A method for generating full quadrilateral mesh based on predefined patterns

Abstract

The present invention discloses a method for generating a full quadrilateral mesh based on a predefined pattern, which specifically relates to the technical field of mesh generation. The specific steps include cleaning a geometric model, setting a predefined pattern for mesh generation according to the cleaned geometric model, defining mesh generation rules, meshing according to the geometric model and the predefined pattern, generating a full quadrilateral mesh based on an overlapping mesh technology, collaboratively scheduling the generation of the full quadrilateral mesh using a distributed architecture platform, optimizing the full quadrilateral mesh using a post-processing technology, performing a quality check on the generated full quadrilateral mesh, evaluating the generation efficiency of the full quadrilateral mesh based on the results of the full quadrilateral mesh quality check and the calculation accuracy in the mesh generation process, and proposing a mesh quality control strategy based on the evaluation results, thereby effectively solving the contradiction between data consistency and processing speed when processing large-scale complex geometric models.

Description

Technical Field

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

[0002] A full quadrilateral mesh is a mesh composed of quadrilateral elements and is used for regional discretization in numerical simulations. Compared with triangular meshes, a full quadrilateral mesh can provide better accuracy and stability, but it is difficult to generate high-quality full quadrilateral meshes in complex geometric shapes. Traditional mesh generation methods may cause the shapes of quadrilateral elements to be severely distorted or irregular when processing complex geometries. Full quadrilateral mesh generation must ensure mesh quality while efficiently handling large-scale computing and storage requirements. This problem is particularly prominent when processing 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 solution: 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 mesh generation rules;

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

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

[0008] The generation efficiency of full quadrilateral mesh is evaluated by combining the results of full quadrilateral mesh quality inspection and the calculation accuracy during mesh generation. 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 by merging multiple coplanar small 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 base mesh density for the cleaned geometry model and the refined mesh density for the refined areas in the geometry 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 the grid application requirements, including the maximum and minimum aspect ratios of the grid cells, the internal angle range of the grid cells, and the smooth transition constraints between grid cells.

[0019] Preferably, the process of generating a full quadrilateral mesh using the overset mesh technique is as follows:

[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 predefined patterns 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 grid, 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] Determine subdomain A and subdomain B. The overlapping area of subdomain A and subdomain B is , with the overlapping area as All grid nodes within are used as interpolation points;

[0026] Assign an interpolation weight to each interpolation point, and the interpolation point is located at the grid node of A and the grid nodes of B Between grid nodes and grid nodes The interpolation weight is and ;

[0027] According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point is , where A is the grid node The calculated value of is the grid node of B The calculated value of .

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

[0029] In the calibration distributed platform, the server processes the subtasks of geometric model cleaning, local refinement, mesh generation, and quality inspection in the order of a, b, c, and d, respectively. The number of times each subtask is processed is ,gather , the total number of subtask processing is , number the servers of the distributed architecture platform as e, the total number of servers is G, and , where G is a positive integer, the average maximum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar2, and the average half-angle value of the inner angle of the grid cells generated by each server in the same predefined mode is Hs. The scheduling difference coefficient is , where and The proportionality coefficient of Are all positive numbers. The maximum number of times a server processes subtasks in a distributed architecture platform is , the minimum value is ;

[0030] A 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, 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 of intermediate data during the mesh generation process, check mesh quality indicators, and check intermediate results during the mesh generation process according to predefined quality standards. Mesh units that do not meet quality standards are identified, and the location and type of problems are recorded. Quality inspection results are fed back, and the generation process is dynamically adjusted and optimized.

[0034] Preferably, the efficiency of generating a 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 grid cell identified in the calibration quality check is Gt, t is the number of the grid cell, the number of times the grid generation process is adjusted and optimized by post-processing technology is Op, and the internal angle difference between adjacent quadrilateral cells is The distortion of adjacent elements is , the gradient of the grid cell size change is Gl, the computational resource consumption of grid generation is Cp, the grid generation time is Tg, and the consistency efficiency coefficient is , the generation efficiency of the full quadrilateral mesh is evaluated, and 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 mesh quality control strategy is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization 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 scheduling of mesh generation, speed up mesh generation efficiency while ensuring data consistency, and use overlapping mesh 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 following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0042] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall 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 mesh generation rules;

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

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

[0046] The generation efficiency of full quadrilateral mesh is evaluated by combining the results of full quadrilateral mesh quality inspection and the calculation accuracy during mesh generation. A mesh quality control strategy is proposed based on the evaluation results.

[0047] A full quadrilateral mesh is a mesh composed of quadrilateral elements and is used for regional discretization in numerical simulations. Compared to triangular meshes, full quadrilateral meshes can provide better accuracy and stability in finite element analysis and fluid dynamics calculations. In typical numerical calculation applications, 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, within the same geometric area, quadrilateral meshes can use fewer elements to achieve the same resolution, thereby reducing computational costs. The uniformity and shape of quadrilateral elements are more regular than those of triangular elements, making them more suitable for high-order numerical methods.

