A grid deformation method, device and electronic equipment

By grouping and greedy algorithm processing in the maximum size direction of geometric components, the number of sample points in the RBF model is reduced, the problem of low efficiency of traditional grid deformation methods is solved, and more efficient grid deformation is achieved.

CN118780215BActive Publication Date: 2024-11-19BEIJING INST OF TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411270465.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-19
Estimated Expiration
2044-09-11

AI Technical Summary

Technical Problem

When the traditional RBF-based grid deformation method has too many sample points, the modeling time is too long, resulting in low grid deformation efficiency.

Method used

The RBF sub-model is obtained by using a uniform grouping method in the direction of the maximum size of the geometric components, and a greedy algorithm is used for the sample points in each sub-mode to obtain the RBF sub-model, reducing the number of sample points and improving modeling efficiency.

Benefits of technology

The modeling time is shortened, the computational memory is saved, the grid deformation efficiency is improved, and the accuracy of the RBF submodel can be adjusted by adjusting the maximum interpolation error in the greedy algorithm.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118780215B_ABST
    Figure CN118780215B_ABST
Patent Text Reader

Abstract

The present invention discloses a grid deformation method, device and electronic device, which determines the coordinates of each grid point and the corresponding grid deformation amount based on the initial calculation grid to obtain the initial grid sample point space and traverse it, and establishes a uniform grouping method in the direction of the maximum size of the geometric component. n subintervals, a greedy algorithm is used for sample points in each subinterval to obtain an RBF submodel, all grid points of the initial calculation grid are partitioned according to the subintervals, and the initial grid deformation is obtained based on the corresponding RBF submodel, and then the target grid deformation of each grid point is obtained based on the smoothing function, and the deformed target grid is obtained by adding each target grid deformation to the corresponding grid point coordinates. The present invention obtains the RBF submodel by using a greedy algorithm for sample points in each subinterval obtained by the uniform grouping method, thereby reducing the number of sample points during modeling, having more efficient modeling efficiency, and improving grid deformation efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of grid deformation, and more specifically, to a grid deformation method, device and electronic equipment. Background Art

[0002] The mesh deformation method can greatly improve the efficiency and accuracy of the coupling calculation of CFD (Computational Fluid Dynamics) / CSD (Computational Solid Dynamics), and is a key technology in the fields of aerodynamic optimization design and fluid-solid coupling numerical simulation. In recent years, the mesh deformation method based on RBF (RadialBasis Function) has attracted widespread attention and has been successfully applied in the optimization design of aircraft aerodynamic shape and aeroelastic mechanics simulation calculation based on high-precision CFD analysis.

[0003] The RBF-based mesh deformation method uses the deformation of the deformation control point as a sample to fit the RBF model, and smoothly interpolates and superimposes the deformation of the control point to the flow field calculation grid, and finally realizes mesh deformation. However, the traditional RBF model needs to use all the surface grid points as sample points for modeling and calculation. The amount of model calculation increases significantly nonlinearly with the increase of the interpolation sample point set, which is N. 3 Therefore, when there are too many sample points, the modeling time of the RBF model will be too long, resulting in low mesh deformation efficiency. Summary of the invention

[0004] In view of this, the present invention discloses a mesh deformation method, device and electronic device to reduce the number of sample points during modeling, shorten the modeling time and improve the mesh deformation efficiency.

[0005] A mesh deformation method, comprising:

[0006] Based on the initial calculation grid, the coordinates of each grid point and the corresponding grid deformation are determined to obtain the initial grid sample point space;

[0007] Traversing the initial grid sample point space, and establishing n subintervals in the direction of the maximum size of the geometric component by using a uniform grouping method, wherein the geometric component corresponds to the initial calculation grid;

[0008] A greedy algorithm is used for each sample point in the subinterval to obtain a corresponding RBF submodel;

[0009] Partitioning all grid points of the initial calculation grid according to corresponding subintervals, and obtaining initial grid deformation of grid points in each subinterval based on the RBF submodel corresponding to each subinterval;

[0010] Based on the initial mesh deformation and a predetermined smoothing function, a target mesh deformation of each mesh point in a three-dimensional coordinate direction is obtained;

[0011] The deformation amount of each target grid is added to the corresponding grid point coordinates to obtain a deformed target grid.

[0012] Optionally, determining the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid to obtain the initial grid sample point space includes:

[0013] Determining the initial calculation grid;

[0014] deriving the grid point coordinates of the model deformation region based on the initial calculation grid;

[0015] Determine the mesh deformation amount of each mesh point in the deformation area of ​​the model in the three-dimensional coordinate direction during the mesh deformation process;

[0016] The initial grid sample point space is obtained according to the coordinates of each grid point and the corresponding grid deformation amount.

[0017] Optionally, traversing the initial grid sample point space and establishing n subintervals by using a uniform grouping method in the direction of the maximum size of the geometric component includes:

[0018] Traversing the initial grid sample point space to determine the maximum sample coordinate value and the minimum sample coordinate value in the direction of the maximum size of the geometric component;

[0019] A uniform grouping method is adopted in the maximum size direction to divide all the grid point coordinates into n sub-intervals for nesting;

[0020] Based on the maximum sample coordinate value and the minimum sample coordinate value, each of the sample points is allocated to the corresponding sub-interval.

[0021] Optionally, allocating each of the sample points to the corresponding subinterval based on the maximum sample coordinate value and the minimum sample coordinate value includes:

[0022] The subinterval determination formula shown below is used to determine the subinterval where each sample point in the initial grid sample point space is located;

[0023] ;

[0024] Where imd is the intermediate parameter of the calculation, floor is the rounding function provided by Fortran, X is the sample point partition judgment coordinate, maxp is the maximum value of the sample coordinate, minp is the minimum value of the sample coordinate, n is the number of sub-intervals divided, imo is the sub-interval number where the sample point is located, and else is others;

[0025] The interval position factor is calculated using the interval position factor calculation formula shown below, and the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor;

[0026] ;

[0027] Wherein, ll is the interval position factor.

