An integrated gradient optimization method for internal and external flow of aircraft based on hybrid parameterization

Abstract

The present invention discloses a hybrid parameterized integrated gradient optimization method for internal and external flows of an aircraft, which is mainly used for the coordinated optimization of the aerodynamic shape and the intake and exhaust system of an aircraft with an integrated flight and launch. The method parameterizes the aerodynamic shape of the aircraft through FFD geometric parameterization technology, and parameterizes the cross-sectional shape, area law and central axis of the intake and exhaust system by combining two-dimensional shape fusion function and B-spline curve, so as to realize the construction of a hybrid parameterized model of the aerodynamic shape and the intake and exhaust system. Then, the objective function and constraint conditions of the optimization problem are defined, and the computational grid of the internal and external flow fields of the aircraft is deformed and updated based on the optimized design variables. Through CFD numerical simulation and discrete adjoint method, the objective function gradient information is obtained and iterative optimization is performed, and finally an aircraft configuration with better performance is obtained. The method can effectively improve the aerodynamic performance and internal flow quality of the aircraft with an integrated flight and launch, and provide strong support for its efficient optimization design.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft design and relates to the aerodynamic optimization design of an integrated flight-engine aircraft. Specifically, it is a method for integrated gradient optimization of internal and external flows of an aircraft based on hybrid parameterization. By combining the parameterization of aerodynamic shape with the geometric parameterization design of intake and exhaust systems, the coordinated optimization of the aerodynamic shape of the aircraft and the intake and exhaust systems is achieved, thereby improving the aerodynamic performance and internal flow quality. Background Art

[0002] Compared with traditional configurations, the flying wing layout has gradually become a research hotspot in modern aircraft design because of its significant advantages in aerodynamic performance and structural performance. This design integrates the fuselage and the wing to achieve coordinated optimization of aerodynamic shape and structural performance, reduce air resistance, improve the endurance and load capacity of the aircraft, and optimize the overall aerodynamic efficiency. However, in order to achieve optimal performance, the flying wing layout usually embeds the engine's intake and exhaust systems into the fuselage, with obvious flight-engine integration characteristics. This feature significantly increases the impact of the intake and exhaust systems on the aerodynamic performance of the flying wing, which requires detailed consideration of the intake and exhaust systems during design.

[0003] A lot of research has been done at home and abroad on the parametric design methods of intake and exhaust systems. For example, the prior art proposes a geometric parametric method for short-diffuser, large-offset S-bend inlets. On this basis, a centerline design method based on polynomial expressions is proposed for the geometric boundary conditions of the S-bend inlet, further expanding the selection range of the centerline. On the basis of the above conventional methods, the prior art also proposes a design method that uses a super ellipse equation to control the generation of the cross section based on the law of change of the centerline and the cross-sectional area, and uses the super ellipse method to carry out the design of S-bend inlets that transition from various shapes of inlets to circular shapes. Although these intake and exhaust system parametric design methods have achieved good results, most of them are only applicable to the design of simpler intake and exhaust systems, and it is difficult to meet the overall optimization design requirements of complex aircraft.

[0004] In addition, there are two main types of commonly used aircraft optimization design frameworks: non-gradient optimization design methods and gradient optimization design methods. Among them, non-gradient optimization methods include inverse design methods, proxy model optimization design methods and intelligent optimization methods. However, non-gradient algorithms usually have too much computation and too high time cost. Although the time cost can be significantly reduced with the help of advanced proxy models, this method can only efficiently handle optimization problems with design variables within 100 dimensions. This type of method does not rely on the gradient information of the design variables and is suitable for optimization problems with fewer design variables. Once there are too many design variables, the optimization method based on the non-gradient algorithm will face the "dimensionality disaster". At the same time, the proxy model generally has insufficient generalization ability. Compared with non-gradient optimization methods, gradient optimization methods have higher convergence speed and computational efficiency. Gradient optimization methods rely on the gradient information of the objective function relative to the design variables, and quickly perform optimization iterations by calculating the impact of changes in design variables on the objective function. In this type of method, the computational cost of the gradient solution method based on discrete adjoint theory is almost independent of the scale of design variables. Therefore, the gradient optimization method based on discrete adjoint is one of the most effective methods to solve optimization problems with large-scale design variables and has been widely used in full turbulence engineering optimization design.

[0005] However, due to the difficulty in accurately solving the gradient of the objective function relative to the design variables, the detailed parameterization methods of the intake and exhaust systems are currently only widely used in non-gradient optimization design, and their application in the field of gradient optimization is limited. Therefore, it is urgent to develop a parameterization method that can accurately capture the geometric characteristics of the aircraft intake and exhaust system and the aerodynamic system, and based on this method, to construct a gradient optimization method that can give full play to the advantages of the aircraft flight and engine integrated design. Summary of the invention

[0006] 1. Purpose of the invention

[0007] In view of the adverse effects on the aerodynamic performance of existing aircraft after considering the integrated design of flight and engine, the existing parameterization methods are difficult to accurately capture the complex geometric characteristics of the aircraft's intake and exhaust system and aerodynamic system. At the same time, the gradient optimization method is limited in the application of detailed parameterization of the intake and exhaust system. In order to solve at least one of the above-mentioned and other technical problems in the prior art, the present invention aims to provide an integrated gradient optimization method for internal and external flows of an aircraft based on hybrid parameterization. By combining the aerodynamic shape parameterization with the geometric parameterization design of the intake and exhaust system, a hybrid parameterization model that can accurately capture the geometric characteristics of the aircraft's intake and exhaust system and aerodynamic system is constructed, and a gradient optimization framework is established based on the model to achieve the integrated optimization design of the aircraft's aerodynamic shape and the intake and exhaust system, thereby significantly improving the aerodynamic performance and inlet flow quality of powered aircraft, giving full play to the advantages of the integrated design of flight and engine, and providing theoretical guidance and methodological support for the efficient optimization design of powered aircraft.

[0008] (II) Technical solution

[0009] In order to achieve the purpose of the invention and solve the technical problems, the present invention adopts the following technical solutions:

[0010] A hybrid parameterized integrated gradient optimization method for internal and external flows of an aircraft is used for the coordinated optimization of the aerodynamic shape and the intake and exhaust system of an integrated flight and engine aircraft, and at least includes the following implementation steps:

[0011] SS1. Hybrid parameterization of the initial configuration of the aircraft to be optimized

[0012] An existing integrated flight and engine vehicle is selected as the initial configuration to be optimized, and the aerodynamic shape of the vehicle is parameterized by using the FFD (Free-FormDeformation) geometric parameterization method. The cross-sectional shape, area ratio and central axis of the intake and exhaust system are parameterized by using a method based on two-dimensional shape fusion function and B-spline curve, so as to realize the hybrid parameterization of the aerodynamic shape of the vehicle and the geometric characteristics of the intake and exhaust system.

