. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 4, characterized in that: The logic of coordinating the generation of full quadrilateral meshes through a distributed architecture is: In the calibration distributed platform, the server processes the geometric model cleaning, local refinement, mesh generation, and quality inspection subtasks in order of a, b, c, and d, respectively. The number of times each subtask is processed is N. a 、N b 、N c 、N d , set Sub = {N a , N b , N c , N d }, the total number of subtask processing is N u , the servers of the distributed architecture platform are numbered as e, the total number of servers is G, and e = {1, 2, 3…G}, where G is a positive integer, the average maximum aspect ratio of the grid units generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid units generated by each server in the same predefined mode is Ar2, and the average half-angle value of the inner angle of the grid units generated by each server in the same predefined mode is Hs, then the scheduling difference coefficient is In the formula, α and β are and The proportional coefficient, and α and β are both positive numbers. The maximum number of times the server processes subtasks in the distributed architecture platform is max(Sub), and the minimum number is min(Sub); The scheduling threshold St is preset. When the calculated scheduling difference coefficient Sd is greater than or equal to the scheduling threshold St, the data consistency of distributed processing is stable, the efficiency of grid generation is stable, and collaborative management is not performed; When the obtained scheduling difference coefficient Sd is less than the scheduling threshold St, the data consistency of distributed processing fluctuates, and the load generated by the grid is unbalanced and inefficient, prompting technicians to perform cross-server collaborative management.

7. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 6, characterized in that: The method for post-processing a full quad mesh is: In-Situ post-processing technology is used to perform real-time analysis on the intermediate data of the mesh generation process, check the mesh quality indicators, check the intermediate results of the mesh generation process according to predefined quality standards, identify the mesh units that do not meet the quality standards, record the location and type of the problem, feedback the quality inspection results, and dynamically adjust and optimize the generation process.

8. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 7, characterized in that: The efficiency of generating full quadrilateral meshes is evaluated, and the logic of mesh quality control strategy is proposed based on the evaluation results: The non-standard mesh unit identified in the calibration quality check is Gt, t is the number of the mesh unit, the number of times the mesh generation process is adjusted and optimized by post-processing technology is Op, the internal angle difference between adjacent quadrilateral units is ΔΘ, the distortion of adjacent units is ΔS, the gradient of the size of the mesh unit is Gl, the computational resource consumption of mesh generation is Cp, the mesh generation time is Tg, the consistency efficiency coefficient is Ec = |St-Sd|, and the generation efficiency of the full quadrilateral mesh is evaluated. The calculation expression is Where, Ev is the evaluation coefficient; The evaluation threshold Et is preset. When the calculated evaluation coefficient Ev is greater than or equal to the evaluation threshold Et, the quality of the generated full quadrilateral mesh is low. The quality control strategy of the informed mesh is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization rate of computing resources. When the calculated evaluation coefficient Ev is less than the evaluation threshold Et, the grid quality control strategy is not notified.

Images