[0028] Optionally, allocating corresponding sample points to corresponding sub-intervals based on the interval position factor includes:

[0029] All sample points are allocated to corresponding sub-intervals according to the following allocation principle;

[0030] like , then the corresponding sample point is only located in the first subinterval;

[0031] like , then the corresponding sample point is only located in the nth subinterval;

[0032] like , then the corresponding sample point partition judgment coordinates are the same as the maximum size coordinates of the geometric component and are only located in the nth subinterval;

[0033] like , then the corresponding sample point is located in the imo-th subinterval;

[0034] like , then the corresponding sample points are located in both the imo-th subinterval and the imo-1-th subinterval;

[0035] like , the corresponding sample points are located in both the imo-th subinterval and the imo+1-th subinterval.

[0036] Optionally, the adopting a greedy algorithm for the sample points in each subinterval to obtain a corresponding RBF submodel includes:

[0037] A greedy algorithm is used for the sample points in each sub-interval to select multiple deformation control points;

[0038] Determine the weight coefficient of each of the deformation control points in each of the subintervals;

[0039] According to each of the weight coefficients in each of the subintervals, the corresponding RBF sub-model is established.

[0040] Optionally, the process of determining the smoothing function includes:

[0041] Using a greedy algorithm to directly screen all sample points in the initial grid sample point space, to obtain target deformation control points whose maximum interpolation errors are lower than a preset deformation amount of the object surface node, wherein the preset deformation amount is determined based on the maximum deformation amount;

[0042] Establishing an extrapolation model according to each of the target deformation control points;

[0043] The extrapolation model is determined as the smoothing function.

[0044] Optionally, obtaining a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function includes:

[0045] The target grid points in the initial calculation grid whose minimum distance value to the object surface nodes is greater than a first length are processed by the smoothing function to obtain corresponding interpolation results, wherein the first length is a first preset multiple of the average partition length;

[0046] When the minimum distance between the to-be-deformed mesh point and the object surface node is not greater than the first length, determining the initial mesh deformation amount as the target mesh deformation amount;

[0047] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the first length and less than the second length, a linear weighted value of the initial mesh deformation amount and the interpolation result is determined as the target mesh deformation amount, wherein the second length is a second preset multiple of the equal partition length;

[0048] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the second length, the interpolation result is determined as the target mesh deformation amount.

[0049] A grid deformation device, comprising:

[0050] A sample point space determination unit is used to determine the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid to obtain an initial grid sample point space;

[0051] A sub-interval determination unit, configured to traverse the initial grid sample point space and establish n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, wherein the geometric component corresponds to the initial calculation grid;

[0052] A sub-model determination unit, used for applying a greedy algorithm to the sample points in each sub-interval to obtain a corresponding RBF sub-model;

[0053] An initial deformation amount determination unit, used for partitioning all grid points of the initial calculation grid according to corresponding subintervals, and obtaining an initial grid deformation amount of the grid points in each subinterval based on the RBF submodel corresponding to each subinterval;

[0054] A target deformation amount determination unit, used for obtaining a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function;

[0055] The target grid determination unit is used to add the deformation amount of each target grid to the corresponding grid point coordinates to obtain a deformed target grid.

[0056] An electronic device, comprising: a memory and a processor;

[0057] The memory is used to store at least one instruction;

[0058] The processor is used to execute the at least one instruction to implement the above-mentioned mesh deformation method.

[0059] As can be seen from the above technical solutions, the present invention discloses a mesh deformation method, device and electronic device, which determines the coordinates of each mesh point and the corresponding mesh deformation amount based on the initial calculation mesh, obtains the initial mesh sample point space, traverses the initial mesh sample point space, establishes n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, obtains the corresponding RBF sub-model by using a greedy algorithm for the sample points in each sub-interval, partitions all mesh points of the initial calculation mesh according to the corresponding sub-intervals, and obtains the initial mesh deformation amount of the mesh points in each sub-interval based on the RBF sub-model corresponding to each sub-interval, obtains the target mesh deformation amount of each mesh point in the three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function, and obtains the deformed target mesh by adding each target mesh deformation amount to the corresponding mesh point coordinates. The present invention obtains the RBF sub-model by using a greedy algorithm for the sample points in each sub-interval obtained by using the uniform grouping method, thereby reducing the number of sample points during modeling, and has a more efficient modeling efficiency compared to directly using all sample points for RBF modeling, thereby greatly saving computing memory and improving mesh deformation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without paying creative work.

[0061] Figure 1 A flow chart of a grid deformation method disclosed in an embodiment of the present invention;

[0062] Figure 2 A flow chart of a method for determining the coordinates of each grid point and the corresponding grid deformation based on an initial calculation grid to obtain an initial grid sample point space disclosed in an embodiment of the present invention;

[0063] Figure 3 A flow chart of a method disclosed in an embodiment of the present invention for traversing an initial grid sample point space and establishing n subintervals by using a uniform grouping method in the direction of the maximum size of a geometric component;

[0064] Figure 4 A flowchart of a method disclosed in an embodiment of the present invention for obtaining a corresponding RBF sub-model by using a greedy algorithm for sample points in each sub-interval;

[0065] Figure 5 A schematic diagram of a blunt body structure disclosed in an embodiment of the present invention;

[0066] Figure 6 A schematic diagram of typical morphological changes caused by ablation of a blunt body structure during hypersonic atmospheric flight disclosed in an embodiment of the present invention;

[0067] Figure 7 A structural mesh diagram generated by meshing a half of a blunt body and exporting a mesh file disclosed in an embodiment of the present invention;

[0068] Figure 8 A computational mesh diagram of different ablation shapes obtained by deforming the initial mesh of a blunt body disclosed in an embodiment of the present invention;

[0069] Fig. 9 A flow field Mach number distribution diagram disclosed in an embodiment of the present invention;

[0070] Fig.10 A wall heat flux distribution diagram disclosed in an embodiment of the present invention;

[0071] Fig.11 A schematic structural diagram of a grid deformation device disclosed in an embodiment of the present invention;

[0072] Fig.12The present invention is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0073] 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.