[0013] SS2. Define the objective function and constraints of the optimization problem

[0014] A certain flight state of the aircraft is selected as an optimization benchmark and used as an optimization state, a number of key aerodynamic parameters in the optimization state are set as fixed conditions in the optimization process, and design variables, objective functions and constraints related to the optimization of the internal and external flow integration of the initial configuration are given, wherein the design variables at least include aerodynamic shape design variables and intake and exhaust system design variables, the objective function at least includes aircraft aerodynamic performance indicators and intake and exhaust system flow quality indicators, and the constraints at least include geometric and / or performance constraints related to the aerodynamic shape and the intake and exhaust system;

[0015] SS3. Deformation and update of the computational grid for the internal and external flow fields of the aircraft

[0016] Based on the hybrid parameterization results of step SS1, according to the changes in the FFD control point parameters and the geometry of the intake and exhaust systems in the design variables, and in combination with the inverse distance weighting method, the computational grids including the surface grid and the space grid of the internal and external flow fields of the aircraft are deformed and updated, and the quality of the computational grid is ensured to meet the accuracy and stability requirements of the subsequent CFD (Computational Fluid Dynamics) calculations;

[0017] SS4. CFD numerical calculation of the internal and external flow fields of aircraft

[0018] Based on the updated computational grid, the RANS equations (Reynolds-averaged Navier-Stokes equations) and the SA turbulence model (i.e., the Spalart-Allmaras turbulence model) are used to perform CFD numerical calculations of the internal and external flow fields of the aircraft. The aerodynamic performance of the aircraft and the flow quality data of the intake and exhaust system are obtained through numerical solution, providing the necessary data for the evaluation and iteration of the objective function.

[0019] SS6. Verification and analysis of internal and external flow integrated design results

[0020] Perform high-precision CFD numerical simulation on the optimized aircraft configuration to evaluate and verify its aerodynamic performance and flow quality of the intake and exhaust system. If the verification result reaches the expected goal, the optimization is completed and the final design scheme is determined. If the expected goal is not reached, adjust the design variables of the aircraft aerodynamic shape and / or the intake and exhaust system, and re-execute steps SS1 to SS6 until a satisfactory optimization result is obtained.

[0021] (III) Technical Effect

[0022] Compared with the prior art, the hybrid parameterized integrated gradient optimization method for internal and external flows of an aircraft of the present invention has the following beneficial and significant technical effects:

[0023] (1) The present invention uses a hybrid parameterization method to coordinately optimize the aerodynamic shape and the intake and exhaust system of the aircraft, thereby effectively improving the problem of mutual influence between the aerodynamic shape and the intake and exhaust system in traditional aircraft design. By performing unified parameterization and gradient optimization calculation on both, it is ensured that the optimization of the aerodynamic shape will not have a significant negative impact on the performance of the intake and exhaust system, and vice versa, effectively improving the aerodynamic performance and internal flow quality of the powered aircraft.

[0024] (2) The present invention can adapt to complex geometric changes and ensure the efficiency and accuracy of the optimization process by accurately modeling and optimizing the aerodynamic shape of the aircraft and the geometric features of the intake and exhaust system. Based on these methods, the aerodynamic shape of the aircraft, the cross-sectional shape, area law and central axis of the intake duct and / or tail nozzle in the intake and exhaust system can be accurately adjusted under multivariable conditions, so that the optimized aerodynamic shape and intake and exhaust system have better performance.

[0025] (3) The present invention uses the adjoint method to solve the gradient information of the optimization target, which can quickly and efficiently calculate the gradient of the objective function to the geometric design variables, significantly improving the convergence speed and computational efficiency of large-scale design variable optimization problems. By combining the adjoint method with the chain derivative rule, rapid iterative optimization can be achieved in the multi-dimensional design space, ensuring the efficiency and accuracy of aircraft design. In addition, the gradient optimization method of the present invention has good applicability for the optimization design of large-scale complex geometries, especially when dealing with the coordinated optimization of the intake and exhaust system and the aerodynamic shape, it can significantly reduce the complexity of the computational grid and the optimization time, thereby improving the design efficiency and performance reliability of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The figure shows a schematic diagram of the implementation process of the aircraft internal and external flow integrated gradient optimization method based on hybrid parameterization of the present invention;

[0027] Figure 2 Shown is a schematic diagram of the corresponding relationship between the inlet and outlet two-dimensional graphics of the air inlet / tail nozzle;

[0028] Figure 3 The figure shows a schematic diagram of a cross-sectional shape set obtained based on a cross-sectional shape parameterization method;

[0029] Figure 4 Schematic diagram of B-spline curve and control points of area law;

[0030] Figure 5 Schematic diagram of the B-spline curve and control points of the central axis;

[0031] Figure 6 Schematic diagram of B-spline and control points before and after deformation;

[0032] Figure 7 Shown are schematic diagrams of configuration optimization constraints (A) and initial configuration optimization FFD box (B);

[0033] Figure 8 Shown is a schematic diagram of the comparison of aerodynamic performance between the initial configuration and the optimized configuration;

[0034] Fig. 9The figure shows the comparison of pressure distribution and airfoil thickness change of the initial configuration (black line) and the optimized configuration (red line). In the figure, (A) is the pressure distribution comparison at the Y / b=2.7 section position of the wing, (B) is the comparison of the airfoil thickness change at the Y / b=2.7 section position, (C) is the pressure distribution comparison at the Y / b=4.5 section position of the wing, and (D) is the comparison of the airfoil thickness change at the Y / b=4.5 section position.

[0035] Fig.10 The figure shows the comparison of pressure distribution and airfoil thickness change between the initial configuration (black line) and the optimized configuration (red line). In the figure, (E) is the pressure distribution comparison at the Y / b=6.2 section position of the wing, (F) is the airfoil thickness change comparison at the Y / b=6.2 section position, (G) is the pressure distribution comparison at the Y / b=7.9 section position of the wing, and (H) is the airfoil thickness change comparison at the Y / b=7.9 section position.