[0048] However, there are still some difficulties in generating full quadrilateral meshes for complex geometric structures and high-precision numerical simulations. Complex geometric shapes are difficult to generate high-quality full quadrilateral meshes. Traditional mesh generation methods may cause the shapes of quadrilateral elements to be severely distorted or irregular when processing complex geometries.

[0049] High-quality meshes require quadrilateral elements to be close to square in shape and have a reasonable aspect ratio. Shape optimization is required during mesh generation to avoid distorted and deformed elements. The mesh must be smooth and uniform, avoiding sudden changes in element size and shape. Smoothness and uniformity are crucial for 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 a high-quality, all-quadrilateral mesh that meets these requirements is extremely challenging.

[0050] Complex geometries often contain multiple topological features. Automatically generating high-quality, fully quadrilateral meshes requires processing complex topological relationships. Developing automated mesh generation algorithms based on predefined patterns that can adaptively generate high-quality, fully quadrilateral meshes under varying geometric and physical conditions is a significant challenge.

[0051] The full-quad 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 fully utilizing multi-core processors and distributed computing resources, it must also ensure data consistency between 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 geometric 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 by merging multiple coplanar small 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 base mesh density for the cleaned geometry model and the refined mesh density for the refined areas in the geometry 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 the grid application requirements, including the maximum and minimum aspect ratios of the grid cells, the internal angle range of the grid cells, and the 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. Common partitioning strategies include mapping method, sweeping method, and region 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 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 the overset mesh 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 partially or completely overlap. 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 the overset mesh technique is as follows:

[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 predefined patterns 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 grid, and the interpolation algorithm is used to perform data transition in the overlapping area.

[0073] Interpolation boundaries are the overlapping areas between different mesh regions in overlapping mesh technology, used for data exchange and transition. By interpolating on interpolation boundaries, we can ensure the consistency and smooth transition of solutions between overlapping mesh regions, thereby improving the accuracy and stability of the overall calculation.

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

[0075] Determine subdomain A and subdomain B. The overlapping area of subdomain A and subdomain B is , with the overlapping area as All grid nodes within are used as interpolation points;

[0076] Assign an interpolation weight to each interpolation point, and the interpolation point is located at the grid node of A and the grid nodes of B Between grid nodes and grid nodes The interpolation weight is and ;

[0077] According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point is , where A is the grid node The calculated value of is the grid node of B The calculated value of .

[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 into quadrilateral meshes 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 calculation results in key areas while maintaining the efficiency and manageability of the overall mesh.

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

[0082] Overlapping grid technology provides greater flexibility in defining grid partitioning strategies. Different grid types and densities can be selected 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. These subtasks are performed simultaneously, and parallel computing is performed on multiple servers.

[0085] Computing resources are dynamically allocated based on 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 inspection on the generated full quadrilateral mesh;

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

[0088] In the calibration distributed platform, the server processes the subtasks of geometric model cleaning, local refinement, mesh generation, and quality inspection in the order of a, b, c, and d, respectively. The number of times each subtask is processed is ,gather , the total number of subtask processing is , number the servers of the distributed architecture platform as e, the total number of servers is G, and , where G is a positive integer, the average maximum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar2, and the average half-angle value of the inner angle of the grid cells generated by each server in the same predefined mode is Hs. The scheduling difference coefficient is , where and The proportionality coefficient of Are all positive numbers. The maximum number of times a server processes subtasks in a distributed architecture platform is , the minimum value is .

[0089] A 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 obtained 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 is a technology that processes and analyzes data generated during the computational process in real time. Compared to traditional post-processing methods, In-Situ post-processing eliminates the need to save data to disk for post-processing after the computation is complete. Instead, it processes data in real time as it is generated. By embedding visualization code into the simulation program, In-Situ post-processing bypasses I / O bottlenecks and shares memory data with the simulation program, enabling real-time data processing and visualization.

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

[0093] In-Situ post-processing technology is used to perform real-time analysis of intermediate data during the mesh generation process, check mesh quality indicators, and check intermediate results during the mesh generation process according to predefined quality standards. Mesh units that do not meet quality standards are identified, and the location and type of problems are recorded. Quality inspection results are fed back, and the generation process is dynamically adjusted and optimized.

[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, and repairing irregular units.

[0095] The efficiency of full quadrilateral mesh generation is evaluated by combining the results of full quadrilateral mesh quality inspection and the computational accuracy during mesh generation. A mesh quality control strategy is proposed based on the evaluation results.