[0074] The embodiment of the present invention discloses a mesh deformation method, device and electronic device. Based on the initial calculation mesh, the coordinates of each mesh point and the corresponding mesh deformation amount are determined to obtain the initial mesh sample point space, the initial mesh sample point space is traversed, and n sub-intervals are established in the direction of the maximum size of the geometric component by using a uniform grouping method. The sample points in each sub-interval are subjected to a greedy algorithm to obtain the corresponding RBF sub-model, and all mesh points of the initial calculation mesh are partitioned according to the corresponding sub-intervals. Based on the RBF sub-model corresponding to each sub-interval, the initial mesh deformation amount of the mesh points in each sub-interval is obtained. Based on the initial mesh deformation amount and a predetermined smoothing function, the target mesh deformation amount of each mesh point in the three-dimensional coordinate direction is obtained, and the deformed target mesh is obtained by adding each target mesh deformation amount to the corresponding mesh point coordinate. The present invention obtains the RBF sub-model by using a greedy algorithm for the sample points in each sub-interval obtained by the uniform grouping method, thereby reducing the number of sample points during modeling, and has a more efficient modeling efficiency compared to directly using all sample points for RBF modeling, and the accuracy of the RBF sub-model can be adjusted by adjusting the maximum interpolation error in the greedy algorithm, thereby greatly saving computing memory and improving mesh deformation efficiency.

[0075] See also Figure 1 , a flow chart of a grid deformation method disclosed in an embodiment of the present invention, the method comprising:

[0076] Step S101, determining the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid to obtain the initial grid sample point space;

[0077] In practical applications, the initial computational grid can be established using existing grid generation software.

[0078] Step S102: traverse the initial grid sample point space and establish n sub-intervals in the direction of the maximum size of the geometric component using a uniform grouping method.

[0079] Among them, the geometric components correspond to the initial computational grid.

[0080] In addition, n in this embodiment is a positive integer, and the specific value depends on actual needs.

[0081] This embodiment uses a uniform grouping method to establish n sub-intervals in the direction of the maximum size of the geometric component. In fact, in the direction of the maximum size of the geometric component, the grid point coordinates are divided into n sub-intervals for nesting, that is, the n sub-intervals can also be called n nested sub-intervals.

[0082] The nested parameter calculation formula is shown in formula (1), which is as follows:

[0083] (1);

[0084] In the formula, l t is the length of the sample point coordinate interval in the direction of the maximum size of the geometric component, n is the number of sub-intervals, Δl is the length of the average sub-interval, l n is the length of the overlapping part of two adjacent nested subintervals, and p is the nesting coefficient.

[0085] Then the length d of the nested subinterval at the end point of the geometric component tip The calculation formula is as follows:

[0086] (2);

[0087] The length of the remaining nested subintervals is d mid The calculation formula is as follows:

[0088] (3);

[0089] Step S103, using a greedy algorithm for each sample point in the subinterval to obtain a corresponding RBF submodel;

[0090] A greedy algorithm means that when solving a problem, it always makes the best choice at the moment. In other words, it does not consider the overall optimal solution, and the algorithm obtains a local optimal solution in a certain sense.

[0091] In practical applications, the accuracy of the RBF sub-model can be adjusted by adjusting the maximum interpolation error in the greedy algorithm.

[0092] Step S104, partitioning all grid points of the initial calculation grid according to corresponding subintervals, and obtaining the initial grid deformation of the grid points in each subinterval based on the RBF submodel corresponding to each subinterval;

[0093] In practical applications, the sub-interval judgment formula and the interval position factor calculation formula can be used to partition all grid points of the initial calculation grid according to the corresponding sub-intervals.

[0094] Step S105, obtaining a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function;

[0095] In this embodiment, the specific process of determining the smoothing function is as follows:

[0096] A greedy algorithm is used to directly screen all sample points in the initial grid sample point space to obtain target deformation control points whose maximum interpolation errors are lower than the preset deformation of the object surface nodes.

[0097] Establish an extrapolation model according to each target deformation control point;

[0098] The extrapolation model is determined as a smoothing function.

[0099] The preset deformation amount is determined based on the maximum deformation amount. For example, the preset deformation amount is one thousandth of the maximum deformation amount.

[0100] That is, the maximum interpolation error in the greedy algorithm can meet the accuracy requirement if it is less than one thousandth of the maximum deformation of the object surface nodes.

[0101] Step S106: adding the deformation amount of each target grid to the corresponding grid point coordinates to obtain a deformed target grid.

[0102] Specifically, the target deformation amount of each grid point in the three directions of x, y, and z is added to the corresponding grid point coordinates to obtain a deformed target grid, and the target grid can be output.

[0103] In summary, the present invention discloses a mesh deformation method, which determines the coordinates of each mesh point and the corresponding mesh deformation amount based on the initial calculation mesh, obtains the initial mesh sample point space, traverses the initial mesh sample point space, establishes n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, obtains the corresponding RBF sub-model by using a greedy algorithm for the sample points in each sub-interval, partitions all mesh points of the initial calculation mesh according to the corresponding sub-intervals, and obtains the initial mesh deformation amount of the mesh points in each sub-interval based on the RBF sub-model corresponding to each sub-interval, obtains the target mesh deformation amount of each mesh point in the three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function, and obtains the deformed target mesh by adding each target mesh deformation amount to the corresponding mesh point coordinates. The present invention obtains the RBF sub-model by using a greedy algorithm for the sample points in each sub-interval obtained by using the uniform grouping method, thereby reducing the number of sample points during modeling, and has a more efficient modeling efficiency compared to directly using all sample points for RBF modeling, thereby greatly saving computing memory and improving mesh deformation efficiency.

[0104] To further optimize the above embodiment, see Figure 2 , a flowchart of a method for determining the coordinates of each grid point and the corresponding grid deformation based on an initial calculation grid to obtain an initial grid sample point space is disclosed in an embodiment of the present invention, that is, step S101 may specifically include:

[0105] Step S201, determining an initial calculation grid;

[0106] In practical applications, the existing commonly used grid generation software can be used to establish the initial calculation grid.