[0036] Fig.11 The figure shows a comparison diagram of the total pressure recovery coefficient of the intake duct before and after optimization;

[0037] Fig.12 Shown is a comparison of the Mach number cloud diagrams of the inlet before and after optimization;

[0038] Fig.13 The figure shows the comparison of the geometric features of the air intake before optimization (black line) and after optimization (red line). In the figure, A~H correspond to different axis positions respectively;

[0039] Fig.14 Shown is a schematic diagram comparing the cross-sectional shapes of the air intake duct at four different axis positions (A), (B), (C), and (D) before optimization (black line) and after optimization (red line);

[0040] Fig.15 The figure shows the comparison of the cross-sectional shapes of the air intake duct at four different axis positions (E), (F), (G), and (H) before optimization (black line) and after optimization (red line);

[0041] Fig.16 Shown is a schematic diagram comparing the cross-sectional area (A) and center axis (B) distribution of the inlet duct along the axis before optimization (black line) and after optimization (red line). DETAILED DESCRIPTION

[0042] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. The described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present invention, and should not be understood as limiting the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0043] The present invention aims to provide an integrated gradient optimization method for internal and external flows of an aircraft based on hybrid parameterization. By combining the aerodynamic shape parameterization with the geometric parameterized design of the intake and exhaust system, a hybrid parameterized model is constructed that can accurately capture the geometric characteristics of the aircraft's intake and exhaust system and the aerodynamic system, thereby realizing the integrated optimization design of the aircraft's aerodynamic shape and the intake and exhaust system, thereby significantly improving the aerodynamic performance and inlet flow quality of powered aircraft, giving full play to the advantages of the integrated design of the aircraft and the engine, and providing theoretical guidance and methodological support for the efficient optimization design of powered aircraft.

[0044] Example 1

[0045] As a specific example, Figure 1 As shown, the hybrid parameterized integrated gradient optimization method for internal and external flows of an aircraft of the present invention is used for the coordinated optimization of the aerodynamic shape and the intake and exhaust system of an integrated flight and engine aircraft. The method comprises at least the following steps when implemented:

[0046] SS1. Hybrid parameterization of the initial configuration of the aircraft to be optimized

[0047] An existing integrated flight and engine vehicle is selected as the initial configuration to be optimized. The aerodynamic shape of the vehicle is parameterized by the FFD parameterization method. The cross-sectional shape and central axis of the intake and exhaust system are geometrically parameterized by a method based on two-dimensional shape fusion function and B-spline curve, thus realizing the hybrid parameterization of the aerodynamic shape of the vehicle and the geometric characteristics of the intake and exhaust system.

[0048] Preferably, in this step, when the aerodynamic shape of the aircraft is subjected to FFD parameterization processing, the aerodynamic shape of the aircraft is embedded in a multi-level FFD control framework, each FFD control framework corresponds to a different local shape area of ​​the aircraft, and the FFD control framework establishes a functional relationship with the nodes on the aerodynamic shape through the Bernstein basis function, and the aerodynamic shape is adjusted by moving the FFD control points.

[0049] More specifically, the FFD method uses the Bernstein basis function to establish the functional relationship between the nodes and the points in the control frame, and its specific expression is:

[0050]

[0051] in, X ffd is the global coordinate value of any point in the control frame; P i,j,k is the local coordinate value of the control point; s , t , u are the points in the local coordinate system O ' STUYZ Middle Edge S , T , U The coordinate value of the direction; They are l , m , n The second-order Bernstein polynomial basis function is defined as:

[0052]

[0053] When the displacement of the control point on the FFD control volume is Δ P i,j,k When , the local coordinates on the control object are ( s , t , u ) point X ffd ( s , t , u ) is:

[0054]

[0055] Preferably, in this step, a method based on a two-dimensional shape fusion function is used to parameterize the cross-sectional shape of the intake duct and / or tail nozzle in the intake and exhaust system, which at least includes the following sub-steps:

[0056] First, extract the two-dimensional cross-sectional shapes at the inlet and / or tail nozzle inlet and outlet, discretize the cross-sectional shapes of the inlet and outlet into a set of point sets, and ensure that the two cross-sectional shapes have the same number of discrete points after discretization. Then, align the centroids of the two cross-sectional shapes through translation operation to ensure the consistency of the two cross-sectional shapes in space, such as Figure 2 As shown, where:

[0057]

[0058] in, f in , f out are the inlet and outlet cross-sectional shapes, respectively. is the discrete point coordinates of the inlet and outlet cross-sectional shapes, preferably d =1,2,…, n 1, n 1 is the preset number of discrete points;

[0059] Next, the correspondence between the graphics is determined based on the geometric features of the two 2D cross-sectional shapes of the inlet and outlet, and the point-to-point correspondence between the two discrete point sets is established based on the key geometric features of the two cross-sectional shapes of the inlet and outlet (such as edge curvature and feature points, etc.), and the intermediate cross-sectional shape is generated through a 2D shape fusion function, the form of which is:

[0060]

[0061] in, f in , f out Respectively represent the inlet and outlet cross-sectional shapes, f ( x i ) indicates the position of the air inlet and / or tail nozzle axis x i The cross-sectional shape at is the weight function to control the shape change;

[0062] Finally, the weight function is applied Perform shape interpolation, where x i is the axis position of the air inlet and / or tail nozzle, x in is the inlet cross-sectional position, x out is the exit section position, α is the control parameter and α >0, by adjusting the parameters α Control the nonlinearity of the shape transition and construct a set of cross-sectional shapes that transition from the inlet to the outlet { f in … f ( x i )… f out},like Figure 3 As shown, f ( x i ) is the position along the axis of the air inlet and / or tail nozzle x i The cross-sectional shape at the location and ensure that the shape transition is smooth and continuous.