[0096] The non-standard grid cell identified in the calibration quality check is Gt, t is the number of the grid cell, the number of times the grid generation process is adjusted and optimized by post-processing technology is Op, and the internal angle difference between adjacent quadrilateral cells is The distortion of adjacent elements is , the gradient of the grid cell size change is Gl, the computational resource consumption of grid generation is Cp, the grid generation time is Tg, and the consistency efficiency coefficient is , the generation efficiency of the full quadrilateral mesh is evaluated, and the calculation expression is , where Ev is the evaluation coefficient, and the preset evaluation threshold Et is set. 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 mesh quality control strategy is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization 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 scheduling of mesh generation, speed up mesh generation efficiency while ensuring data consistency, and use overlapping mesh 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 formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas 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 using 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 processes or functions described in the embodiments of the present application are 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, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. 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 drive.

[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 skilled 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 beyond the scope of this application.

[0104] Those skilled in the art will 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, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of software goods. The computer software goods are stored in a storage medium and include several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0106] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection 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 based on the geometric model and predefined patterns, and full quadrilateral meshes are generated based on the overset mesh technology. The generation of full quadrilateral meshes is coordinated and scheduled using a distributed architecture platform. Use post-processing technology to optimize the full quadrilateral mesh and perform quality inspection on the generated full quadrilateral mesh; The efficiency of full quadrilateral mesh generation is evaluated by combining the results of full quadrilateral mesh quality inspection and the computational accuracy during mesh generation. A mesh quality control strategy is proposed based on the evaluation results. The predefined modes for mesh generation are set as follows: Set the base mesh density for the cleaned geometry model and the refined mesh density for the refined areas in the geometry 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; Set grid generation rules and constraints based on 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; The logic for coordinating the generation of full quadrilateral meshes through a distributed architecture is as follows: In the calibration distributed platform, the server processes the subtasks of geometric model cleaning, local refinement, mesh generation, and quality inspection in the order of a, b, c, and d, respectively. The number of times each subtask is processed is ,gather , the total number of subtask processing is , number the servers of the distributed architecture platform as e, the total number of servers is G, and , where G is a positive integer, the average maximum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar1, the average minimum aspect ratio of the grid cells generated by each server in the same predefined mode is Ar2, and the average half-angle value of the inner angle of the grid cells generated by each server in the same predefined mode is Hs. The scheduling difference coefficient is , where and The proportionality coefficient of Are all positive numbers. The maximum number of times a server processes subtasks in a distributed architecture platform is , the minimum value is ; A 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, the load generated by the grid is unbalanced and inefficient, prompting technicians to perform cross-server collaborative management.

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 geometric model are: Delete small features, fill small holes in geometric models, delete or merge small edges and faces in geometric models; Simplify the model by merging multiple coplanar small 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. The method for generating a full quadrilateral mesh based on a predefined pattern according to claim 1, wherein: The process of generating a full quadrilateral mesh using the overset mesh technique is as follows: 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 predefined patterns 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 grid, and the interpolation algorithm is used to perform data transition in the overlapping area.

4. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 3, characterized in that: The method of using interpolation algorithm to perform data transition is: Determine subdomain A and subdomain B. The overlapping area of subdomain A and subdomain B is , with the overlapping area as All grid nodes within are used as interpolation points; Assign an interpolation weight to each interpolation point, and the interpolation point is located at the grid node of A and the grid nodes of B Between grid nodes and grid nodes The interpolation weight is and ; According to the interpolation weight, the data is interpolated at the interpolation point, and the calculated value Q of the interpolation point is , where A is the grid node The calculated value of is the grid node of B The calculated value of .

5. The method for generating a full quadrilateral mesh based on a predefined pattern according to claim 1, wherein: The post-processing method for a full quad mesh is: In-Situ post-processing technology is used to perform real-time analysis of intermediate data during the mesh generation process, check mesh quality indicators, and check intermediate results during the mesh generation process according to predefined quality standards. Mesh units that do not meet quality standards are identified, and the location and type of problems are recorded. Quality inspection results are fed back, and the generation process is dynamically adjusted and optimized.

6. A method for generating a full quadrilateral mesh based on a predefined pattern according to claim 5, characterized in that: The efficiency of full quadrilateral mesh generation is evaluated, and the logic of mesh quality control strategy is proposed based on the evaluation results: The non-standard grid cell identified in the calibration quality check is Gt, t is the number of the grid cell, the number of times the grid generation process is adjusted and optimized by post-processing technology is Op, and the internal angle difference between adjacent quadrilateral cells is The distortion of adjacent elements is , the gradient of the grid cell size change is Gl, the computational resource consumption of grid generation is Cp, the grid generation time is Tg, and the consistency efficiency coefficient is , the generation efficiency of the full quadrilateral mesh is evaluated, and 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 mesh quality control strategy is to optimize the geometric model, perform local refinement control, smooth the mesh shape, verify the data simulation convergence effect, and improve the utilization 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