A large-scale unstructured grid wind resistance calculation method for the whole vehicle

By introducing distortion correction method and non-orthogonal correction method in CFD software, combined with the OpenFOAM platform, the kernel black box problem in large-scale grid resistance calculation of the whole vehicle is solved, the calculation stability and accuracy are improved, and the automotive industry is adapted to Requirements have formed a set of common and sustainable improvement analytical methods.

CN118520672BActive Publication Date: 2024-11-12CHONGQING UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410603904.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-15
Publication Date
2024-11-12
Estimated Expiration
2044-05-15

AI Technical Summary

Technical Problem

Existing CFD commercial software has a kernel black box problem in large-scale grid resistance calculations for the whole vehicle, resulting in a decrease in the credibility of the calculation results. Open source platforms such as OpenFOAM have low stability in high-precision calculations, making it difficult to meet the automotive industry. Actual demand.

Method used

The large-scale unstructured grid resistance calculation method of the whole vehicle is adopted, and the convective term and surface normal gradient are corrected by introducing improved distortion correction method and non-orthogonal correction method, combined with the incompressible flow solver of the OpenFOAM platform. Perform numerical calculations to improve the stability and accuracy of the calculation.

Benefits of technology

It realizes the independent control of CFD software, improves the stability and accuracy of calculations, adapts to practical problems in the automotive industry, and forms a set of general and sustainable iterative and improved large-scale grid resistance analysis methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118520672B_ABST
    Figure CN118520672B_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of automobile aerodynamics, and discloses a method for calculating the wind resistance of a large-scale unstructured grid of a whole vehicle, comprising the following steps: establishing a whole vehicle model and a three-dimensional calculation domain and performing grid division, and establishing an incompressible whole vehicle external flow field computational fluid dynamics model; setting boundary conditions and calculation conditions, introducing an improved distortion correction method to correct the convection term, and introducing an improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and the Laplace term; based on the set boundary conditions, calculation conditions, and the corrected convection term and the corrected surface normal gradient and the Laplace term, performing whole vehicle aerodynamic analysis numerical calculation on the incompressible whole vehicle external flow field computational fluid dynamics model, and obtaining the wind resistance coefficient, lift coefficient and pitch force coefficient in a stable state; the method improves the accuracy and stability of the whole vehicle external flow field and aerodynamic analysis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of automobile aerodynamics, and in particular to a method for calculating the wind resistance of a large-scale unstructured grid of a whole vehicle. Background Art

[0002] Simulation analysis is one of the important links in the modern automobile design industry. It can effectively improve the efficiency of the design process of automobile structure and appearance, and can effectively reduce the test cost compared with traditional experimental methods. The stability, accuracy and calculation efficiency of simulation analysis directly affect the length of the automobile design process.

[0003] At present, most automobile companies use foreign commercial software for simulation analysis, especially for CFD simulation of automobile external flow field, which basically relies on software such as starccm+ and fluent. The kernel code of these commercial software is invisible, and only the algorithms provided by them can be used for simulation. Engineers cannot fully master the underlying code of commercial software, which will inevitably lead to a decrease in the credibility of calculation results.

[0004] Some existing open source platforms, such as OpenFOAM, can perform CFD simulations, and the underlying code is transparent and has a variety of optional calculation schemes. However, they are basically limited to small-scale laboratory-level grid simulations. Some high-precision calculation schemes have low stability on large-scale grids of the entire vehicle. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a large-scale unstructured grid wind resistance calculation method for a whole vehicle. By improving the calculation conditions, the black box problem of the existing CFD commercial software kernel is solved, the autonomous control of the CFD software is achieved, and it is adapted to the calculation of actual engineering problems in the automotive industry, while ensuring the stability, accuracy and efficiency of the calculation, thereby forming a large-scale grid whole vehicle wind resistance analysis method with strong versatility and continuous iterative improvement.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0007] A method for calculating wind resistance of a large-scale unstructured grid of a whole vehicle comprises the following steps:

[0008] S1. Establish a vehicle model and a three-dimensional computational domain and perform meshing, and establish an incompressible vehicle external flow field computational fluid dynamics model;

[0009] S2. Setting boundary conditions and calculation conditions, introducing an improved distortion correction method to correct the convection term, introducing an improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and Laplace term, and obtaining the corrected convection term and the corrected surface normal gradient and Laplace term;

[0010] S3. Based on the boundary conditions and calculation conditions set in step S2 and the corrected convection terms and the corrected surface normal gradients and Laplace terms obtained, the incompressible whole vehicle external flow field computational fluid dynamics model established in step S1 is used to perform numerical calculations for the whole vehicle aerodynamic analysis, the flow characteristics of the whole vehicle external flow field are obtained, the velocity distribution and pressure distribution of the whole vehicle external flow field are obtained and post-processed, and the drag coefficient, lift coefficient and pitch force coefficient under stable state are obtained.

