A parameter design method, system, medium and device for an aircraft

Abstract

The present invention discloses a method, system, medium and device for parameter design of an aircraft, relating to the technical field of aircraft design. An initial shape design space of the aircraft is constructed; according to the initial shape design space, iterative optimization is performed on the design parameters of the aircraft to determine the optimal solution of the initial shape design space; in the case that the initial shape design space does not meet the convergence condition, the effectiveness index of each candidate point in the initial shape design space is calculated, and the initial shape design space is refined according to the candidate point with the maximum effectiveness index to update the initial shape design space; the updated initial shape design space is used as the new initial shape design space, and iterative optimization is continued on the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition, and the optimal solution of the shape design space that meets the convergence condition is determined as the parameter design strategy of the aircraft. This method can improve the design accuracy of the aircraft.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft design, and particularly relates to a method, system, medium, and device for parameter design of an aircraft. Background Art

[0002] Currently, the aerodynamic shape optimization design is one of the core research topics in the aerospace field. Its purpose is to improve the performance of the aircraft, such as lift, drag, and stability, by adjusting the parameters of the airfoil or other aerodynamic shapes.

[0003] With the continuous progress of high-performance computing hardware and advanced modeling techniques, the accuracy and authenticity of aircraft design optimization have been significantly improved, and the enhanced computing power provides more powerful support for optimization design. However, the engineering design optimization still faces many challenges due to the difficult problem of efficiently exploring complex design spaces and ensuring global optimal solutions.

[0004] Therefore, there is an urgent need for a method that can improve the design accuracy of aircraft. Summary of the Invention

[0005] Based on this, in view of the above technical problems, it is necessary to provide a method, system, medium, and device for parameter design of an aircraft, which can improve the design accuracy of the aircraft.

[0006] The present invention adopts the following technical solutions:

[0007] The present invention provides a method for parameter design of an aircraft, including:

[0008] According to the reference shape of the aircraft, construct an initial shape design space of the aircraft; the initial shape design space includes the design parameters of multiple candidate points on the aircraft;

[0009] According to the initial shape design space, perform iterative optimization on the design parameters of the aircraft to determine the optimal solution of the initial shape design space;

[0010] In the case where the initial shape design space does not meet the convergence condition, calculate the effectiveness index of each candidate point in the initial shape design space, and refine the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space; refining the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points;

[0011] Take the updated initial shape design space as the new initial shape design space, and continue to perform iterative optimization on the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition, and determine the shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

[0012] Optionally, calculate the effectiveness index of each candidate point in the initial shape design space, including:

[0013] Obtain the Spearman rank correlation coefficient between the optimization objective of the aircraft and each candidate point in the initial shape design space, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point;

[0014] Determine the effectiveness index of each candidate point according to the Spearman rank correlation coefficient between the optimization objective of the aircraft and each candidate point, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point.

[0015] Optionally, refine the initial shape design space according to the candidate point with the maximum effectiveness index to update the initial shape design space, including:

[0016] Obtain the design parameters of two new candidate points according to the index and design parameters of the candidate point with the maximum effectiveness index;

[0017] Determine the updated initial shape design space according to the design parameters of the two new candidate points and the design parameters of the candidate points in the initial shape design space other than the candidate point with the maximum effectiveness index.

[0018] Optionally, the index of the candidate point with the maximum effectiveness index , and the calculation formula for the design parameters of the updated initial shape design space is:

[0019] ;

[0020] where the initial shape design space is , and the updated initial shape design space is , represents the number of refinement rounds, represents the design parameters of the first candidate point in the th round of refinement, represents the design parameters of the first candidate point in the th round of refinement, represents the design parameters of the second candidate point in the th round of refinement, represents the design parameters of the second candidate point in the th round of refinement, represents the design parameters of the th candidate point in the th round of refinement, represents the design parameters of the th candidate point in the th round of refinement, represents the th candidate point in the Design parameters of the candidate points Indicates the Round of refinement, the Design parameters of the th candidate point; is the number of candidate points in the initial outer shape design space; and are the design parameters of two new candidate points;

[0021] Index of the candidate point with the maximum effectiveness index Satisfies Then the calculation formula for the design parameters of the updated initial outer shape design space is:

[0022] ;

[0023] Among them, Indicates the Round of refinement, the Design parameters of the th candidate point; Indicates the Round of refinement, the th candidate point; Indicates the Round of refinement, the Design parameters of the th candidate point; Indicates the Round of refinement, the Design parameters of the th candidate point; Round of refinement, the Design parameters of the Round of refinement, the th candidate point; Indicates the Round of refinement, the Design parameters of the th candidate point; and

[0024] Index of the candidate point with the maximum effectiveness index Satisfies

[0025] ;

[0026] Among them, Indicates the Round of refinement, the Design parameters of the th candidate point; The design parameters of the th candidate point during the round of refinement, indicating the design parameters of the th candidate point during the round of refinement, indicating the design parameters of the th candidate point during the round of refinement, and are the design parameters of two new candidate points.

[0027] Optionally, the method for determining whether the initial outer shape design space meets the convergence condition includes:

[0028] Obtain the optimized outer shape corresponding to the optimal solution of the aircraft in the initial outer shape design space and the initial configuration of the initial outer shape design space; the initial configuration is the reference outer shape of the aircraft corresponding to the initial outer shape design space before iterative optimization;

[0029] If the optimized outer shape is the same as the initial configuration, it is determined that the initial outer shape design space meets the convergence condition;

[0030] If the optimized outer shape is different from the initial configuration, it is determined that the initial outer shape design space does not meet the convergence condition.

