Automatic optimization method for independence of supersonic pneumatic simulation grid

Abstract

The invention discloses a supersonic pneumatic simulation grid independence automatic optimization method. The method comprises the following steps: importing target geometry; the importance degrees of different parts of the model are distinguished, and in supersonic pneumatic simulation, a fluid domain is divided into a boundary layer, a shock wave region, a mainstream region and a far field part; determining the grid scale of each fluid region under different grid precisions, determining the boundary layer scale according to the incoming flow Mach number in the supersonic pneumatic simulation process, and keeping the first boundary layer scale unchanged no matter how to change the grid precision; generating three sets of rough, medium and fine initial grids according to the determined grid scales; carrying out simulation calculation on the basis of the three sets of grids; comparing simulation results of the three sets of grids, and optimizing the grids by automatically adjusting grid division parameters and a grid division algorithm; and selecting a grid with medium quality in an iteration result for subsequent calculation. According to the invention, the efficiency of pneumatic simulation work can be improved.

Description

Technical Field

[0001] The invention relates to the field of aerodynamic simulation, and in particular to a supersonic aerodynamic simulation grid-independence automatic optimization method. Background Art

[0002] Meshing is an important step in starting a simulation. From the principle of simulation analysis, the finer the mesh, the higher the accuracy of the simulation results. However, in actual engineering design and application, the computational cost will increase sharply with the increase in the number of meshes, and after the mesh accuracy reaches a certain number, further increasing the number of meshes will not have a significant effect on the increase in simulation accuracy. Therefore, in actual engineering applications, the appropriate mesh should be selected according to the calculation accuracy, so as to obtain the best possible results with a lower calculation cost.

[0003] Grid independence is a basic requirement for the reliability of aerodynamic simulation results. When a set of grids is so fine that the impact on the results can be ignored, the grid can be considered to have grid independence. In actual operation, grid division is mainly based on practical experience and requires manual iteration, which makes the efficiency of aerodynamic simulation work low. Summary of the invention

[0004] The invention provides a supersonic aerodynamic simulation grid independence automatic optimization method, and the invention aims to automate the process to improve the efficiency of aerodynamic simulation work.

[0005] The technical solutions to the above problems are as follows: A supersonic aerodynamic simulation grid-independence automatic optimization method is used to automate grid division in actual operation, comprising the following steps: Step 1: Import the target geometry; Step 2: Distinguish the importance of different parts of the model. In supersonic aerodynamic simulation, the fluid domain is divided into boundary layer, shock wave region, mainstream region and far field; Step 3: Determine the grid scale of each fluid region under different grid accuracy. In the supersonic aerodynamic simulation process, the boundary layer scale is determined by the incoming flow Mach number. No matter how the grid accuracy is changed, the scale of the first boundary layer remains unchanged. Step 4: Generate three sets of initial grids, namely coarse, medium and fine, according to the determined grid scale; Step 5: Perform simulation calculations based on the above three sets of grids; Step 6: Compare the simulation results of the three sets of meshes and optimize the meshes by automatically adjusting the meshing parameters and meshing algorithm; Step 7: Select the medium-quality mesh from the iterative results for subsequent calculations.

[0006] Furthermore, in step 6, the process of comparing the simulation results of the three sets of grids is as follows: When there is a significant difference between the simulation results of the fine grid and the medium grid, it means that the grid is too coarse, so the grid should be refined and the simulation should be performed again. When the simulation results of the coarse and medium grids are close, it means that the grid accuracy is too high, and the grid size should be increased to perform simulation calculations again; Iteration is performed automatically according to the simulation results until the three sets of grids meet the requirements that the simulation results of the coarse grid are significantly different from those of the medium grid, and the simulation results of the medium grid and the fine grid are consistent within the accuracy range.

[0007] Furthermore, in step 6, the grid partitioning algorithm includes the following methods: binary algorithm, quadtree algorithm, octree algorithm, centroid segmentation algorithm and maximum-minimum weight spanning tree algorithm.

[0008] Furthermore, the mesh division parameters include the following indicators: aspect ratio of the division unit, deformation degree of the division unit, minimum angle of the division unit, volume ratio of adjacent units and number of slender units.

[0009] Furthermore, the division unit is any one of a data structure consisting of division node coordinates and connection relationships, a grid file, and a grid image.

[0010] The present invention automatically iterates according to the simulation results until the three sets of grids meet the requirements that the simulation results of the coarse grid are significantly different from those of the medium grid, and the simulation results of the medium grid and the fine grid are consistent within the accuracy range. Therefore, the present invention has a wide range of industry application value in improving the efficiency and accuracy of simulation analysis and reducing development costs by building an automated process for grid division. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0012] In order to more clearly illustrate the examples of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other implementation methods can be obtained according to the drawings in the specification without paying creative work, and these implementation methods are still within the protection scope of the claims of the present invention.