[0011] Furthermore, in step S1, the computational fluid dynamics model of the incompressible vehicle external flow field adopts the k-ε turbulence model or the k-ω turbulence model during calculation, and performs numerical calculation based on the incompressible flow solver of the OpenFOAM platform.

[0012] Furthermore, step S2 specifically includes:

[0013] S21. Set the boundary conditions of the three-dimensional calculation domain, where the inlet adopts the velocity boundary condition, the outlet adopts the pressure boundary condition, the wall and the body surface adopt the no-slip boundary condition, and the semi-implicit method of the pressure coupling equation group is used to solve the velocity and pressure coupling;

[0014] S22. The calculation conditions are set as follows: the time term is calculated using the steady-state method, the gradient term is calculated using the Gaussian linear interpolation method after being limited by the unit-limited Minmod gradient limiter, the convection term is calculated using the second-order upwind interpolation method with distortion correction, and an improved non-orthogonal correction method is introduced to perform non-orthogonal correction on the surface normal gradient and the Laplace term, and an improved distortion correction method is introduced to correct the convection term.

[0015] Furthermore, the specific process of introducing the improved distortion correction method to correct the convection term is as follows:

[0016] In the distortion correction method, a limiter is used to limit the gradient term in the distortion correction term, and its calculation formula is:

[0017]

[0018] Where j represents the center point of the interface, j′ represents the body center line PP j The intersection point with the interface, P represents the center point of unit 1, P j represents the center point of unit 2, represents the interface value, λ j represents the distance factor, Φ P represents the variable value of unit 1, represents the variable value of unit 2, represents the corrected surface gradient value, Represents the surface gradient value, d j′jThe body-center line PP j The vector connecting the intersection point j′ of the interface to the center point j of the interface, represents the gradient value of the center point P of unit 1, Represents the center point P of unit 2 j The gradient value of Represents the center point j of the interface to the center point P of unit 2 j The vector of the connection, Represents the center point P of unit 1 to the center point P of unit 2 j For the connected vector, min(·) means taking the minimum value, and max(·) means taking the maximum value.

[0019] Furthermore, the specific process of introducing the improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and the Laplace term is as follows:

[0020] First, when the cells in a three-dimensional unstructured grid with in-plane thickness are in the same plane, the non-orthogonal angle in the original non-orthogonal correction method is the angle between the normal unit vector of the interface and the vector connecting the center points of the cells, that is:

[0021]

[0022] Where n represents the normal unit vector of the interface between unit 1 and unit 2, Represents the center point P of unit 1 to the center point P of unit 2 j The vector of the connecting line, θ1 represents the normal unit vector n of the interface between unit 1 and unit 2 and the center point P of unit 1 to the center point P of unit 2 j Vector of the connection , cos(·) represents the cosine function;

[0023] Secondly, when the cells in a three-dimensional unstructured grid with in-plane thickness are perpendicular to each other, the cells are non-orthogonal grids, and the non-orthogonal angle is redefined as the angle between the vector connecting the center point of the cell to the center point of the interface and the unit vector normal to the interface, that is:

[0024]

[0025] Among them, d Pj represents the vector of the line connecting the center point P of unit 1 to the center point j of the interface, θ2 represents the angle between the vector of the line connecting the center point P of unit 1 to the center point k of the interface and the normal unit vector n of the interface;

[0026] Finally, the maximum value of θ1 and θ2 is taken as the corrected non-orthogonal angle, that is:

[0027] θ=max[θ1,θ2]

[0028] Among them, max[·] means taking the maximum value.