[0107] Step S202, deriving the grid point coordinates of the model deformation area based on the initial calculation grid;

[0108] Step S203, determining the mesh deformation amount of each mesh point in the deformation area of ​​the model in the three-dimensional coordinate direction during the mesh deformation process;

[0109] The mesh deformation amount of each mesh point in the three-dimensional coordinate direction during the mesh deformation process refers to: the mesh deformation amount of each mesh point in the three coordinate directions of x, y, and z during the mesh deformation process.

[0110] Step S204: obtaining an initial grid sample point space according to the coordinates of each grid point and the corresponding grid deformation amount.

[0111] To further optimize the above embodiment, see Figure 3 , a flowchart of a method disclosed in an embodiment of the present invention for traversing an initial grid sample point space and establishing n subintervals by using a uniform grouping method in the direction of the maximum size of a geometric component, the method comprising:

[0112] Step S301, traverse the initial grid sample point space to determine the maximum value and minimum value of the sample coordinates in the direction of the maximum size of the geometric component;

[0113] In this embodiment, the initial grid sample point space I 0 By traversing, the maximum value maxp of the sample coordinates and the minimum value minp of the sample coordinates in the direction of the maximum size of the geometric component can be determined.

[0114] Step S302: using a uniform grouping method in the maximum size direction, all grid point coordinates are divided into n sub-intervals for nesting;

[0115] Step S303: Allocate each sample point to a corresponding sub-interval based on the maximum sample coordinate value and the minimum sample coordinate value.

[0116] Specifically, the subinterval determination formula shown in formula (4) is used to determine the subinterval where each sample point in the initial grid sample point space is located;

[0117] (4);

[0118] Where imd is the intermediate parameter of the calculation, floor is the integer function provided by Fortran, X is the sample point partition judgment coordinate (that is, the sample point coordinate in the direction of the maximum size of the geometric component), maxp is the maximum value of the sample coordinate, minp is the minimum value of the sample coordinate, n is the number of sub-intervals divided, imo is the sub-interval number where the sample point is located, and else is others.

[0119] The interval position factor is calculated using the interval position factor calculation formula shown in formula (5), and the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor;

[0120] (5);

[0121] Where ll is the interval location factor.

[0122] Furthermore, the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor, specifically including:

[0123] All sample points are allocated to corresponding sub-intervals according to the following allocation principle;

[0124] like , then the corresponding sample point is only located in the first subinterval;

[0125] like , then the corresponding sample point is only located in the nth subinterval;

[0126] like , then the corresponding sample point partition judgment coordinates are the same as the maximum size coordinates of the geometric component and are only located in the nth subinterval;

[0127] like , then the corresponding sample point is located in the imo-th subinterval;

[0128] like , then the corresponding sample points are located in both the imo-th subinterval and the imo-1-th subinterval;

[0129] like , the corresponding sample points are located in both the imo-th subinterval and the imo+1-th subinterval.

[0130] To further optimize the above embodiment, see Figure 4 , a flowchart of a method for obtaining a corresponding RBF sub-model by using a greedy algorithm for sample points in each sub-interval disclosed in an embodiment of the present invention, the method comprising:

[0131] Step S401, using a greedy algorithm for sample points in each sub-interval to select multiple deformation control points;

[0132] Step S402, determining the weight coefficient of each deformation control point in each sub-interval;

[0133] Step S403: Establish the corresponding RBF sub-model according to each weight coefficient in each sub-interval.

[0134] Specifically, assuming that the number of samples in a subinterval is i, first randomly select K (usually K=3) from the i sample points to form the sub-control point space M 0 , based on M 0 The radial basis interpolation function calculation formula (6) is constructed as follows:

[0135] (6);

[0136] Where F(r) is the interpolation function, ‖rr i ‖ is the distance from the position vector r to ri, ri represents the position of the support point of the i-th radial basis function, ωi is the weight coefficient corresponding to the i-th radial basis function, and φ is Wendlland's C2 compact radial basis function.

[0137] The expression of Wendlland's C2 compact radial basis function is shown in formula (7):

[0138] (7);

[0139] In the formula, , d is the radius of action of the radial basis function, when Mandatory The value of is zero.

[0140] The interpolation condition of the above radial basis function interpolation problem can be further expressed in matrix form as follows:

[0141] The matrix expression of the interpolation condition in the x-direction of radial basis function interpolation is shown in formula (8):

[0142] (8);

[0143] The matrix expression of the interpolation condition in the y direction of radial basis function interpolation is shown in formula (9):

[0144] (9);

[0145] The matrix expression of the radial basis function interpolation z-direction interpolation condition is shown in formula (10):

[0146] (10);

[0147] In formulas (8) to (10), the subscript S represents the surface properties.

[0148] , , They represent the displacement components of the N object surface nodes in the x, y and z directions respectively.

[0149] is the sequence of weight coefficients to be determined;

[0150] The specific form of the matrix Φ is the K-dimensional radial basis function interpolation condition matrix formula, as shown in formula (11):

[0151] (11);

[0152] By solving formulas (8) to (10), we can get W k X , W k Y and W k Z After that, the current K-dimensional interpolation weight coefficient is further extended to N-dimensional, that is, the N-dimensional interpolation weight coefficient formula (12) is as follows:

[0153] (12);

[0154] Obviously, in the above formula are all 0, Substitute it into the equation shown in formula (13) as an approximate solution.

[0155] The interpolation calculation formula of the object surface node displacement is shown in formula (13), which is as follows:

[0156] (13);

[0157] The interpolation error description formula (14) is as follows:

[0158] (14);

[0159] In the formula, is the displacement of N object surface nodes, is the description error of the displacement of each object surface mesh node by the current M-dimensional radial basis function interpolation, is the conditional matrix of N-dimensional radial basis function interpolation, and its specific expression is the N-dimensional radial basis function interpolation conditional matrix formula shown in formula (15), and formula (15) is as follows:

[0160] (15);

[0161] Displacement description error of the current object surface node Traverse to find the maximum displacement description error and the corresponding object surface node, and include the object surface node into the current control point set to form a new deformation control point space.

[0162] Repeat formulas (8) to (15) until the maximum displacement description error of the object surface node in the subinterval meets the accuracy requirement, that is, terminate the traversal, obtain the final approximate solution of the weight coefficient of the deformation control point of the current subinterval, and establish the RBF submodel of the current subinterval. Further, repeat the above steps for each subinterval to establish the RBF submodel of n subintervals.