[0063] Preferably, in this step, the B-spline curve method is used to parameterize the area law along the inlet duct and / or the tail nozzle, including:

[0064] First, extract the cross-sectional area values ​​of the inlet duct and / or tail nozzle at different flow direction sections and form a point set {( x i , S i )},in S i is the position along the axis of the inlet and / or tail nozzle x i The cross-sectional area at the axis position x i By selecting uniform or non-uniform distribution, ensure that the area information covers the key positions of the entire inlet and / or tail nozzle, and preferably i =1,2,…, n 2, n 2 is the preset number of cross-section sampling points;

[0065] Next, from the point set {( x i , S i )} to extract key feature points that represent the main trend of area change along the route and can reflect the key change position of the area (such as area positive value points, inflection points of area change, etc.), and form a new feature point set {( x j , S j )},in S j The critical change position of the area on the axis of the inlet and / or tail nozzle x j The cross-sectional area at j =1,2,…, n 3, among which n 3 is the number of feature points extracted and n 3≤ n 2;

[0066] Afterwards, the B-spline theory is used to extract the feature point set {( x j , S j )} to interpolate and obtain several benchmark control points along the area change P k and node vector U 1, where P k =( x k , S k), S k Axis position x k The cross-sectional area at k =1,2,…, n 4. n 4 is the number of reference control points and n 4= n 3. Vector U 1 is used to determine the segmentation and interpolation interval of the curve, defined as

[0067]

[0068] in x min , x max are the starting position and ending position of the axis of the air inlet and / or tail nozzle, respectively. p is the order of the B-spline curve, m +1 is the length of the node vector and satisfies m = n 4+ p +1, and the benchmark control point P k and node vector U 1 satisfies the following B-spline curve equation:

[0069]

[0070] in, S ( x ) indicates the axis position x The cross-sectional area at N k,p ( x ) is the k Control Points P k About the axis position x of p The B-spline basis function is Figure 4 As shown;

[0071] Finally, a new design variable Δ is introduced S k , and obtain a new set of control points {( x k , S k +Δ S k )}, and generate a new along-the-way area distribution curve by adjusting the position of the control points:

[0072]

[0073] Among them, the design variable Δ S k Satisfy ΔS k_min ≤ Δ S k ≤ΔS k_max , Δ S k_min , Δ S k_max The value of is set according to the actual engineering requirements, and the design variable Δ S k Optimize the area distribution curve, B-spline and control points before and after deformation Figure 6 shown.

[0074] Preferably, in this step, the central axis of the air inlet and / or the tail nozzle is parameterized using a B-spline curve method, including:

[0075] First, based on the geometric structure of the air inlet and / or tail nozzle, the center points at different flow sections are extracted to form a center point set {( x l , y l , z l )},in( x l , y l , z l ) is the cross-sectional position on the axis of the air inlet and / or tail nozzle x l The spatial coordinates of the center point at the position are selected based on geometric symmetry and must ensure that the curvature and changes along the axis can be accurately described, and preferably l =1,2,…, n 5. n 5 is the preset number of center point samples;

[0076] Next, from the center point set {( x l , y l , z l )} to extract key feature points that represent the main change trend of the central axis (such as axis extreme points, inflection points with large curvature changes, etc.), and form a new feature point set {( x m , y m , z m )}, feature point( x m ,y m , z m ) is the cross-sectional position on the axis of the air inlet and / or tail nozzle x m The spatial coordinates at the position are extracted based on the curvature change, spatial displacement and / or fluid dynamics optimization requirements, and it is necessary to ensure that the key geometric characteristics of the central axis can be accurately captured, wherein the preferred m =1,2,…, n 6. n 6≤ n 5;

[0077] Afterwards, the B-spline theory is used to extract the feature point set {( x m , y m , z m )} to interpolate and generate the B-spline curve of the central axis, and obtain the reference control point set {( x n , y n , z n )},like Figure 5 As shown, each benchmark control point ( x n , y n , z n ) is used to control the geometry of the B-spline curve, preferably n =1,2,…, n 7, n 7 is the number of reference control points and n 7= n 6, and by reasonably choosing the node vector U 2 to control the segmentation and interpolation interval of the B-spline curve, the length of the node vector and the number of control points n 7 and the degree of the curve p Related;

[0078] Finally, a new design variable (Δ y n ,Δ z n ), and obtain a new set of control points by adjusting the positions of the control points {( x n , y n +Δ y n , z n +Δ zn )}, generate a new B-spline curve about the central axis:

[0079]

[0080] in, C new ( x ) indicates the position of the air inlet and / or tail nozzle axis x The spatial position of N n,p ( x ) is the n The position of the control points about the axis x of p B-spline basis function, design variable (Δ y n ,Δ z n ) is used to optimize the position of the control points, improve the local geometry of the central axis, and ensure that the central axis can accurately describe the geometric changes at key positions. The B-spline and control points before and after deformation are as follows: Figure 6 shown.

[0081] SS2. Define the objective function and constraints of the optimization problem

[0082] A certain flight state of the aircraft is selected as an optimization benchmark and used as the optimization state, and several key aerodynamic parameters under the optimization state are set as fixed conditions in the optimization process. At the same time, design variables, objective functions and constraints related to the integrated optimization of internal and external flows of the initial configuration are given, wherein the design variables at least include aerodynamic shape design variables and intake and exhaust system design variables, the objective function at least includes aircraft aerodynamic performance indicators and intake and exhaust system flow quality indicators, and the constraints at least include geometric and / or performance constraints related to the aerodynamic shape and the intake and exhaust system.

[0083] Preferably, in this step, the key aerodynamic parameters set in the optimization state include at least the incoming flow Mach number Ma, the airflow angle of attack AoA, the lift coefficient C L and / or drag coefficient C D The aerodynamic shape design variables are the selected FFD control point parameters related to the aerodynamic shape geometric parameters of the aircraft (such as wingspan, chord length, etc.), and the intake and exhaust system design variables are the geometric parameters related to the cross-sectional shape, along-the-path area law and central axis of the intake and exhaust system and the tail nozzle profile.

[0084] At the same time, the aerodynamic performance index of the aircraft in the objective function at least includes the minimized drag coefficient C under the optimized state. D , maximize the lift coefficient C L, maximize lift-to-drag ratio C L / C D The flow quality index of the intake and exhaust system at least includes the total pressure recovery coefficient of the intake duct under the optimized state. σ , Inlet distortion coefficient σ d The objective function comprehensively considers multiple performance indicators in a weighted summation manner, and different performance indicators have different weight coefficients to achieve multi-objective optimization of the aircraft's aerodynamic performance and the flow quality of the intake and exhaust system.

[0085] In addition, the constraints include at least one or more of the geometric dimension constraints related to the aerodynamic shape of the aircraft (such as the maximum and minimum values ​​of the maximum thickness, wingspan, fuselage length, etc.), aerodynamic performance constraints (such as the minimum lift coefficient, maximum drag coefficient, etc.), pitch moment constraints, and flow stability constraints related to the intake and exhaust systems (such as static pressure, total temperature, total pressure, flow coefficient, etc.).