[0029] The present invention has the following beneficial effects:

[0030] The present invention proposes a method for calculating the wind resistance of a large-scale unstructured grid of a whole vehicle. The method has high configurability and can be calculated for different types of automobile shapes and structures, thereby ensuring the stability of the calculation. At the same time, an improved distortion correction method is introduced to correct the convection term, and an improved non-orthogonal correction method is introduced to perform non-orthogonal correction on the surface normal gradient and the Laplace term, thereby improving the accuracy and stability of the external flow field and aerodynamic analysis of the whole vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of the flow of a large-scale unstructured grid wind resistance calculation method for a whole vehicle proposed in the present invention;

[0032] Figure 2 Schematic diagram of correction of distortion of convection term in a three-dimensional unstructured grid with in-plane thickness in an embodiment;

[0033] Figure 3 Schematic diagram of non-orthogonal correction of a three-dimensional unstructured grid with in-plane thickness in an embodiment Figure 1 ;

[0034] Figure 4 Schematic diagram of non-orthogonal correction of a three-dimensional unstructured grid with in-plane thickness in an embodiment Figure 2 ;

[0035] Figure 5 Schematic diagram of non-orthogonal correction of a three-dimensional unstructured grid with in-plane thickness in an embodiment Figure 3 ; DETAILED DESCRIPTION

[0036] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0037] like Figure 1 As shown, a large-scale unstructured grid wind resistance calculation method for a whole vehicle includes the following steps S1-S3:

[0038] S1. Establish a vehicle model and a three-dimensional computational domain and perform meshing, and establish an incompressible vehicle external flow field computational fluid dynamics model.

[0039] In the present embodiment, modeling software is adopted to carry out whole vehicle CAD modeling, and then Star ccm+ is utilized to wrap surface processing using different triangular facets to different automobile component boundaries, and whole vehicle surface mesh model is generated and surface mesh file (STL file) is exported, and the boundary names of each component of automobile and corresponding mesh refinement level are extracted from the surface mesh file. The background hexahedral mesh in the calculation area is generated by the block mesh generator (blockMesh) of the OpenFOAM platform, and its purpose is to generate a basic mesh, so that the mesh division processor (snappyHexMeshDict) file is convenient to remove and process the non-fluid area according to the surface mesh model. Then the surface feature extractor (surfaceFeatureExtractDict) file of the OpenFOAM platform is used to extract feature surface, and according to the point coordinates in the surface mesh model, the edge and surface information therein are extracted, such as the direction of the face, the curvature problem, the position of the face, so as to generate the calculation domain body mesh of good quality. Finally, the meshing processor (snappyHexMeshDict) file of the OpenFOAM platform is used to mesh the characteristic edges and characteristic faces in the geometry according to the extracted characteristic surfaces, remove the meshes in the non-fluid area, and only keep the meshes in the fluid area; then the mesh is continuously refined. The specific process is: based on the background hexahedral mesh in the calculation area, according to the component boundary name and the corresponding mesh refinement level, set the encrypted boundary, mesh encryption level, number of transition mesh layers, and basic parameters of the boundary layer mesh; when the meshing processor is running, the background hexahedral mesh in the calculation area is gradually divided into two finer meshes, and the mesh refinement level is increased by one level when the mesh size is divided into two. After the refinement is performed, according to the geometric feature information of the car surface extracted by the surface feature extractor, the mesh near the refined car body is fitted to the car body surface. After the mesh is fitted, the boundary layer mesh is generated on the car body surface according to the set boundary layer parameters. After refinement, fitting, and boundary layer generation, the processed mesh is quality checked according to mesh quality parameters, including non-orthogonal angles, distortion, aspect ratio, and skew, and meshes that do not meet quality standards are screened out, and the process of refining, fitting, and boundary layer generation is repeated before refinement. For meshes that meet quality standards, the iterative refinement process is skipped, and finally a set of body meshes for the vehicle external flow field calculation domain with component boundary refinement is generated based on the surface mesh file information.

[0040] Specifically, the computational fluid dynamics model of the incompressible vehicle external flow field in step S1 adopts the k-ε turbulence model or the k-ω turbulence model during calculation, and performs numerical calculation based on the incompressible flow solver of the OpenFOAM platform.

[0041] In this embodiment, in order to accurately analyze the aerodynamic characteristics of the whole vehicle, the incompressible flow solver based on the OpenFOAM platform is used to perform numerical calculations on the established incompressible computational fluid dynamics model of the external flow field of the whole vehicle, wherein the calculation formula of the incompressible computational fluid dynamics model of the external flow field of the whole vehicle can adopt the k-ε turbulence model or the k-ω turbulence model.