[0031] Optionally, based on the initial outer shape design space, iterative optimization is performed on the design parameters of the aircraft to determine the optimal solution of the initial outer shape design space, including:

[0032] Extract a geometric sample set in the initial outer shape design space by using the Latin hypercube sampling method;

[0033] Calculate the aerodynamic characteristics of the geometric sample set to obtain a geometric aerodynamic sample library;

[0034] Import the geometric aerodynamic sample library into the Kriging model to obtain a surrogate model;

[0035] Based on the geometric and aerodynamic constraints of the aircraft, use the genetic algorithm to perform iterative optimization on the design parameters of the aircraft according to the surrogate model until the optimal solution of the initial outer shape design space is found.

[0036] Optionally, based on the reference outer shape of the aircraft, construct the initial outer shape design space of the aircraft, including:

[0037] Characterize the reference outer shape of the aircraft by using the Hicks-Henne parameterization method to construct the initial outer shape design space.

[0038] The present invention provides a parameter design system for an aircraft, including:

[0039] An initialization module for constructing an initial shape design space of an aircraft according to the reference shape of the aircraft; the initial shape design space includes design parameters of multiple candidate points on the aircraft;

[0040] An optimization module for iteratively optimizing the design parameters of the aircraft according to the initial shape design space to determine the optimal solution of the initial shape design space;

[0041] A refinement module for calculating the effectiveness index of each candidate point in the initial shape design space and refining the initial shape design space according to the candidate point with the largest effectiveness index in the case where the initial shape design space does not meet the convergence condition, so as to update the initial shape design space; the refinement of the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points;

[0042] An iteration module for using the updated initial shape design space as a new initial shape design space to continue iteratively optimizing the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition, and determining the shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

[0043] The present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the parameter design method of the above-mentioned aircraft is implemented.

[0044] The present invention provides a computer device including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the parameter design method of the above-mentioned aircraft is implemented.

[0045] The above at least one technical solution adopted by the present invention can achieve the following beneficial effects:

[0046] In the present invention, during the optimization process, the effectiveness index of each candidate point is calculated, and the design space is refined according to the candidate point with the largest effectiveness index. In this way, the design space is gradually refined by continuously inserting candidate points, effectively reducing the dependence of the optimization result on the initial shape design space, making the optimization design time of the present invention basically the same as that of the traditional fixed parameterization method, but the optimization result is significantly improved, thereby improving the design accuracy of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0048] Figure 1Schematic diagram of the parameter design method for an aircraft provided by the present invention;

[0049] Figure 2 Schematic diagram of another parameter design method for an aircraft provided by the present invention;

[0050] Figure 3 Schematic diagram of the variation relationship of an improved 5th-order Hicks-Henne type function provided by the present invention;

[0051] Figure 4 Schematic diagram of the process for determining the initial shape design space in an optimizer provided by the present invention;

[0052] Figure 5 Schematic diagram of the position distribution of multiple candidate points on the geometric shape of an aircraft in the initial shape design space provided by the present invention;

[0053] Figure 6 Schematic diagram of a RAE 2822 airfoil grid provided by the present invention;

[0054] Figure 7 Schematic diagram of the comparison of the result pressure coefficients of three groups of grids provided by the present invention;

[0055] Figure 8 Schematic diagram of the effectiveness index of each candidate point provided by the present invention;

[0056] Figure 9 Schematic diagram of the effect of inserting candidate points in four rounds of refinement provided by the present invention;

[0057] Figure 10 Schematic diagram of the next-stage optimization iteration curve caused by triggering space refinement at different times provided by the present invention;

[0058] Figure 11 Schematic diagram of the convergence history of the fixed and adaptive parameterization optimization process provided by the present invention;

[0059] Figure 12 Schematic diagram of the comparison of the geometric and pressure distributions of the optimization results of the fixed and adaptive parameterization methods provided by the present invention;

[0060] Figure 13 Schematic diagram of the optimization result and refinement process of each round of the adaptive parameterization method provided by the present invention;

[0061] Figure 14 Schematic diagram of the optimization result and refinement process of each round of the adaptive parameterization method starting from NACA 0012 provided by the present invention;

[0062] Figure 15Schematic diagram of the comparison of the geometric shapes and pressure distributions of the two initial shape optimization results provided by the present invention;

[0063] Figure 16 Schematic diagram of a parameter design system for an aircraft provided by the present invention;

[0064] Figure 17 Schematic diagram of a computer device for implementing a method for parameter design of an aircraft provided by the present invention. Detailed implementation manners

[0065] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0066] Currently, in the design process of aircraft, depending on whether gradient information is dominant, aerodynamic optimization design is divided into two technical routes: namely, optimization design dominated by the adjoint method and optimization design dominated by the evolutionary method. The adjoint method can handle a large number of design variables. However, since gradient-based algorithms are weak in dealing with multi-extremum problems, in practical applications, they often face the dilemma of local optimal solutions and cannot fully explore the potential of the design space. To achieve the global optimization effect, heuristic algorithms that imitate biological evolution or swarm behavior phenomena in nature to improve the global optimization characteristics have attracted people's attention. For example, common genetic algorithms and particle swarm optimization algorithms. Further, an efficient optimization system formed by combining heuristic algorithms with surrogate models can quickly find the global optimal solution.