[0086] SS3. Deformation and update of the computational grid for the internal and external flow fields of the aircraft

[0087] Based on the hybrid parameterization results of step SS1, according to the changes in the FFD control point parameters and the intake and exhaust system geometry in the design variables, and in combination with the inverse distance weighting method, the computational grids including the surface grid and the space grid of the internal and external flow fields of the aircraft are deformed and updated, and it is ensured that the quality of the computational grid meets the accuracy and stability requirements of subsequent CFD calculations.

[0088] Preferably, in this step, based on the parameterized results of the cross-sectional shape, area law and central axis of the intake duct and / or tail nozzle in the intake and exhaust system completed in step SS1, the profile of the intake duct and / or tail nozzle is formed by combining these parameterized results, and then the grid coordinates related to the profile of the intake duct and / or tail nozzle in the calculation grid are updated according to the change of the design variables, including:

[0089] First, import the computational mesh of the inlet and / or tail nozzle G and simplified surface meshes of inlet and / or nozzle profiles He , from the computational mesh by keyword extraction methods (e.g. geometric feature recognition methods) G Extract the grid coordinate set related to the inlet and / or tail nozzle profile H , and establish a grid coordinate set H With simplified surface mesh He The mapping relationship between H = L ( He ),in LFor the mapping function, ensure that the mesh data remains consistent during the geometry adjustment process;

[0090] Secondly, based on the simplified surface mesh He Extract the cross-sectional shapes at the inlet and / or the tail nozzle at the inlet and the outlet, and construct a cross-sectional shape set from the inlet to the outlet according to the parameterized results of the cross-sectional shapes of the inlet and / or the tail nozzle. f in … f ( x i )… f out}, then according to the position of each axis x i The cross-sectional shape at f ( x i ) extracts its corresponding center point ( x i , y i , z i ) and area S i ;

[0091] Next, according to the changes in the design variables, the area law along the path and the B-spline curve equation of the central axis after parameterization in step SS1 are optimized and adjusted respectively. Then, according to the new area law and the B-spline curve equation of the central axis, the position of each axis is optimized. x i The corresponding target center point positions are obtained by interpolation calculation ( x i , y ' i , z ' i ) and target area A i ;

[0092] Afterwards, the target center point position ( x i , y ' i , z ' i ) and target area A i , for each axis position x i The original cross-sectional shape at f ( x i ) to scale and translate, with its original center point ( x i ,y i , z i ) as the center to scale the area and obtain the cross-sectional area that conforms to the new area law distribution. A i , the zoom ratio is , the cross-sectional shape after scaling is , and translate the scaled cross-section shape to the target center point position ( x i , y ' i , z ' i ) to generate an updated simplified surface mesh He' ;

[0093] Finally, through the mapping relationship H = L ( He ), the updated simplified surface mesh He' Mapping to the computational grid, during the mapping process, ensure that the geometrically adjusted grid coordinates have sufficient smoothness and continuity, and ultimately achieve the update of the grid coordinates related to the inlet and / or tail nozzle profile.

[0094] SS4. CFD numerical calculation of the internal and external flow fields of aircraft

[0095] Based on the updated computational grid, the RANS equations and SA turbulence model are used to perform CFD numerical calculations of the internal and external flow fields of the aircraft. The aerodynamic performance of the aircraft and the flow quality data of the intake and exhaust systems are obtained through numerical solution, providing the necessary data for the evaluation and iteration of the objective function.

[0096] Preferably, in this step, the RANS equation is discretized using the finite volume method, the spatial discretization uses the second-order upwind format, and the time discretization uses the implicit double time stepping method; and in terms of boundary condition setting, the static pressure boundary condition is used at the inlet outlet (engine inlet), the total temperature and total pressure boundary condition is used at the tail nozzle inlet (engine outlet), and the velocity boundary condition is used in the far field.

[0097] SS5. Objective function gradient information solution and iterative optimization

[0098] The gradient information of the objective function relative to the design variables is solved based on the discrete adjoint method, and the optimization problem defined in step SS2 is iteratively solved by combining the CFD calculation results and the obtained gradient information in step SS4 through the gradient optimization algorithm, and the geometric characteristics of the aerodynamic shape and the intake and exhaust system are gradually optimized based on the iteratively adjusted design variables. The CFD calculation is re-performed after each round of iteration until the objective function converges or reaches the predetermined optimization standard;

[0099] Preferably, in this step, the implementation of the discrete adjoint method includes: first constructing a discrete flow field control equation and its corresponding adjoint equation, then solving the discrete adjoint equation to obtain adjoint variables, and finally using the adjoint variables to calculate the gradient of the objective function with respect to the design variables; when solving the adjoint equation, using the same discrete format and solver as the original flow field equation to ensure the consistency and accuracy of the gradient calculation; and using a matrix-free calculation method in the gradient calculation process to improve the calculation efficiency and avoid directly storing and inverting large-scale Jacobian matrices.

[0100] Preferably, in this step, the gradient optimization algorithm adopts a sequential quadratic programming (SQP) method, and gradually approaches the optimal solution by solving quadratic programming subproblems in each iteration. In specific implementation, the BFGS (Broyden-Fletcher-Goldfarb-Shanno) algorithm is used to update the Hessian matrix approximation, and a line search strategy is used to determine the step size of each iteration; at the same time, in order to deal with nonlinear constraints, an augmented Lagrangian function is introduced to transform the constrained optimization problem into a series of unconstrained optimization subproblems; and in the optimization process, a multi-starting point strategy is combined to start the optimization from different initial points to avoid falling into the local optimum and increase the possibility of obtaining the global optimal solution.

[0101] Preferably, in this step, the gradient information of the objective function relative to the design variables is obtained by decomposing the geometric design variables of the aerodynamic shape of the aircraft and the intake and exhaust system, specifically including:

[0102] First, the objective function I For geometric design variables Y Gradient information dI / d Y Decompose into objective function I Aerodynamic shape design variables Y Gradient of 1 dI / d Y 1 and the objective function I Design variables for intake and exhaust systems Y Gradient 2 dI / d Y 2, expressed as:

[0103]

[0104] Secondly, the optimization objective is calculated with respect to the surface coordinate variables by the adjoint method X Gradient information dI / dX , where the surface coordinate variables X Contains the geometric grid coordinates corresponding to the aerodynamic shape of the aircraft X 1 and the geometric grid coordinates corresponding to the intake and exhaust system X2. The transmission relationship of gradient information is described as:

[0105]

[0106] in, dI / dX 1 and dI / dX 2 by dI / dX Decomposed into, dX 1 / d Y 1 and dX 2 / d Y 2 are respectively based on the geometric parameterization of the aircraft aerodynamic shape and the intake and exhaust system and are analytically derived using the chain rule.