[0013] like Figure 1 As shown, a supersonic aerodynamic simulation grid-independence automatic optimization method of the present invention is used to automatically divide the grid in actual operation to improve the efficiency of aerodynamic simulation work, including the following steps: Step 1: Import the target geometry. In this step, the imported target geometry file is a CAD file or an iges file.

[0014] Step 2: Distinguish the importance of different parts of the model. In supersonic aerodynamic simulation, the fluid domain is divided into boundary layer, shock wave region, mainstream region and far field.

[0015] Step 3: Determine the grid scale of each fluid region under different grid accuracy. In the supersonic aerodynamic simulation process, the boundary layer scale is determined by the incoming flow Mach number. No matter how the grid accuracy is changed, the scale of the first boundary layer remains unchanged.

[0016] Step 4: Generate three sets of initial grids, namely coarse, medium and fine, according to the determined grid scale. This step is implemented by macro-controlled commercial software.

[0017] Step 5: Perform simulation calculations based on the above three sets of grids. In this step, the simulation calculation process is implemented through macro control commercial software.

[0018] Step 6: Compare the simulation results of the three sets of grids, optimize the grids by automatically adjusting the grid division parameters and the grid division algorithm, reduce the number of manual iterations, and improve the efficiency of aerodynamic simulation.

[0019] The process of comparing the simulation results of the three sets of meshes is as follows: (1) When there is a significant difference between the simulation results of the fine grid and the medium grid, it means that the grid is too coarse. Increase the grid density and perform the simulation again.

[0020] (2) When the simulation results of the coarse and medium grids are close, it means that the grid accuracy is too high. Increase the grid size and perform the simulation again.

[0021] (3) Iterate automatically according to the simulation results until the three sets of grids meet the requirements that the simulation results of the coarse grid are significantly different from those of the medium grid, and the simulation results of the medium grid and the fine grid are consistent within the accuracy range.

[0022] Based on practical experience, identify and apply appropriate meshing parameters and algorithms to generate high-quality meshes.

[0023] In step 6, the grid partitioning algorithm includes the following methods: binary algorithm, quadtree algorithm, octree algorithm, centroid segmentation algorithm and maximum and minimum weight spanning tree algorithm.

[0024] In step 6, the mesh partitioning parameters include the following indicators: aspect ratio of the partitioning unit, deformation degree of the partitioning unit, minimum angle of the partitioning unit, volume ratio of adjacent units and number of slender units. The partitioning unit is any one of a data structure consisting of partitioning node coordinates and connection relationships, a mesh file and a mesh image.

[0025] Step 7: Select the medium-quality mesh from the iterative results for subsequent calculations.

Claims

1. A supersonic aerodynamic simulation grid-independence automatic optimization method for automatic grid division in actual operation, characterized in that: The following steps are involved: Step 1: Import the target geometry; Step 2: Distinguish the importance of different parts of the model. In supersonic aerodynamic simulation, the fluid domain is divided into boundary layer, shock wave region, mainstream region and far field; Step 3: Determine the grid scale of each fluid region under different grid accuracy. In the supersonic aerodynamic simulation process, the boundary layer scale is determined by the incoming flow Mach number. No matter how the grid accuracy is changed, the scale of the first boundary layer remains unchanged. Step 4: Generate three sets of initial grids, namely coarse, medium and fine, according to the determined grid scale; Step 5: Perform simulation calculations based on the above three sets of grids; Step 6: Compare the simulation results of the three sets of meshes and optimize the meshes by automatically adjusting the meshing parameters and meshing algorithm; Step 7: Select the medium-quality mesh from the iterative results for subsequent calculations.

2. According to the supersonic aerodynamic simulation grid-independence automatic optimization method of claim 1, in step 6, the process of comparing the simulation results of the three sets of grids is as follows: When there is a significant difference between the simulation results of the fine grid and the medium grid, it means that the grid is too coarse, so the grid should be refined and the simulation should be performed again. When the simulation results of the coarse and medium grids are close, it means that the grid accuracy is too high, and the grid size should be increased to perform simulation calculations again; Iteration is performed automatically according to the simulation results until the three sets of grids meet the requirements that the simulation results of the coarse grid are significantly different from those of the medium grid, and the simulation results of the medium grid and the fine grid are consistent within the accuracy range.

3. The method for automatic optimization of supersonic aerodynamic simulation grid independence according to claim 1, characterized in that: In step 6, the grid partitioning algorithm includes the following methods: binary algorithm, quadtree algorithm, octree algorithm, centroid segmentation algorithm and maximum and minimum weight spanning tree algorithm.

4. According to the supersonic aerodynamic simulation grid-independence automatic optimization method of claim 1, in step 6, the grid division parameters include the following indicators: the aspect ratio of the division unit, the deformation degree of the division unit, the minimum angle of the division unit, the volume ratio of adjacent units and the number of slender units.

5. The method for automatic optimization of supersonic aerodynamic simulation grid independence according to claim 4, wherein the division unit is any one of a data structure consisting of division node coordinates and connection relationships, a grid file, and a grid image.

Images