[0067] Existing research has shown that in aerodynamic optimization, common parameterization methods such as airfoil parameterization CST (Class Shape Transformation), Hicks-Henne type functions, and B-splines, although having good generality and flexible shape representation capabilities, also have some defects. On the one hand, the existing parameterization methods largely depend on the selection of the initial shape and the construction of the design space, which requires high experience of the designer and pre-setting. Especially in unconventional configurations with limited design experience, the pre-determined geometric parameterization is likely to limit the ability of the optimizer to find the best geometric shape, resulting in unstable optimization results. On the other hand, these methods usually require a large number of design variables to accurately describe the airfoil shape, thus significantly increasing the computational complexity and time cost of the optimization process.

[0068] Based on this, the present invention provides a parameter design method, system, medium and device for an aircraft. By constructing an initial shape design space and introducing a candidate point insertion technique, the design space is continuously refined, and the continuous insertion of candidate points is automatically realized during the optimization process, greatly reducing the dependence of the optimization result on the initial shape design space and effectively improving the efficiency and engineering practicability of aircraft optimization design.

[0069] In the present invention, the server can be used as the execution entity. The server mentioned in the present invention can be a server set up on the service platform or a device such as a desktop computer or a laptop computer that can execute the solution of the present invention.

[0070] The following will, in conjunction with the accompanying drawings, detail the technical solutions provided by the embodiments of the present invention.

[0071] Figure 1 It is a schematic flow chart of a parameter design method for an aircraft in the present invention, specifically including the following steps:

[0072] S101. According to the reference shape of the aircraft, construct the initial shape design space of the aircraft; the initial shape design space includes the design parameters of multiple candidate points on the aircraft.

[0073] In one embodiment, according to the reference shape of the aircraft, constructing the initial shape design space of the aircraft includes: using the Hicks-Henne parameterization method to characterize the reference shape of the aircraft and constructing the initial shape design space.

[0074] Among them, the reference shape can be designed independently or an existing aircraft model can be used; using the Hicks-Henne parameterization method to characterize the reference shape, the expression of the characterization function is as follows:

[0075] (1);

[0076] Among them, and respectively represent the upper and lower surface functions of the airfoil, and respectively represent the upper and lower surface functions of the reference airfoil, is the chord length of the airfoil, and its value range is generally . is the shape function, is the weight coefficient of the shape function (i.e., the candidate point), is the number of shape functions (i.e., candidate points), which is determined according to the design requirements. The candidate points are sorted in sequence from the leading edge to the trailing edge, is the index of the candidate point. is used to control the slope change of the shape function, and its value range is generally , which can control the change at the trailing edge of the airfoil geometric shape, is the decay rate coefficient of the control function, and the additional term is used to control the decay at the trailing edge of the airfoil geometry. Assume that when = 2, 3, 4, 5, 6, 7, they are 0.15, 0.30, 0.45, 0.60, 0.75, 0.90 respectively.

[0077] The Hicks-Henne parameterization method realizes the deformation of the aircraft baseline shape by selecting appropriate candidate point positions and adjusting the weight coefficients of the shape functions. That is, the design parameters of the candidate points are the positions of the candidate points.

[0078] S102, according to the initial shape design space, iteratively optimize the design parameters of the aircraft to determine the optimal solution of the initial shape design space.

[0079] In one embodiment, according to the initial shape design space, iteratively optimize the design parameters of the aircraft to determine the optimal solution of the initial shape design space, including: extracting a geometric sample set in the initial shape design space by using the Latin hypercube sampling method; calculating the aerodynamic characteristics of the geometric sample set to obtain a geometric-aerodynamic sample library; importing the geometric-aerodynamic sample library into the Kriging model to obtain a surrogate model; using the genetic algorithm to iteratively optimize the design parameters of the aircraft according to the surrogate model until the optimal solution of the initial shape design space is found.

[0080] Among them, the construction process of the surrogate model includes: preprocessing each sample in the geometric sample set to form a watertight geometric model, and using the script of the preprocessing software (The Integrated Computer Engineering and Manufacturing code for Computational Fluid Dynamics, ICEMCFD) to realize the generation of the geometric model into a computational grid; performing accurate numerical simulation of computational fluid dynamics (CFD) using the Fluent commercial software to obtain high-fidelity aerodynamic characteristics; importing the geometric samples and aerodynamic characteristics (geometric-aerodynamic sample library) into the Kriging model to construct an initial surrogate model.

[0081] Among them, the surrogate model can be updated by adopting an improved expected point addition criterion to improve the global accuracy of the surrogate model.

[0082] Through selection, crossover, and mutation of the genetic algorithm, iterative optimization is performed on the design parameters of the aircraft. The geometric and aerodynamic characteristics of the found design parameters are evaluated, and it is checked whether the optimization tends to converge. The design parameters of the last 30 iterations can be left unchanged as a way of evaluation. If the optimization has not converged, the design sample is filled into the existing sample set to correct the surrogate model. If the optimization converges, the optimal design parameters are output as the optimal solution.

[0083] S103. When the initial shape design space does not meet the convergence condition, calculate the effectiveness index of each candidate point in the initial shape design space, and refine the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space; the refinement of the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points.