[0163] In practical applications, the initial mesh deformation of the mesh points in each sub-interval is determined in step S104. When , the displacement of the grid points in the subinterval is obtained by weighting the two adjacent RBF sub-models. The specific weighting relationship is as follows:

[0164] The calculation formula of the model weight coefficient is shown in formula (16), which is as follows:

[0165] (16);

[0166] The calculation formula of the nested regional interpolation result is shown in formula (17). Formula (17) is as follows:

[0167] (17);

[0168] In the formula, is the deformation interpolation result of the overlapped area between the Ith sub-model and the I+1th sub-model, and They are the interpolation results of the grid points in the Ith sub-model and the I+1th sub-model respectively. is the characteristic coordinate value of the left endpoint of the I-th sub-model, is the characteristic coordinate value of the right endpoint of the I+1th sub-model, θ is the weight coefficient of the model, It is the intermediate calculation amount. At the right endpoint of the Ith sub-model, the interpolation result is independently determined by the I+1th sub-model, and at the left endpoint of the I+1th sub-model, the interpolation result is independently determined by the Ith sub-model.

[0169] To further optimize the above embodiment, step S105 may specifically include:

[0170] The target grid points in the initial calculation grid whose minimum distance value to the object surface nodes is greater than a first length are processed by the smoothing function to obtain corresponding interpolation results, wherein the first length is a first preset multiple of the average partition length;

[0171] When the minimum distance between the to-be-deformed mesh point and the object surface node is not greater than the first length, determining the initial mesh deformation amount as the target mesh deformation amount;

[0172] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the first length and less than the second length, a linear weighted value of the initial mesh deformation amount and the interpolation result is determined as the target mesh deformation amount, wherein the second length is a second preset multiple of the equal partition length;

[0173] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the second length, the interpolation result is determined as the target mesh deformation amount.

[0174] In this embodiment, the value of the first preset multiple is determined according to actual needs, such as 0.01, and the value of the second preset multiple is determined according to actual needs, such as 0.1.

[0175] In this embodiment, it is assumed that the initial grid deformation is The interpolation result is expressed as Indicates that the average partition length is It means that the first preset multiple is 0.01 and the second preset multiple is 0.1.

[0176] For the grid points whose minimum distance to the object surface nodes is less than 0.01 times the average partition length, the initial grid deformation obtained based on the RBF sub-model with high fitting accuracy is used. .

[0177] For grid points whose minimum distance to the object surface nodes is greater than 0.1 times the average partition length, the interpolation result based on the smoothing function with low fitting accuracy is used. .

[0178] For the grid points in the transition interval, that is, the grid points whose minimum distance to the object surface nodes is greater than 0.01 times the average partition length and less than 0.1 times the average partition length, the initial grid deformation is used. And the interpolation result The linear weighted value is determined by

[0179] Therefore, the calculation formula of the target mesh deformation is shown in formula (18), which is as follows:

[0180] (18);

[0181] In the formula, l s is the minimum distance between the grid point to be deformed and the object surface node, is the weighting factor, and the calculation formula is shown in formula (19). Formula (19) is as follows:

[0182] (19);

[0183] To facilitate understanding of the mesh deformation process disclosed in the present invention, the present invention also provides a specific embodiment as follows:

[0184] Computational mesh deformation in simulation of typical ablation morphology of a blunt body.

[0185] When an aircraft passes through the atmosphere at high speed, due to the strong compression of the airflow, a severe high temperature and high pressure environment is formed on the frontal surface of the aircraft shell, causing severe thermochemical ablation, resulting in a significant change in the aircraft's aerodynamic shape and overall aerodynamic characteristics. Figure 5 The typical morphological changes of the blunt body structure caused by ablation during hypersonic atmospheric flight ( Figure 6 A series of mesh deformation and aerodynamic simulation calculations are performed. Figure 5 A and B are typical laminar ablation profiles, C and D are typical transitional ablation profiles, and E and F are typical turbulent ablation profiles. Figure 6 In the figure, 1000mm is the diameter of the interface between the ball head and the cone body; 9.5 degrees is the semi-cone angle of the cone body; 182mm is the radius R of the ball head; and 1300mm is the diameter of the tail end face of the cone body.

[0186] The mesh generation software pointwise is used to generate the initial mesh of the blunt body shape. Considering the symmetry of the model, only half of the blunt body is meshed and the mesh file is exported. The generated structural mesh is as follows Figure 7 As shown; at the same time, the grid point coordinates of the deformation area of ​​the blunt body surface are output, the deformation amount of each grid point at different ablation times is determined, and the control file of the deformation program is generated.

[0187] The nested interval division is performed along the flow direction of the blunt body, and a total of 8 regions are divided. The nesting coefficient is set to 0.5, the radius of action of the radial basis function is set to 1.0, the interpolation accuracy of the nested greedy RBF model is set to 0.0001, the interpolation accuracy of the smoothing function is set to 0.01, and the transition interval between the nested greedy RBF function and the smoothing function is set to 0.01~0.1 times the average interval length. Finally, the computational grids of different ablation shapes obtained by deforming the initial model grid of the blunt body are shown in Figure 1. Figure 8 As shown, it includes: mesh of deformed surface 1, mesh of deformed surface 2, mesh of deformed surface 3, mesh of deformed surface 4 and mesh of deformed surface 5. As can be seen from the figure, the meshes obtained by each deformation have high mesh quality, which shows the universality of this method for large-scale deformation of structural meshes.

[0188] For the hypersonic conditions of Mach number Mach=6.04, Reynolds number Re=2.18e7, inflow temperature 62.6K, wall temperature Tw=300K, the series of meshes of the initial shape grid of the blunt body and its deformation at different typical ablation moments are simulated and calculated, and the flow field Mach number distribution (such as Fig. 9 as shown) and the wall heat flux distribution (as shown Fig.10 ). Among them, Fig. 9 The initial model flow field, the deformed profile 1 flow field, the deformed profile 2 flow field, the deformed profile 3 flow field, the deformed profile 4 flow field and the deformed profile 5 flow field are shown in FIG. Fig.10 The initial model heat flow, deformation surface 1 heat flow, deformation surface 2 heat flow, deformation surface 3 heat flow, deformation surface 4 heat flow and deformation surface 5 heat flow are shown in the figure. The initial mesh and the deformed mesh both well reflect the calculated flow field distribution of different blunt body morphologies under this working condition, further proving the robustness and accuracy of the mesh deformation method proposed in the present invention.