[0107] SS6. Verification and analysis of internal and external flow integrated design results

[0108] Perform high-precision CFD numerical simulation on the optimized aircraft configuration to evaluate and verify its aerodynamic performance and flow quality of the intake and exhaust system. If the verification result reaches the expected goal, the optimization process is completed and the final design scheme is determined. If the verification result fails to reach the expected goal, the specific factors causing the performance deviation are analyzed, and the design variables of the aerodynamic shape and / or the intake and exhaust system are automatically adjusted. Steps SS1 to SS6 are re-executed and the optimization process is repeated until a satisfactory optimization result is obtained.

[0109] Example 2

[0110] Based on the above-mentioned embodiment 1, as a more specific application example, this embodiment 2 briefly describes the aerodynamic optimization problem of the X1 flying wing layout by applying the above-mentioned internal and external flow integrated gradient optimization method based on hybrid parameterization, and the implementation process is as follows:

[0111] Select X1 configuration for optimization design, the optimization state is Ma=0.70, C L =0.20, the optimization target is the optimal resistance coefficient C D , total pressure recovery coefficient σ And the distortion coefficient σ d Combination 100C D - σ+σ d The moment coefficient C of the aerodynamic constraint in the optimized state is myBalance. The optimization variables are: ① The wing profiles at y=0.9m, y=1.83m, y=2.7m, y=3.6m, y=4.5m, y=5.4m, y=6.2m, y=7.0m, y=7.9m, and y=9.0m; ② The air inlet profiles at x=2.56m, x=2.73m, x=3.14m, x=3.58m, x=3.90m, x=4.25m, x=4.71m, and x=4.93m. The optimized wings are profiles at y=0.9m, y=1.83m, y=2.7m, y=3.6m, y=4.5m, y=5.4m, y=6.2m, y=7.0m, y=7.9m, and y=9.0m, and thickness constraints are imposed on the above airfoils: the leading edge is from 3%-8%, the middle section is from 8%-85%, and the trailing edge is from 85%-92%. The upper limit of the thickness is 3 times the local thickness, and the lower limit is 0.98 times the local thickness. Figure 7 The following is a schematic diagram of the configuration optimization constraint (A) and the initial configuration optimization FFD box (B). The comparison of relevant performance parameters before and after optimization is shown in Table 1.

[0112] Table 1 Comparison of performance parameters before and after optimization

[0113]

[0114] As can be seen from Table 1, the angle of attack (AoA) is reduced from 3.3° before optimization to 3.065821°, which shows that the optimized aerodynamic shape of the aircraft can maintain the same lift coefficient (C L =0.200). Secondly, the resistance coefficient C D The total pressure recovery coefficient decreased from 0.010583 to 0.010042, indicating that aerodynamic optimization significantly reduced the drag of the aircraft. σ The improvement from 0.9276 to 0.9751 shows the improvement of airflow quality after the intake duct optimization. σ d The decrease from 0.0807 to 0.0578 indicates that the uniformity of the airflow distribution in the inlet is further improved. In summary, the aerodynamic optimization scheme significantly improves the overall performance of the aircraft and effectively achieves the optimization goal.

[0115] Figures 8-10 The aerodynamic performance and geometric characteristics of the configurations before and after optimization are compared. It can be seen from the pressure coefficient cloud map and the wing section pressure coefficient distribution map in the figure that optimization can significantly improve the lift distribution of the wing, thereby reducing the induced drag of the flying wing. In addition, through optimization, the low-pressure area at the leading edge of the outer wing section of the flying wing is improved, and the pressure difference drag of the flying wing is also reduced, which ultimately significantly reduces the drag of the aircraft. At the same time, from Fig. 9 , 10It can be seen that the pressure distribution of the optimized configuration at different cross-sectional positions is more uniform, and the airfoil thickness is also adjusted accordingly, which effectively improves the aerodynamic performance of the configuration and further reduces the drag, verifying the effectiveness of the optimization method.

[0116] pass Fig.11 , 12 The total pressure recovery coefficient comparison chart and the inlet Mach number cloud chart show that the optimized inlet has a lower total pressure recovery coefficient. σ The total pressure recovery coefficient before optimization was unevenly distributed, showing a large airflow loss, while the optimized intake duct maintained a high total pressure recovery coefficient in most areas, especially in the center area, and the airflow quality was significantly improved. In addition, further combined with Figures 13-16 It can be seen that the cross-sectional area of ​​the flying wing inlet is significantly increased, and the central axis moves upward. This change effectively improves the airflow separation at the lower surface of the second bend of the inlet and the boundary layer separation downstream, and increases the airflow Mach number at the inlet outlet, thereby significantly improving the total pressure recovery coefficient at the inlet outlet. Compared with the initial configuration, the total pressure recovery coefficient cloud diagram of the optimized configuration at the inlet outlet is more uniform and the distortion coefficient is smaller.