[0084] Optionally, the method for determining whether the initial shape design space meets the convergence condition includes: obtaining the optimized shape corresponding to the optimal solution of the aircraft in the initial shape design space and the initial configuration of the initial shape design space; the initial configuration is the reference shape of the aircraft corresponding to the initial shape design space before iterative optimization; if the optimized shape is the same as the initial configuration, it is determined that the initial shape design space meets the convergence condition; if the optimized shape is different from the initial configuration, it is determined that the initial shape design space does not meet the convergence condition.

[0085] Specifically, according to the reference shape of the aircraft, the initial shape design space is designed. The initial shape design space includes the design parameters of 6 candidate points. The optimal solution of the initial shape design space is obtained through iterative optimization for the 6 candidate points, and the optimized shape corresponding to the 6 candidate points is determined accordingly. It is judged whether the convergence condition is met (whether the optimized shape corresponding to the initial shape design space with 6 candidate points is the same as the reference shape. If it is the same, it is determined that the convergence condition is met). If the convergence condition is not met, the initial shape design space is refined and updated. At this time, the initial shape design space has 7 candidate points. After finding the optimal solution of this initial shape design space, it is judged whether the convergence condition is met. The method for judging whether the convergence condition is met at this time is: judging whether the optimized shape corresponding to the optimal solution of the initial shape design space with 7 candidate points is the same as the initial configuration of the initial shape design space with 7 candidate points; the initial configuration is the reference shape of the aircraft corresponding to the initial shape design space with 7 candidate points before iterative optimization, that is, the optimized shape corresponding to the optimal solution of the initial shape design space with 6 candidate points. And so on, when judging whether the initial shape design space meets the convergence condition, the optimized shape corresponding to the optimal solution of the current initial shape design space is compared with the reference shape of the aircraft corresponding to the current initial shape design space before iterative optimization to judge whether they are the same.

[0086] If the initial shape design space satisfies the convergence condition, the optimal solution of the initial shape design space is used as the global optimal solution, and the optimal solution of the initial shape design space that satisfies the convergence condition is determined as the parameter design strategy of the aircraft.

[0087] In the case where the initial shape design space does not satisfy the convergence condition, the initial shape design space refinement is triggered, that is, the effectiveness index considering constraints is calculated for each candidate point, and nodes are inserted at the position of the best-performing candidate point to achieve adaptive shape design space refinement, thereby ensuring a significant improvement in the optimization efficiency while introducing a small number of design variables.

[0088] That is, the specific time to trigger the initial shape design space refinement is to find the optimal solution in the current shape design space. There are two main reasons: First, only when the existing shape design space can no longer improve the geometric and aerodynamic characteristics of the aircraft is it necessary to add additional candidate points. Second, the effectiveness index composed of the Spearman rank correlation coefficient is accurate only when the optimization reaches full convergence.

[0089] Among them, calculating the effectiveness index of each candidate point in the initial shape design space includes: obtaining the Spearman rank correlation coefficient between the optimization objective of the aircraft and each candidate point in the initial shape design space, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point; determining the effectiveness index of each candidate point according to the Spearman rank correlation coefficient between the optimization objective of the aircraft and each candidate point, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point.

[0090] Specifically, the Spearman rank correlation coefficient is used to represent the correlation between two variables, and the calculation formula of the Spearman rank correlation coefficient is as follows:

[0091] (2);

[0092] Among them, is the variable and between the Spearman rank correlation coefficient, , , is the number of samples, , , , , is rank, is rank.

[0093] Therefore, the Spearman rank correlation coefficient between the optimization objective of the aircraft and each candidate point in the initial shape design space, as well as the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point, can be calculated using formula (2). For example, when calculating the Spearman rank correlation coefficient between the optimization objective of the aircraft and any candidate point in the initial shape design space, is used as the design parameter of the candidate point, is the number of samples in the initial shape design space, is the optimization objective. The optimization objective can be the lift coefficient , the drag coefficient , etc., and the design constraints can be the area , the moment constraint , etc.

[0094] The calculation formula for the effectiveness index of the candidate point is:

[0095] (3);

[0096] Among them, represents the effectiveness index of the th candidate point , is the Spearman rank correlation coefficient between the optimization objective of the aircraft and the candidate point , is the Spearman rank correlation coefficient between the design constraints of the aircraft and the candidate point , is the number of design constraints, is the weight coefficient of the th design constraint, that is, if there are multiple design constraints, the Spearman rank correlation coefficient between any design constraint and each candidate point is calculated. When it is a lower-bound design constraint , and when it is an upper-bound design constraint .

[0097] In one embodiment, the initial shape design space is refined according to the candidate point with the largest effectiveness index to update the initial shape design space, including: obtaining the design parameters of two new candidate points according to the index and design parameters of the candidate point with the largest effectiveness index; determining the updated initial shape design space according to the design parameters of the two new candidate points and the design parameters of the candidate points in the initial shape design space other than the candidate point with the largest effectiveness index.

[0098] The refinement of the initial shape design space refers to introducing new candidate points by inserting nodes to refine the initial shape design space, that is, increasing the dimension of the design space, aiming to significantly improve the optimization efficiency while introducing a small number of candidate points.

[0099] Optionally, the index of the candidate point with the largest validity index , the calculation formula for the design parameters of the updated initial shape design space is:

[0100] (4);

[0101] Wherein, the initial shape design space is , the updated initial shape design space is , represents the number of refinement rounds, represents the design parameters of the first candidate point in the th refinement round, represents the design parameters of the first candidate point in the th refinement round, represents the design parameters of the second candidate point in the th refinement round, represents the design parameters of the second candidate point in the th refinement round, represents the design parameters of the th candidate point in the th refinement round, represents the design parameters of the th candidate point in the th refinement round, represents the design parameters of the th candidate point in the th refinement round, represents the design parameters of the th candidate point in the th refinement round, , is the number of candidate points in the initial shape design space; and are the design parameters of two new candidate points.