[0189] Corresponding to the above method embodiment, the present invention also discloses a grid deformation device.

[0190] See also Fig.11 , a schematic diagram of a grid deformation device disclosed in an embodiment of the present invention, the device may include:

[0191] The sample point space determination unit 501 is used to determine the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid to obtain the initial grid sample point space;

[0192] In practical applications, the initial computational grid can be established using existing grid generation software.

[0193] A sub-interval determination unit 502 is used to traverse the initial grid sample point space and establish n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, wherein the geometric component corresponds to the initial calculation grid;

[0194] Among them, the geometric components correspond to the initial computational grid.

[0195] In addition, n in this embodiment is a positive integer, and the specific value depends on actual needs.

[0196] This embodiment uses a uniform grouping method to establish n sub-intervals in the direction of the maximum size of the geometric component. In fact, in the direction of the maximum size of the geometric component, the grid point coordinates are divided into n sub-intervals for nesting, that is, the n sub-intervals can also be called n nested sub-intervals.

[0197] A sub-model determination unit 503, configured to use a greedy algorithm for each sample point in the sub-interval to obtain a corresponding RBF sub-model;

[0198] An initial deformation amount determination unit 504 is used to partition all grid points of the initial calculation grid according to corresponding subintervals, and obtain an initial grid deformation amount of the grid points in each subinterval based on the RBF submodel corresponding to each subinterval;

[0199] In practical applications, the sub-interval judgment formula and the interval position factor calculation formula can be used to partition all grid points of the initial calculation grid according to the corresponding sub-intervals.

[0200] A target deformation amount determination unit 505 is used to obtain a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function;

[0201] The target grid determination unit 506 is used to add the deformation amount of each target grid to the corresponding grid point coordinates to obtain a deformed target grid.

[0202] Specifically, the target deformation amount of each grid point in the three directions of x, y, and z is added to the corresponding grid point coordinates to obtain a deformed target grid, and the target grid can be output.

[0203] In summary, the present invention discloses a mesh deformation device, which determines the coordinates of each mesh point and the corresponding mesh deformation amount based on the initial calculation mesh, obtains the initial mesh sample point space, traverses the initial mesh sample point space, establishes n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, obtains the corresponding RBF sub-model by using a greedy algorithm for the sample points in each sub-interval, partitions all the mesh points of the initial calculation mesh according to the corresponding sub-intervals, and obtains the initial mesh deformation amount of the mesh points in each sub-interval based on the RBF sub-model corresponding to each sub-interval, obtains the target mesh deformation amount of each mesh point in the three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function, and obtains the deformed target mesh by adding each target mesh deformation amount to the corresponding mesh point coordinates. The present invention obtains the RBF sub-model by using a greedy algorithm for the sample points in each sub-interval obtained by using the uniform grouping method, thereby reducing the number of sample points during modeling, and has a more efficient modeling efficiency compared to directly using all sample points for RBF modeling, thereby greatly saving computing memory and improving mesh deformation efficiency.

[0204] To further optimize the above embodiment, the sample point space determination unit 501 may be specifically used for:

[0205] Determine the initial computational grid;

[0206] deriving the grid point coordinates of the model deformation region based on the initial calculation grid;

[0207] Determine the mesh deformation amount of each mesh point in the deformation area of ​​the model in the three-dimensional coordinate direction during the mesh deformation process;

[0208] The initial grid sample point space is obtained according to the coordinates of each grid point and the corresponding grid deformation amount.

[0209] To further optimize the above embodiment, the sub-interval determining unit 502 may specifically include:

[0210] A traversal subunit, used to traverse the initial grid sample point space to determine the maximum value and the minimum value of the sample coordinates in the direction of the maximum size of the geometric component;

[0211] A uniform grouping subunit, used for adopting a uniform grouping method in the direction of the maximum size to divide all the grid point coordinates into n subintervals for nesting;

[0212] An allocating subunit is used to allocate each of the sample points to the corresponding sub-interval based on the maximum sample coordinate value and the minimum sample coordinate value.

[0213] To further optimize the above embodiment, the allocation subunit may be specifically used for:

[0214] The subinterval determination formula shown below is used to determine the subinterval where each sample point in the initial grid sample point space is located;

[0215] ;

[0216] Where imd is the intermediate parameter of the calculation, floor is the rounding function provided by Fortran, X is the sample point partition judgment coordinate, maxp is the maximum value of the sample coordinate, minp is the minimum value of the sample coordinate, n is the number of sub-intervals divided, imo is the sub-interval number where the sample point is located, and else is others;

[0217] The interval position factor is calculated using the interval position factor calculation formula shown below, and the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor;

[0218] ;

[0219] Wherein, ll is the interval position factor.

[0220] To further optimize the above embodiment, the allocation subunit may also be used for:

[0221] All sample points are allocated to corresponding sub-intervals according to the following allocation principle;

[0222] like , then the corresponding sample point is only located in the first subinterval;

[0223] like , then the corresponding sample point is only located in the nth subinterval;

[0224] like , then the corresponding sample point partition judgment coordinates are the same as the maximum size coordinates of the geometric component and are only located in the nth subinterval;

[0225] like , then the corresponding sample point is located in the imo-th subinterval;

[0226] like , then the corresponding sample points are located in both the imo-th subinterval and the imo-1-th subinterval;

[0227] like , the corresponding sample points are located in both the imo-th subinterval and the imo+1-th subinterval.

[0228] To further optimize the above embodiment, the sub-model determination unit 503 may be specifically used for:

[0229] A greedy algorithm is used for the sample points in each sub-interval to select multiple deformation control points;

[0230] Determine the weight coefficient of each of the deformation control points in each of the subintervals;

[0231] According to each of the weight coefficients in each of the subintervals, the corresponding RBF sub-model is established.