[0042] S2. Set boundary conditions and calculation conditions, introduce an improved distortion correction method to correct the convection term, introduce an improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and Laplace term, and obtain the corrected convection term and the corrected surface normal gradient and Laplace term.

[0043] Specifically, step S2 specifically includes S21-S22:

[0044] S21. Set the boundary conditions of the three-dimensional calculation domain, where the inlet adopts the velocity boundary condition, the outlet adopts the pressure boundary condition, the wall and the body surface adopt the no-slip boundary condition, and the semi-implicit method of the pressure coupling equation group is used to solve the velocity and pressure coupling;

[0045] S22. The calculation conditions are set as follows: the time term is calculated using the steady-state method, the gradient term is calculated using the Gaussian linear interpolation method after being limited by the unit-limited Minmod gradient limiter, the convection term is calculated using the second-order upwind interpolation method with distortion correction, and an improved non-orthogonal correction method is introduced to perform non-orthogonal correction on the surface normal gradient and the Laplace term, and an improved distortion correction method is introduced to correct the convection term.

[0046] Specifically, the specific process of introducing the improved distortion correction method to correct the convection term is as follows:

[0047] In the distortion correction method, a limiter is used to limit the gradient term in the distortion correction term, and its calculation formula is:

[0048]

[0049] Where j represents the center point of the interface, j′ represents the body center line PP j The intersection point with the interface, P represents the center point of unit 1, P j represents the center point of unit 2, represents the interface value, λ j represents the distance factor, Φ P represents the variable value of unit 1, represents the variable value of unit 2, represents the corrected surface gradient value, Represents the surface gradient value, d j′j The body-center line PP jThe vector connecting the intersection point j′ of the interface to the center point j of the interface, represents the gradient value of the center point P of unit 1, Represents the center point P of unit 2 j The gradient value of Represents the center point j of the interface to the center point P of unit 2 j The vector of the connection, Represents the center point P of unit 1 to the center point P of unit 2 j For the connected vector, min(·) means taking the minimum value, and max(·) means taking the maximum value.

[0050] like Figure 2 As shown, Figure 2 Schematic diagram of the correction of the distortion of the convection term in a three-dimensional unstructured grid with in-plane thickness, where unit 1 and unit 2 are three-dimensional unstructured grids with in-plane thickness, S j represents the interface normal vector between element 1 and element 2, and n represents the interface normal unit vector between element 1 and element 2. The formula for interpolating the unit node value to the interface value using traditional linear interpolation is: When the surface gradient value is too large, the calculation using this formula will produce errors. Therefore, an improved distortion correction method is introduced into the traditional linear interpolation. The limiter is used to limit the gradient term in the distortion correction term to strictly limit the surface gradient value within the range of the adjacent unit gradient value. The formula is: This achieves the correction of the distortion of the convection term. Then the corrected surface gradient value Add formula On the right, a new calculation formula is formed That is, the final formula for calculating the interface value by linear interpolation including the distortion correction method is obtained.

[0051] In this embodiment, since the non-orthogonal correction method is the key to affecting the stability of calculation, it is mainly used to correct the calculation error caused by the unstructured grid. However, the definition of non-orthogonal angles in the OpenFOAM platform will cause some unstructured grids to be mistaken for orthogonal grids, which seriously affects the calculation accuracy. Therefore, an improved non-orthogonal correction method is introduced to perform non-orthogonal correction on the surface normal gradient and the Laplace term.

[0052] Specifically, the specific process of introducing the improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and the Laplace term is as follows:

[0053] First, when the cells in a three-dimensional unstructured grid with in-plane thickness are in the same plane, the non-orthogonal angle in the original non-orthogonal correction method is the angle between the normal unit vector of the interface and the vector connecting the center points of the cells, that is:

[0054]

[0055] Where n represents the normal unit vector of the interface between unit 1 and unit 2, Represents the center point P of unit 1 to the center point P of unit 2 j The vector of the connecting line, θ1 represents the normal unit vector n of the interface between unit 1 and unit 2 and the center point P of unit 1 to the center point P of unit 2 j Vector of the connection , cos(·) represents the cosine function.