[0102] The index of the candidate point with the largest validity index satisfies , then the calculation formula for the design parameters of the updated initial shape design space is:

[0103] (5);

[0104] Wherein, represents the design parameters of the th candidate point in the th refinement round, represents the design parameters of the th candidate point in the th refinement round, , Represents the design parameters of the th candidate point during the th round of refinement, represents the design parameters of the th candidate point during the th round of refinement, represents the design parameters of the th candidate point during the th round of refinement, represents the design parameters of the th candidate point during the th round of refinement; and

[0105] are the design parameters of two new candidate points. If the index of the candidate point with the maximum effectiveness index

[0106] is (6);

[0107] where represents the design parameters of the th candidate point during the th round of refinement, represents the design parameters of the th candidate point during the th round of refinement, represents the design parameters of the th candidate point during the th round of refinement, and are the design parameters of two new candidate points.

[0108] S104. Take the updated initial shape design space as the new initial shape design space, and continue to iteratively optimize the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition. Determine the shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

[0109] Take the updated initial shape design space as the new initial shape design space, continue to iteratively optimize the design parameters of the aircraft, re-determine the optimal solution of the initial shape design space. When the initial shape design space does not meet the convergence condition, calculate the effectiveness index of each candidate point in the initial shape design space, and refine the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space. Then, take the updated initial shape design space as the new initial shape design space again, and continue to iteratively optimize the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition, and determine the shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

[0110] In one embodiment, the present invention also provides a parameter design method for an aircraft, as Figure 2 shown, this embodiment includes the following steps:

[0111] S201, obtain a reference shape, and determine a low-dimensional initial shape design space according to the reference shape.

[0112] Among them, the Hicks-Henne parameterization method can be used to characterize the reference shape. The Hicks-Henne parameterization method realizes the deformation of the geometric shape by selecting appropriate candidate point positions and adjusting the weight coefficients of the shape functions. In this solution, the slope change coefficient takes , and the attenuation speed coefficient takes . Refer to Figure 3 , which shows a schematic diagram of the variation relationship of the improved 5th-order Hicks-Henne shape function. The candidate point position , and it can be seen from the distribution law of the function curve that mainly affects the leading-edge line type of the airfoil, mainly affects the middle line type, mainly affects the trailing-edge line type.

[0113] S202, input the initial shape design space into the optimizer.

[0114] S203, iteratively optimize the design parameters of the aircraft through the optimizer to obtain the optimal solution of the initial shape design space output by the optimizer.

[0115] The process of iteratively optimizing the design parameters of the aircraft through the optimizer to obtain the initial shape design space output by the optimizer is as Figure 4 shown, and this process specifically includes:

[0116] S401, adopt the Latin hypercube sampling method to extract a geometric sample set in the initial shape design space.

[0117] As Figure 5 shown,Figure 5 It is a schematic diagram of the position distribution of multiple candidate points on the geometric shape of the aircraft in the initial shape design space, and the positions of the candidate points on the upper and lower surfaces of the airfoil , a total of 6 candidate points

[0118] S402. Conduct CFD numerical simulation on the geometric sample set to calculate its aerodynamic characteristics, and form a geometric-aerodynamic sample library

[0119] Among them, the process of extracting the geometric-aerodynamic sample library from the initial shape design space can be called the DOE process

[0120] S403. Import the geometric-aerodynamic sample library into the Kriging model to obtain a surrogate model

[0121] Among them, the construction process of the surrogate model includes

[0122] S501. Perform preprocessing on each sample in the geometric sample set to form a watertight geometric model

[0123] S502. Use ICEM scripts to realize the generation of the geometric model into computational grids

[0124] To verify the sensitivity of the calculation results to the change of grid density, grid independence verification is carried out under the design conditions. Three sets of grids are designed respectively, and the number of grids is 32000, 62000, and 120000 respectively. The division of the medium grid is referred to Figure 6 , Figure 6 It is a schematic diagram of a RAE 2822 airfoil grid. Compare the lift and drag coefficients calculated by the three sets of grids and the pressure distribution of the symmetric section to conduct grid independence verification

[0125] S503. Use Fluent software to perform CFD accurate numerical simulation to obtain high-fidelity aerodynamic characteristics

[0126] The turbulence model uses the Shear Stress Transport (SST) k model, the convective flux adopts the 2nd-order accurate Advection Upstream Splitting Method (AUSM+) format, the viscous term is obtained through the central difference format, among which, the laminar viscosity coefficient is calculated by the Sutherland formula, and the time advancement adopts the implicit advancement. In terms of boundary condition setting, the far front adopts the pressure far-field boundary, and the fuselage wall surface adopts the adiabatic no-slip wall surface boundary

[0127] The oncoming flow condition is: Mach number = 0.734, Reynolds number = 6.5×10 6, lift coefficient = 0.824. The comparison of the angle of attack and drag coefficient calculated by three groups of meshes (coarse mesh, medium mesh, and fine mesh) is shown in Table 1. The error of the angle of attack when the lift coefficient is satisfied for the three groups of meshes is less than 0.02°, and the error of the corresponding drag coefficient is less than 1 count. The comparison of the result pressure coefficients of the three groups of meshes is shown in Figure 7 , Figure 7 where the abscissa is the position of the candidate point and the ordinate is the pressure coefficient . It can be seen that the consistency of the pressure coefficient is good. Therefore, in order to ensure the calculation accuracy and efficiency during the optimization process, we choose the medium mesh for the subsequent aerodynamic optimization design.