[0232] To further optimize the above embodiment, the mesh deformation may further include:

[0233] The function determination unit is used to determine the smoothing function.

[0234] The function determination unit can be specifically used for:

[0235] Using a greedy algorithm to directly screen all sample points in the initial grid sample point space, to obtain target deformation control points whose maximum interpolation errors are lower than a preset deformation amount of the object surface node, wherein the preset deformation amount is determined based on the maximum deformation amount;

[0236] Establishing an extrapolation model according to each of the target deformation control points;

[0237] The extrapolation model is determined as the smoothing function.

[0238] To further optimize the above embodiment, the target deformation amount determination unit 505 can be specifically used for:

[0239] The target grid points in the initial calculation grid whose minimum distance value to the object surface nodes is greater than a first length are processed by the smoothing function to obtain corresponding interpolation results, wherein the first length is a first preset multiple of the average partition length;

[0240] When the minimum distance between the to-be-deformed mesh point and the object surface node is not greater than the first length, determining the initial mesh deformation amount as the target mesh deformation amount;

[0241] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the first length and less than the second length, a linear weighted value of the initial mesh deformation amount and the interpolation result is determined as the target mesh deformation amount, wherein the second length is a second preset multiple of the equal partition length;

[0242] When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the second length, the interpolation result is determined as the target mesh deformation amount.

[0243] It should be noted that for the specific working principles of each component in the device embodiment, please refer to the corresponding part of the method embodiment, which will not be repeated here.

[0244] Corresponding to the above embodiment, Fig.12 As shown, the present invention also provides an electronic device, which may include: a processor 1 and a memory 2;

[0245] The processor 1 and the memory 2 communicate with each other via a communication bus 3;

[0246] Processor 1, configured to execute at least one instruction;

[0247] Memory 2, used to store at least one instruction;

[0248] The processor 1 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0249] The memory 2 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0250] The processor executes at least one instruction to implement the steps shown in the grid deformation method embodiment.

[0251] In summary, the present invention discloses an electronic device, which determines the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid, obtains the initial grid sample point space, traverses the initial grid sample point space, establishes n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, obtains the corresponding RBF sub-model by using a greedy algorithm for the sample points in each sub-interval, partitions all the grid points of the initial calculation grid according to the corresponding sub-intervals, and obtains the initial grid deformation of the grid points in each sub-interval based on the RBF sub-model corresponding to each sub-interval, obtains the target grid deformation of each grid point in the three-dimensional coordinate direction based on the initial grid deformation and a predetermined smoothing function, and obtains the deformed target grid by adding each target grid deformation to the corresponding grid point coordinates. The present invention obtains the RBF sub-model by using a greedy algorithm for the sample points in each sub-interval obtained by using the uniform grouping method, thereby reducing the number of sample points during modeling, and has a more efficient modeling efficiency compared to directly using all sample points for RBF modeling, thereby greatly saving computing memory and improving grid deformation efficiency.

[0252] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0253] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0254] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A mesh deformation method, characterized in that: include: Based on the initial calculation grid, the grid point coordinates and corresponding grid deformation amount of each aircraft blunt body deformation area are determined to obtain the initial grid sample point space of the aircraft blunt body shape; Traversing the initial grid sample point space, and establishing n subintervals in the direction of the maximum size of the geometric component by using a uniform grouping method, wherein the geometric component corresponds to the initial calculation grid; A greedy algorithm is used for each sample point in the subinterval to obtain a corresponding RBF submodel; Partitioning all grid points of the initial calculation grid according to corresponding subintervals, and obtaining initial grid deformation of grid points in each subinterval based on the RBF submodel corresponding to each subinterval; Based on the initial mesh deformation and a predetermined smoothing function, a target mesh deformation of each mesh point in a three-dimensional coordinate direction is obtained; Adding each target grid deformation amount to the corresponding grid point coordinates to obtain a deformed target grid; The traversal of the initial grid sample point space and the establishment of n sub-intervals by a uniform grouping method in the direction of the maximum size of the geometric component include: Traversing the initial grid sample point space to determine the maximum sample coordinate value and the minimum sample coordinate value in the direction of the maximum size of the geometric component; A uniform grouping method is adopted in the maximum size direction to divide all the grid point coordinates into n sub-intervals for nesting; Allocating each of the sample points to the corresponding subinterval based on the maximum sample coordinate value and the minimum sample coordinate value, including: The subinterval determination formula shown below is used to determine the subinterval where each sample point in the initial grid sample point space is located; Where imd is the intermediate parameter of the calculation, floor is the rounding function provided by Fortran, X is the sample point partition judgment coordinate, maxp is the maximum value of the sample coordinate, minp is the minimum value of the sample coordinate, n is the number of sub-intervals divided, imo is the sub-interval number where the sample point is located, and else is others; The interval position factor is calculated using the interval position factor calculation formula shown below, and the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor; Wherein, ll is the interval position factor; The allocating corresponding sample points to corresponding sub-intervals based on the interval position factor includes: All sample points are allocated to corresponding sub-intervals according to the following allocation principle; If imo=1, ll<1-0.5p, the corresponding sample point is only located in the first subinterval; If imo=n, ll>0.5p, the corresponding sample point is only located in the nth subinterval; If imo=n, ll=0, the corresponding sample point partition judgment coordinates are the same as the maximum size coordinates of the geometric component and are only located in the nth subinterval; If 2≤imo≤n-1, 0.5p<ll<1-0.5p, then the corresponding sample point is located in the imo-th subinterval; If 2≤imo≤n, ll≤0.5p, the corresponding sample point is located in both the imo-th subinterval and the imo-1-th subinterval; If 1≤imo≤n-1, ll≥1-0.5p, the corresponding sample point is located in both the imo-th subinterval and the imo+1-th subinterval; The step of obtaining a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function includes: The target grid points in the initial calculation grid whose minimum distance value to the object surface nodes is greater than a first length are processed by the smoothing function to obtain corresponding interpolation results, wherein the first length is a first preset multiple of the average partition length; When the minimum distance between the to-be-deformed mesh point and the object surface node is not greater than the first length, determining the initial mesh deformation amount as the target mesh deformation amount; When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the first length and less than the second length, a linear weighted value of the initial mesh deformation amount and the interpolation result is determined as the target mesh deformation amount, wherein the second length is a second preset multiple of the average partition length; When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the second length, the interpolation result is determined as the target mesh deformation amount.