[0056] Secondly, when the cells in a three-dimensional unstructured grid with in-plane thickness are perpendicular to each other, the cells are non-orthogonal grids, and the non-orthogonal angle is redefined as the angle between the vector connecting the center point of the cell to the center point of the interface and the unit vector normal to the interface, that is:

[0057]

[0058] Among them, d Pj represents the vector of the line connecting the center point P of unit 1 to the center point j of the interface, and θ2 represents the angle between the vector of the line connecting the center point P of unit 1 to the center point j of the interface and the normal unit vector n of the interface.

[0059] Finally, the maximum value of θ1 and θ2 is taken as the corrected non-orthogonal angle, that is:

[0060] θ=max[θ1,θ2]

[0061] Among them, max[·] means taking the maximum value.

[0062] like Figure 3-Figure 5 As shown in the figure, they are three-dimensional unstructured grids with in-plane thickness in different cases. Figure 3 Unit 1 and unit 2 are three-dimensional unstructured grids with in-plane thickness, and unit 1 and unit 2 are in the same plane. Figure 3 The non-orthogonal angle in is θ1, that is, the non-orthogonal angle is defined as the angle between the normal unit vector of the interface and the vector connecting the unit center points; however, when Figure 4 In the case shown, that is, unit 1 and unit 2 are perpendicular to each other or not in the same plane, if the angle between the normal unit vector of the interface and the vector connecting the unit center points is still used as the non-orthogonal angle, then Figure 4 The non-orthogonal angle between the corresponding unit 1 and unit 2 is 0, so the unit 1 and unit 2 will be defined as an orthogonal grid, but in fact, Figure 4 Unit 1 and unit 2 in the grid are non-orthogonal, so the non-orthogonal angle needs to be redefined. Figure 5As shown in the figure, θ2 is used as the non-orthogonal angle, which is defined as the angle between the vector connecting the center point of the unit to the center point of the interface and the normal unit vector of the interface; finally, by calculating θ1 and θ2 between the units, the maximum value is selected as the corrected non-orthogonal angle to ensure the stability of the calculation.

[0063] S3. Based on the boundary conditions and calculation conditions set in step S2 and the corrected convection terms and the corrected surface normal gradients and Laplace terms obtained, the incompressible whole vehicle external flow field computational fluid dynamics model established in step S1 is used to perform numerical calculations for the whole vehicle aerodynamic analysis, the flow characteristics of the whole vehicle external flow field are obtained, the velocity distribution and pressure distribution of the whole vehicle external flow field are obtained and post-processed, and the drag coefficient, lift coefficient and pitch force coefficient under stable state are obtained.

[0064] In this embodiment, based on the incompressible flow solver of the OpenFOAM platform, the numerical calculation of the aerodynamic analysis of the whole vehicle is performed with the boundary conditions and calculation conditions set in the above steps to obtain the numerical calculation results of the flow of the external flow field of the whole vehicle. By obtaining the flow characteristics of the external flow field of the whole vehicle, the velocity distribution and pressure distribution in the flow field of the calculation domain are obtained, and the flow field of the whole vehicle is post-processed to obtain the drag coefficient, lift coefficient, and pitch moment coefficient in the stable state.

[0065] A large-scale unstructured grid wind resistance calculation method for a whole vehicle proposed in the present invention has the following characteristics: firstly, the software code is transparent and highly adjustable, and the code level can be optimized and adjusted according to different calculation tasks; secondly, the calculation method is highly adaptable and can be calculated for different types of automobile shapes and structures, and the calculation stability can be guaranteed; finally, the incompressible flow solver based on the OpenFOAM platform can improve the accuracy and stability of the incompressible flow solver for the external flow field and aerodynamic analysis of the whole vehicle by improving the distortion correction and non-orthogonal correction methods.