[0128] Table 1

[0129]

[0130] S504, Import the geometric samples and aerodynamic characteristics into the Kriging model to construct the initial surrogate model.

[0131] S404, Based on the surrogate model, using the point addition criterion, perform selection, crossover, and mutation operations on the initial samples of random sampling.

[0132] S405, Evaluate the fitness of the samples and find the sample with the highest fitness.

[0133] S406, Determine whether the sample with the highest fitness converges to the optimal solution of the current shape design space.

[0134] Among them, whether it converges can be determined by whether the sample with the highest fitness changes within 30 - 50 iterations. If it does not change, it converges to the optimal solution; if it does not converge to the global optimal solution, execute step S407, and if it converges to the global optimal solution, execute step S408.

[0135] S407, Add the sample with the highest fitness to the existing sample set to correct the surrogate model.

[0136] S408, Output the sample with the highest fitness.

[0137] Based on the surrogate model, use the genetic algorithm to perform iterative optimization in the current shape design space until the sample with the highest fitness is the optimal solution of the current shape design space.

[0138] S204, Obtain the optimized shape of the aircraft according to the optimal solution of the initial shape design space.

[0139] S205, Determine whether the optimized shape meets the convergence conditions.

[0140] If the optimized shape does not meet the convergence condition, step S206 is executed; if the optimized shape meets the convergence condition, step S208 is executed.

[0141] S206, Calculate the effectiveness index of each candidate point in the initial shape design space.

[0142] As Figure 8 shown, Figure 8 is a schematic diagram of the effectiveness index of each candidate point, showing the position of the candidate points When, the effectiveness index of each candidate point can be found. It can be found that the position of the leading edge of the upper surface is strongly negatively correlated with the target variable and weakly positively correlated with the area. If the target variable is to be reduced, the corresponding weight needs to be increased, and the area will also increase accordingly, meeting the upper boundary constraint of the area. Therefore, when there is a lower boundary constraint , such as the area constraint, and an upper boundary constraint , such as the drag constraint. It can be calculated from the above formula that is the best candidate point, and refining at the position can improve the design space to the greatest extent.

[0143] S207, According to the effectiveness index, refine the initial shape design space to update the initial shape design space, and use the updated initial shape design space as the new initial shape design space and input it into the optimizer.

[0144] As Figure 9 shown, Figure 9 is a schematic diagram of the effect of inserting candidate points in a four-round refinement. Specifically, the way of inserting candidate points at the front and rear edges is significantly different from other positions. It should be noted that in order to prevent unreasonable close spacing in shape control, a minimum spacing is applied between adjacent candidate points. If the best candidate point does not meet the spacing requirement, other candidate points need to be considered to maintain the effectiveness of adding points.

[0145] Triggering the space refinement algorithm is a stop condition, which terminates the optimization of the current shape design space and initiates parameter refinement. The trigger is crucial for efficiency. See Figure 10 , Figure 10 is a schematic diagram of the next-stage optimization iteration curve caused by triggering space refinement at different times. Figure 10 The abscissa of is the number of iterations, and the ordinate is the drag coefficient , for an airfoil drag reduction optimization problem, two branches show the impact of triggering at different times on performance, indicating that different refinement positions are selected at different times, and the optimization results of the next stage are also different. This is because the effectiveness index of candidate points is determined by the Spearman rank correlation coefficient, and the database size will affect the effectiveness index, thereby affecting the position of the best candidate point.

[0146] The specific time to trigger the refinement of the shape design space is to find the optimal solution in the current shape design space and start parameter refinement when full convergence is reached in each cycle. There are two main reasons: First, only when the existing design parameters can no longer improve the objective function is it necessary to add additional candidate points. Second, the effectiveness index composed of the Spearman rank correlation coefficient is accurate only when the optimization reaches full convergence.

[0147] S208, determine the shape design space corresponding to the optimized shape that meets the convergence condition as the best design space of the aircraft, and the optimal solution of this best design space is the global optimal solution.

[0148] In one embodiment, the design constraints include moment constraints and area constraints, and transonic drag reduction aerodynamic optimization design under area constraints and moment constraints is carried out for RAE 2822. The mathematical model of the optimization is:

[0149] (7);

[0150] Among them, is the drag coefficient, is the lift coefficient, is the moment coefficient, is the area of the optimized configuration, is the area of the initial configuration.

[0151] As Figure 11 shown, Figure 11 is a schematic diagram of the convergence history of a fixed candidate point number and adaptive parameterization optimization process. Table 2 shows a comparison of the optimization results of the fixed candidate point number and adaptive parameterization methods. When the candidate point number is of the fixed parameterization method converges in 72 steps, and the after convergence is 145.79 kuang; when is of the fixed parameterization method converges in 146 steps, and the after convergence is 143.62 kuang; the adaptive Hicks-Henne parameterization method with the initial shape design space dimension converges in 80 steps, and the optimization process experiences three rounds of refinement. The after convergence is 137.28 kuang. Generally speaking, the optimization time of the adaptive parameterization method is related to It is basically the same as the fixed parameterization method at that time, and the drag coefficient of the optimization result is reduced by 8.51%.