2. The grid deformation method according to claim 1, characterized in that: Determining the coordinates of each grid point and the corresponding grid deformation based on the initial calculation grid to obtain the initial grid sample point space includes: Determining the initial calculation grid; deriving the grid point coordinates of the model deformation region based on the initial calculation grid; Determine the mesh deformation amount of each mesh point in the deformation area of ​​the model in the three-dimensional coordinate direction during the mesh deformation process; The initial grid sample point space is obtained according to the coordinates of each grid point and the corresponding grid deformation amount.

3. The grid deformation method according to claim 1, characterized in that: The method of using a greedy algorithm for each sample point in the subinterval to obtain a corresponding RBF submodel includes: A greedy algorithm is used for the sample points in each sub-interval to select multiple deformation control points; Determine the weight coefficient of each of the deformation control points in each of the subintervals; According to each of the weight coefficients in each of the subintervals, the corresponding RBF sub-model is established.

4. The grid deformation method according to claim 1, characterized in that: The process of determining the smoothing function includes: Directly screening all sample points in the initial grid sample point space using a greedy algorithm to obtain target deformation control points whose maximum interpolation errors are lower than a preset deformation amount of the object surface node, wherein the preset deformation amount is determined based on the maximum deformation amount; Establishing an extrapolation model according to each of the target deformation control points; The extrapolation model is determined as the smoothing function.

5. A grid deformation device, characterized in that: include: A sample point space determination unit is used to determine the grid point coordinates and corresponding grid deformation amounts of each aircraft blunt body surface deformation area based on the initial calculation grid, and obtain the initial grid sample point space of the aircraft blunt body shape; A sub-interval determination unit, configured to traverse the initial grid sample point space and establish n sub-intervals in the direction of the maximum size of the geometric component by using a uniform grouping method, wherein the geometric component corresponds to the initial calculation grid; A sub-model determination unit, used for applying a greedy algorithm to the sample points in each sub-interval to obtain a corresponding RBF sub-model; An initial deformation amount determination unit, used for partitioning all grid points of the initial calculation grid according to corresponding subintervals, and obtaining an initial grid deformation amount of the grid points in each subinterval based on the RBF submodel corresponding to each subinterval; A target deformation amount determination unit, used for obtaining a target mesh deformation amount of each mesh point in a three-dimensional coordinate direction based on the initial mesh deformation amount and a predetermined smoothing function; A target grid determination unit, used for adding each target grid deformation amount to the corresponding grid point coordinates to obtain a deformed target grid; Wherein, the sub-interval determining unit includes: A traversal subunit, used to traverse the initial grid sample point space to determine the maximum value and the minimum value of the sample coordinates in the direction of the maximum size of the geometric component; A uniform grouping subunit, used for adopting a uniform grouping method in the direction of the maximum size to divide all the grid point coordinates into n subintervals for nesting; an allocating subunit, configured to allocate each of the sample points to the corresponding subinterval based on the maximum sample coordinate value and the minimum sample coordinate value; The allocation subunit is specifically used for: The subinterval determination formula shown below is used to determine the subinterval where each sample point in the initial grid sample point space is located; Where imd is the intermediate parameter of the calculation, floor is the rounding function provided by Fortran, X is the sample point partition judgment coordinate, maxp is the maximum value of the sample coordinate, minp is the minimum value of the sample coordinate, n is the number of sub-intervals divided, imo is the sub-interval number where the sample point is located, and else is others; The interval position factor is calculated using the interval position factor calculation formula shown below, and the corresponding sample points are allocated to the corresponding sub-intervals based on the interval position factor; Wherein, ll is the interval position factor; The allocation subunit is further specifically used for: All sample points are allocated to corresponding sub-intervals according to the following allocation principle; If imo=1, ll<0.5p, the corresponding sample point is only located in the first subinterval; If imo=n, ll>0.5p, the corresponding sample point is only located in the nth subinterval; If imo=n, ll=0, the corresponding sample point partition judgment coordinates are the same as the maximum size coordinates of the geometric component and are only located in the nth subinterval; If 2≤imo≤n-1, 0.5p<ll<1-0.5p, then the corresponding sample point is located in the imo-th subinterval; If 2≤imo≤n, ll≤0.5p, the corresponding sample point is located in both the imo-th subinterval and the imo-1-th subinterval; If 1≤imo≤n-1, ll≥1-0.5p, the corresponding sample point is located in both the imo-th subinterval and the imo+1-th subinterval; The target deformation amount determination unit is specifically used for: The target grid points in the initial calculation grid whose minimum distance value to the object surface nodes is greater than a first length are processed by the smoothing function to obtain corresponding interpolation results, wherein the first length is a first preset multiple of the average partition length; When the minimum distance between the to-be-deformed mesh point and the object surface node is not greater than the first length, determining the initial mesh deformation amount as the target mesh deformation amount; When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the first length and less than the second length, a linear weighted value of the initial mesh deformation amount and the interpolation result is determined as the target mesh deformation amount, wherein the second length is a second preset multiple of the average partition length; When the minimum distance between the to-be-deformed mesh point and the object surface node is greater than the second length, the interpolation result is determined as the target mesh deformation amount.

6. An electronic device, characterized in that: The electronic device comprises: a memory and a processor; The memory is used to store at least one instruction; The processor is used to execute the at least one instruction to implement the mesh deformation method according to any one of claims 1 to 4.

Citation Information

Patent Citations

  • Space grid deformation method based on parallelized radial basis function

    CN107391871A

  • Improved radial basis function grid deformation method based on modal space and greedy algorithm

    CN116796811A

  • Method and device for determining grid for fluid calculation

    CN118114580A

  • System and method for diffusing curvature

    US20050243102A1