[0117] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A hybrid parameterized integrated gradient optimization method for internal and external flows of an aircraft, characterized in that: The method comprises at least the following steps when implemented: SS1. Hybrid parameterization of the initial configuration of the aircraft to be optimized An existing integrated flight vehicle is selected as the initial configuration to be optimized, and its aerodynamic shape is parameterized by using the FFD geometric parameterization method. The cross-sectional shape, along-the-path area law and central axis of the intake and exhaust system are parameterized by using a method based on two-dimensional shape fusion function and B-spline curve. SS2. Define the objective function and constraints of the optimization problem A certain flight state of the aircraft is selected as the optimization state, and design variables, objective functions and constraints related to the optimization of the internal and external flow integration of the initial configuration are given, wherein the design variables at least include aerodynamic shape design variables and intake and exhaust system design variables, the objective function at least includes an aircraft aerodynamic performance index and an intake and exhaust system flow quality index, and the constraints at least include geometric and / or performance constraints related to the aerodynamic shape and the intake and exhaust system; SS3. Deformation and update of the computational grid for the internal and external flow fields of the aircraft Based on the hybrid parameterization results of step SS1, the computational grids including the surface grid and the space grid of the internal and external flow fields of the aircraft are deformed and updated according to the changes of the FFD parameters and the geometric parameters of the intake and exhaust systems in the design variables, and in combination with the inverse distance weighting method; SS4. CFD numerical calculation of the internal and external flow fields of aircraft Based on the updated computational grid, the RANS equations and SA turbulence model are used to perform CFD numerical calculations of the internal and external flow fields of the aircraft to obtain the aerodynamic performance of the aircraft and the flow quality data of the intake and exhaust system; SS5. Objective function gradient information solution and iterative optimization The gradient information of the objective function relative to the design variables is solved based on the discrete adjoint method. The optimization problem is solved iteratively through the gradient optimization algorithm combined with the CFD calculation results and the obtained gradient information. The geometric characteristics of the aerodynamic shape and the intake and exhaust system are gradually optimized based on the iteratively adjusted design variables. The CFD calculation is repeated after each round of iteration until the objective function converges or reaches the predetermined optimization standard. The implementation of the discrete adjoint method includes: firstly constructing a discrete flow field control equation and its corresponding adjoint equation, then solving the discrete adjoint equation to obtain adjoint variables, and finally using the adjoint variables to calculate the gradient of the objective function to the design variables; when solving the adjoint equation, the same discrete format and solver as the original flow field equation are used, and a matrix-free calculation method is used in the gradient calculation process; The gradient optimization algorithm adopts a sequential quadratic programming method, and gradually approaches the optimal solution by solving quadratic programming sub-problems in each iteration; at the same time, in order to deal with nonlinear constraints, an augmented Lagrangian function is introduced to transform the constrained optimization problem into a series of unconstrained optimization sub-problems; and in the optimization process, a multi-starting point strategy is combined to start the optimization from different initial points to avoid falling into the local optimum; SS6. Verification and analysis of internal and external flow integrated design results Perform CFD numerical simulation on the optimized aircraft configuration to evaluate and verify its aerodynamic performance and flow quality of the intake and exhaust systems. If the verification results meet the expected goals, the optimization is completed; if not, adjust the design variables and repeat steps SS1 to SS6 until a satisfactory optimization result is obtained.

2. The method for optimizing the integrated gradient of internal and external flows of an aircraft based on hybrid parameterization according to claim 1, characterized in that: In step SS1, when the aerodynamic shape of the aircraft is subjected to FFD parameterization, the aerodynamic shape of the aircraft is embedded in a multi-level FFD control framework, each FFD control framework corresponds to a different local shape area of ​​the aerodynamic shape of the aircraft, and each FFD control frame establishes a functional relationship with the nodes on the aerodynamic shape through the Bernstein basis function, and the aerodynamic shape is adjusted by moving the FFD control points.

3. The method for optimizing the integrated gradient of internal and external flows of an aircraft based on hybrid parameterization according to claim 1, characterized in that: In step SS1, a method based on a two-dimensional shape fusion function is used to parameterize the cross-sectional shape of the intake duct and / or tail nozzle in the intake and exhaust system, which at least includes: First, extract the two-dimensional cross-sectional shapes at the inlet and outlet of the inlet and / or tail nozzle, discretize the cross-sectional shapes of the inlet and outlet into a set of point sets, and the two cross-sectional shapes have the same number of discrete points after discretization. Then, align the centroids of the two cross-sectional shapes through translation operation, where: ， in, f in , f out are the inlet and outlet cross-sectional shapes, respectively. , are the discrete point coordinates of the inlet and outlet cross-sectional shapes, d =1,2,…, n 1, n 1 is the preset number of discrete points; Next, the correspondence between the shapes is determined based on the geometric features of the two 2D cross-sectional shapes of the inlet and outlet, and the point-to-point correspondence between the two discrete point sets is established based on the key geometric features of the two cross-sectional shapes of the inlet and outlet. The shapes of the intermediate cross-sectional shapes are generated through the 2D shape fusion function: in, f in , f out Respectively represent the inlet and outlet cross-sectional shapes, f ( x i ) indicates the position of the air inlet and / or tail nozzle axis x i The cross-sectional shape at is the weight function to control the shape change; Finally, the weight function is applied Perform shape interpolation, where x i is the axis position of the air inlet and / or tail nozzle, x in is the inlet cross-sectional position, x out is the exit section position, α is the control parameter and α >0, by adjusting the parameters α Control the nonlinearity of the shape transition and construct a set of cross-sectional shapes that transition from the inlet to the outlet { f in … f ( x i )… f out }.

4. The method for integrated gradient optimization of internal and external flows of an aircraft based on hybrid parameterization according to claim 3 is characterized in that: In step SS1, the area law along the inlet duct and / or the tail nozzle is parameterized by using a B-spline curve method, including: First, the cross-sectional area values ​​of the inlet duct and / or tail nozzle at different flow direction sections are extracted and a point set {( x i , S i )},in S i is the position of the air inlet and / or tail nozzle axis x i The cross-sectional area at Next, from the point set {( x i , S i )} to extract key feature points that represent the main trend of area changes along the route and reflect the key change positions of the area, forming a new feature point set {( x j , S j )},in S j The critical change position of the area on the axis of the inlet and / or tail nozzle x j The cross-sectional area at Afterwards, the B-spline theory is used to extract the feature point set {( x j , S j )} to interpolate and obtain several benchmark control points along the area change P k and node vector U 1, where P k =( x k , S k ), S k Axis position x k The cross-sectional area at the node vector U 1 is used to determine the segmentation and interpolation interval of the curve and is defined as ,in x min , x max are the starting and ending positions of the inlet and / or tail nozzle axis, p is the order of the B-spline curve, the reference control point P k and node vector U 1 satisfies the following B-spline curve equation: in, S ( x ) indicates the axis position x The cross-sectional area at N k,p ( x ) is the k Control Points P k About the axis position x of p Sub-B-spline basis function; Finally, a new design variable Δ is introduced S k , and obtain a new set of control points {( x k , S k +Δ S k )}, and generate a new along-the-way area distribution curve by adjusting the position of the control points: Among them, the design variable Δ S k Satisfy ΔS k_min ≤ Δ S k ≤ΔS k_max , Δ S k_min , Δ S k_max The value is set according to actual project requirements.