[0152] Table 2

[0153]

[0154] For the comparison schematic diagram of the geometric and pressure distributions of the optimization results of the fixed and adaptive parameterization methods, see Figure 12 , the optimized geometries all show that the camber of the front half of the upper surface decreases, the position of the maximum thickness moves backward, and the trailing edge has a reverse bend. Compared with the baseline configuration, the shock wave intensity of the three parameterization methods is weakened, and the shock wave drag is significantly reduced. Among them, the shock wave of the adaptive parameterization method is basically eliminated, so the drag coefficient is the lowest. See Figure 13 shows the optimization results and refinement process of each round of the adaptive parameterization method. First, it is refined at the leading edge position of the upper surface. This is because the formation of the shock wave is mainly affected by the camber of the leading edge of the upper surface, making the effectiveness index at this position the largest. Secondly, it is refined at the leading edge of the lower surface and twice. After that, the refinement does not improve the optimization result. Therefore, three rounds of refinement are sufficient to find the optimal airfoil. The adaptive parameterization method better solves the problem that the traditional parameterization method cannot take into account both low-dimensional design variables and the best deformation space.

[0155] Now replace the initial airfoil with NACA 0012, and carry out the transonic drag reduction aerodynamic optimization design under the same working conditions and constraints of RAE 2822. Refer to Figure 14 , which shows the optimization results and refinement process of each round of the adaptive parameterization method starting from NACA 0012. The main refinement positions are the leading edge of the upper surface and the trailing edge of the lower surface, and it converges to the optimal configuration after 8 refinements.

[0156] Table 3 shows the aerodynamic performance of the baseline airfoil (RAE 2822) and the optimization results from two initial shapes of RAE 2822 and NACA 0012. Refer to Figure 15 , which shows the comparison schematic diagram of the geometric shape and pressure distribution of the baseline and the optimization results of the two initial shapes. It is not difficult to find that starting from the two initial shapes of RAE 2822 and NACA 0012 for optimization has little effect on the final optimization result. Except for a slight deviation at the leading edge position, the other positions of the optimized geometric shapes are basically the same. Therefore, the adaptive parameterization method better solves the problem that the optimization result of the traditional parameterization method depends on the initial configuration.

[0157] Table 3

[0158]

[0159] When applying the parameter design method of the aircraft provided by the present invention, it is not necessary toFigure 1 The steps shown are executed in sequence, and the specific execution sequence of each step can be determined as needed, and the present invention does not limit this.

[0160] The above is the parameter design method for an aircraft provided by one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding parameter design system for an aircraft, as Figure 16 shown.

[0161] Figure 16 Schematic diagram of a parameter design system for an aircraft provided by the present invention. The system 1600 includes:

[0162] An initialization module 1601 for constructing an initial shape design space of the aircraft according to the reference shape of the aircraft; the initial shape design space includes design parameters of a plurality of candidate points on the aircraft.

[0163] An optimization module 1602 for iteratively optimizing the design parameters of the aircraft according to the initial shape design space to determine the initial shape design space.

[0164] A refinement module 1603 for calculating the effectiveness index of each candidate point in the initial shape design space and refining the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space when the initial shape design space does not meet the convergence condition; the refinement of the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points.

[0165] An iteration module 1604 for using the updated initial shape design space as the initial shape design space to continue iteratively optimizing the design parameters of the aircraft until the initial shape design space of the aircraft meets the convergence condition, and determining the initial shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

[0166] For the specific limitations on the parameter design system of the aircraft, reference can be made to the limitations on the parameter design method of the aircraft in the above text, which will not be elaborated here. Each module in the above parameter design system of the aircraft can be implemented in whole or in part by software, hardware and their combinations. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0167] The present invention also provides a computer-readable storage medium, which stores a computer program, and the computer program can be used to execute the above Figure 1 provided parameter design method for an aircraft.

[0168] The present invention also providesFigure 17 Schematic structural diagram of the computer device shown, such as Figure 17 shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory. Of course, it may also include other hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the above Figure 1 parameter design method of the aircraft provided.

[0169] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided by the present invention can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0170] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope recorded by the present invention.

Claims

1. A method for designing aircraft parameters, characterized in that: include: According to the reference shape of the aircraft, construct the initial shape design space of the aircraft; The initial shape design space includes design parameters of a plurality of candidate points on the aircraft; According to the initial shape design space, the design parameters of the aircraft are iteratively optimized to determine the optimal solution of the initial shape design space; When the initial shape design space does not meet the convergence condition, calculating the effectiveness index of each candidate point in the initial shape design space, and refining the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space; The refinement of the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points; The updated initial shape design space is used as a new initial shape design space, and the design parameters of the aircraft are continuously optimized iteratively until the shape design space of the aircraft satisfies the convergence condition, and the optimal solution of the shape design space that satisfies the convergence condition is determined as the parameter design strategy of the aircraft; The calculating of the effectiveness index of each candidate point in the initial shape design space includes: obtaining the Spearman rank correlation coefficient between the optimization target of the aircraft and each candidate point in the initial shape design space, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point; determining the effectiveness index of each candidate point according to the Spearman rank correlation coefficient between the optimization target of the aircraft and each candidate point, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point; The method of refining the initial shape design space according to the candidate point with the largest validity index to update the initial shape design space includes: obtaining design parameters of two new candidate points according to the index and design parameters of the candidate point with the largest validity index; and determining an updated initial shape design space according to the design parameters of the two new candidate points and the design parameters of candidate points in the initial shape design space except the candidate point with the largest validity index.