[0066] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0067] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A large-scale unstructured grid wind resistance calculation method for a whole vehicle, characterized in that: The following steps are involved: S1. Establish a vehicle model and a three-dimensional computational domain and perform meshing, and establish an incompressible vehicle external flow field computational fluid dynamics model; S2. Setting boundary conditions and calculation conditions, introducing an improved distortion correction method to correct the convection term, introducing an improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and Laplace term, and obtaining the corrected convection term and the corrected surface normal gradient and Laplace term; S3. Based on the boundary conditions and calculation conditions set in step S2 and the corrected convection terms and the corrected surface normal gradients and Laplace terms obtained, the incompressible whole vehicle external flow field computational fluid dynamics model established in step S1 is used to perform numerical calculations for the whole vehicle aerodynamic analysis, the flow characteristics of the whole vehicle external flow field are obtained, the velocity distribution and pressure distribution of the whole vehicle external flow field are obtained and post-processed, and the drag coefficient, lift coefficient and pitch force coefficient under stable state are obtained.

2. The method for calculating the wind resistance of a vehicle using large-scale unstructured grids according to claim 1 is characterized in that: In step S1, the computational fluid dynamics model of the incompressible vehicle external flow field adopts the k-ε turbulence model or the k-ω turbulence model during calculation, and performs numerical calculation based on the incompressible flow solver of the OpenFOAM platform.

3. The method for calculating the wind resistance of a vehicle using large-scale unstructured grids according to claim 1 is characterized in that: Step S2 specifically includes: S21. Set the boundary conditions of the three-dimensional calculation domain, where the inlet adopts the velocity boundary condition, the outlet adopts the pressure boundary condition, the wall and the body surface adopt the no-slip boundary condition, and the semi-implicit method of the pressure coupling equation group is used to solve the velocity and pressure coupling; S22. The calculation conditions are set as follows: the time term is calculated using the steady-state method, the gradient term is calculated using the Gaussian linear interpolation method after being limited by the unit-limited Minmod gradient limiter, the convection term is calculated using the second-order upwind interpolation method with distortion correction, and an improved non-orthogonal correction method is introduced to perform non-orthogonal correction on the surface normal gradient and the Laplace term, and an improved distortion correction method is introduced to correct the convection term.

4. The method for calculating the wind resistance of a vehicle using large-scale unstructured grids according to claim 3 is characterized in that: The specific process of introducing the improved distortion correction method to correct the convection term is as follows: In the distortion correction method, a limiter is used to limit the gradient term in the distortion correction term, and its calculation formula is: Where i represents the center point of the interface, i′ represents the body center line PP j The intersection point with the interface, P represents the center point of unit 1, P j represents the center point of unit 2, represents the interface value, λ j represents the distance factor, Φ P represents the variable value of unit 1, represents the variable value of unit 2, represents the corrected surface gradient value, Represents the surface gradient value, d j′j The body-center line PP j The vector connecting the intersection point j′ of the interface to the center point j of the interface, represents the gradient value of the center point P of unit 1, Represents the center point P of unit 2 j The gradient value of Represents the center point j of the interface to the center point P of unit 2 j The vector of the connection, Represents the center point P of unit 1 to the center point P of unit 2 j For the connected vector, min(·) means taking the minimum value, and max(·) means taking the maximum value.

5. The method for calculating the wind resistance of a vehicle using large-scale unstructured grids according to claim 3 is characterized in that: The specific process of introducing the improved non-orthogonal correction method to perform non-orthogonal correction on the surface normal gradient and the Laplace term is as follows: First, when the cells in a three-dimensional unstructured grid with in-plane thickness are in the same plane, the non-orthogonal angle in the original non-orthogonal correction method is the angle between the normal unit vector of the interface and the vector connecting the center points of the cells, that is: Where n represents the normal unit vector of the interface between unit 1 and unit 2, Represents the center point P of unit 1 to the center point P of unit 2 j The vector of the connecting line, θ1 represents the normal unit vector n of the interface between unit 1 and unit 2 and the center point P of unit 1 to the center point P of unit 2 j Vector of the connection , cos(·) represents the cosine function; Secondly, when the cells in a three-dimensional unstructured grid with in-plane thickness are perpendicular to each other, the cells are non-orthogonal grids, and the non-orthogonal angle is redefined as the angle between the vector connecting the center point of the cell to the center point of the interface and the unit vector normal to the interface, that is: Among them, d Pj represents the vector of the line connecting the center point P of unit 1 to the center point j of the interface, θ2 represents the angle between the vector of the line connecting the center point P of unit 1 to the center point j of the interface and the normal unit vector n of the interface; Finally, the maximum value of θ1 and θ2 is taken as the corrected non-orthogonal angle, that is: θ=max[θ1,θ2] Among them, max[·] means taking the maximum value.