5. The method for integrated gradient optimization of internal and external flows of an aircraft based on hybrid parameterization according to claim 4, characterized in that: In step SS1, the central axis of the air inlet and / or the tail nozzle is parameterized using a B-spline curve method, including: First, based on the geometric structure of the air inlet and / or tail nozzle, the center points at different flow sections are extracted to form a center point set {( x l , y l , z l )},in( x l , y l , z l ) is the cross-sectional position on the axis of the air inlet and / or tail nozzle x l The spatial coordinates of the center point at ; Next, from the center point set {( x l , y l , z l )} to extract the key feature points that represent the main change trend of the central axis and form a new feature point set {( x m , y m , z m )}, feature point( x m , y m , z m ) is the cross-sectional position on the axis of the air inlet and / or tail nozzle x m The spatial coordinates of Afterwards, the B-spline theory is used to extract the feature point set {( x m , y m , z m )} to interpolate and generate the B-spline curve of the central axis, and obtain the reference control point set {( x n , y n , z n )}, where each benchmark control point ( x n , y n , z n ) is used to control the geometric shape of the B-spline curve and to select the node vector U 2 to control the segmentation and interpolation interval of the B-spline curve; Finally, a new design variable (Δ y n ,Δ z n ), and obtain a new set of control points by adjusting the positions of the control points {( x n , y n +Δ y n , z n +Δ z n )}, generate a new B-spline curve about the central axis: in, C new ( x ) indicates the position of the air inlet and / or tail nozzle axis x The spatial position of N n,p ( x ) is the n The position of the control points about the axis x of p Sub-B-spline basis functions.

6. The method for integrated gradient optimization of internal and external flows of an aircraft based on hybrid parameterization according to claim 1, characterized in that: In step SS2, the key aerodynamic parameters set in the optimization state include at least the incoming flow Mach number Ma, the airflow angle of attack AoA, the lift coefficient C L and / or drag coefficient C D The aerodynamic shape design variables are selected FFD control point parameters related to the geometric parameters of the aerodynamic shape of the aircraft, and the intake and exhaust system design variables are geometric parameters related to the cross-sectional shape, along-the-path area law and central axis of the intake and exhaust system; The aerodynamic performance index of the aircraft in the objective function at least includes the minimum drag coefficient C under the optimization state. D , maximize the lift coefficient C L , maximize lift-to-drag ratio C L / C D One or more of the following, the flow quality index of the intake and exhaust system at least includes the intake duct total pressure recovery coefficient under the optimized state σ , Inlet distortion coefficient σ d One or more of; the constraints include at least one or more of geometric size constraints, aerodynamic performance constraints, pitch moment constraints related to the aerodynamic shape of the aircraft, and flow stability constraints related to the intake and exhaust system.

7. The method for integrated gradient optimization of internal and external flows of an aircraft based on hybrid parameterization according to claim 5, characterized in that: In step SS3, based on the parameterized results of the cross-sectional shape, the area law along the path, and the central axis of the intake duct and / or the tail nozzle in the intake and exhaust system completed in step SS1, the profile of the intake duct and / or the tail nozzle is formed by combining the parameterized results, and then the grid coordinates related to the profile of the intake duct and / or the tail nozzle in the calculation grid are updated according to the change of the design variables: First, import the computational mesh of the inlet and / or tail nozzle G and simplified surface meshes of inlet and / or nozzle profiles He , from the computational grid by keyword extraction method G Extract the grid coordinate set related to the inlet and / or tail nozzle profile H , and establish a grid coordinate set H With simplified surface mesh He The mapping relationship between H = L ( He ),in L is the mapping function; Secondly, based on the simplified surface mesh He Extract the cross-sectional shapes at the inlet and / or the tail nozzle at the inlet and the outlet, and construct a cross-sectional shape set from the inlet to the outlet according to the parameterized results of the cross-sectional shapes of the inlet and / or the tail nozzle. f in … f ( x i )… f out }, then according to the position of each axis x i The cross-sectional shape at f ( x i ) extracts its corresponding center point ( x i , y i , z i ) and area S i ; Next, according to the changes in the design variables, the area law along the path and the B-spline curve equation of the central axis after parameterization in step SS1 are optimized and adjusted respectively. Then, according to the new area law and the B-spline curve equation of the central axis, the position of each axis is optimized. x i The corresponding target center point positions are obtained by interpolation calculation ( x i , y ' i , z ' i ) and target area A i ; Afterwards, the target center point position ( x i , y ' i , z ' i ) and target area A i , for each axis position x i The original cross-sectional shape at f ( x i ), with its original center point ( x i , y i , z i ) as the center to scale the area and obtain the cross-sectional area that conforms to the new area law distribution. A i , the zoom ratio is , the cross-sectional shape after scaling is , and the scaled cross-sectional shape Move to the target center point ( x i , y ' i , z ' i ) to generate an updated simplified surface mesh He' ; at last , through the mapping relationship H = L ( He ), the updated grid He' Mapped to the computational grid.

8. The method for optimizing the integrated gradient of internal and external flows of an aircraft based on hybrid parameterization according to claim 1, characterized in that: In step SS4, the RANS equations are discretized using the finite volume method, the spatial discretization uses the second-order upwind scheme, the time discretization uses the implicit double time stepping method, and the static pressure boundary condition is used at the inlet outlet, the total temperature and total pressure boundary conditions are used at the tail nozzle inlet, and the velocity boundary condition is used in the far field.

9. The method for optimizing the integrated gradient of internal and external flows of an aircraft based on hybrid parameterization according to claim 1, characterized in that: In step SS5, the gradient information of the objective function relative to the design variables is obtained by decomposing the geometric design variables of the aircraft aerodynamic shape and the intake and exhaust system: First, the objective function I For geometric design variables Y Gradient information dI / d Y Decompose into objective function I Aerodynamic shape design variables Y Gradient of 1 dI / d Y 1 and the objective function I Design variables for intake and exhaust systems Y Gradient 2 dI / d Y 2, expressed as ; Secondly, the optimization objective is calculated with respect to the surface coordinate variables by the adjoint method X Gradient information dI / dX , where the surface coordinate variables X Contains the geometric grid coordinates corresponding to the aerodynamic shape of the aircraft X 1 and the geometric grid coordinates corresponding to the intake and exhaust system X 2. The transmission relationship of gradient information is described as: ， in, dI / dX 1 and dI / dX 2 by dI / dX Decomposed into, dX 1 / d Y 1 and dX 2 / d Y 2 are respectively based on the geometric parameterization of the aircraft aerodynamic shape and the intake and exhaust system and are analytically derived using the chain rule.

Images