2. The method according to claim 1, characterized in that The index P of the candidate point with the largest effectiveness index is 1, and the calculation formula of the design parameters of the updated initial shape design space is: Among them, the initial shape design space is The updated initial shape design space is d represents the number of refinement rounds, represents the design parameters of the first candidate point in the d-th round of refinement, represents the design parameters of the first candidate point in the d+1th round of refinement, represents the design parameters of the second candidate point in the dth round of refinement, represents the design parameters of the second candidate point in the d+1th round of refinement, represents the design parameters of the jth candidate point in the dth round of refinement, represents the design parameters of the nth candidate point in the dth round of refinement, represents the design parameters of the j+1th candidate point in the d+1th round of refinement, represents the design parameters of the n+1th candidate point in the d+1th round of refinement, j∈[P+1,n], where n is the number of candidate points in the initial shape design space; and These are the design parameters of the two new candidate points; The index P of the candidate point with the largest effectiveness index satisfies 1<P<n, and the calculation formula of the design parameters of the updated initial shape design space is: in, represents the design parameters of the i-th candidate point in the d-th round of refinement, represents the design parameters of the i-th candidate point in the d+1-th round of refinement, i∈[1,P-1], represents the design parameters of the P-1th candidate point in the dth round of refinement, represents the design parameters of the P+1th candidate point in the dth round of refinement, represents the design parameters of the Pth candidate point in the d+1th round of refinement, represents the design parameters of the jth candidate point in the dth round of refinement, represents the design parameters of the P+1th candidate point in the d+1th round of refinement; and These are the design parameters of the two new candidate points; The index of the candidate point with the largest effectiveness index P=n, then the calculation formula of the design parameters of the updated initial shape design space is: in, represents the design parameters of the n-1th candidate point in the dth round of refinement, represents the design parameters of the nth candidate point in the d+1th round of refinement, represents the design parameters of the nth candidate point in the dth round of refinement, represents the design parameters of the n+1th candidate point in the d+1th round of refinement, and These are the design parameters of the two new candidate points.

3. The method according to claim 1, characterized in that Methods for determining whether the initial shape design space meets the convergence conditions include: Obtaining an optimized shape corresponding to an optimal solution of the aircraft in the initial shape design space and an initial configuration of the initial shape design space; the initial configuration is a reference shape of the aircraft corresponding to the initial shape design space when no iterative optimization is performed; If the optimized shape is consistent with the initial configuration, it is determined that the initial shape design space meets the convergence condition; If the optimized shape is inconsistent with the initial configuration, it is determined that the initial shape design space does not meet the convergence condition.

4. The method according to claim 1, characterized in that: The iterative optimization of the design parameters of the aircraft according to the initial shape design space to determine the optimal solution of the initial shape design space includes: A Latin hypercube sampling method is used to extract a geometric sample set in the initial shape design space; Calculating the aerodynamic characteristics of the geometric sample set to obtain a geometric aerodynamic sample library; Import the geometric aerodynamic sample library into the Kriging model to obtain a proxy model; Based on the geometric and aerodynamic constraints of the aircraft, a genetic algorithm is used according to the surrogate model to iteratively optimize the design parameters of the aircraft until the optimal solution of the initial shape design space is found.

5. The method according to claim 1, characterized in that The step of constructing an initial shape design space of the aircraft according to the reference shape of the aircraft includes: The Hicks-Henne parameterization method is used to characterize the reference shape of the aircraft and construct the initial shape design space.

6. A parameter design system for an aircraft, characterized in that: The system comprises: An initialization module, used to construct an initial shape design space of the aircraft according to a reference shape of the aircraft; the initial shape design space includes design parameters of the aircraft at multiple candidate points; An optimization module is used to iteratively optimize the design parameters of the aircraft according to the initial shape design space to determine the optimal solution of the initial shape design space; A refinement module is used for calculating the effectiveness index of each candidate point in the initial shape design space when the initial shape design space does not meet the convergence condition, and refining the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space; the refinement of the initial shape design space means introducing new design parameters into the initial shape design space by inserting candidate points; the calculation of the effectiveness index of each candidate point in the initial shape design space includes: obtaining the Spearman rank correlation coefficient between the optimization target of the aircraft and each candidate point in the initial shape design space, and the Spearman rank correlation coefficient between the design constraint of the aircraft and each candidate point Spearman rank correlation coefficient; determining the effectiveness index of each candidate point according to the Spearman rank correlation coefficient between the optimization target of the aircraft and each candidate point, and the Spearman rank correlation coefficient between the design constraints of the aircraft and each candidate point; the refining of the initial shape design space according to the candidate point with the largest effectiveness index to update the initial shape design space includes: obtaining the design parameters of two new candidate points according to the index and design parameters of the candidate point with the largest effectiveness index; determining the updated initial shape design space according to the design parameters of the two new candidate points and the design parameters of the candidate points in the initial shape design space except the candidate point with the largest effectiveness index; The iterative module is used to use the updated initial shape design space as a new initial shape design space, continue to iteratively optimize the design parameters of the aircraft until the shape design space of the aircraft meets the convergence condition, and determine the optimal solution of the shape design space that meets the convergence condition as the parameter design strategy of the aircraft.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

Images