Aircraft tail retreating separation grid force measurement pneumatic modeling sample compression method and aircraft tail retreating separation grid force measurement pneumatic modeling sample compression system

Abstract

The invention provides a method and a system for compressing a force-measuring pneumatic modeling sample of an aircraft tail retreating separation grid, and the method comprises the steps: S1, solving a separation track of a separation body based on a separation state range given by a task; s2, based on the separation track, determining a position beam and an attitude beam of the separated body, and further selecting a sampling range of pneumatic modeling of the separated body; and S3, arranging sample points on the basis of the sampling range, complementing the data of the sample points which are not arranged, and constructing a pneumatic model of the separation body. According to the sample point arrangement provided by the invention, the distribution range constraint of the separation trajectory is considered, and the effectiveness and pertinence of sample arrangement are improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of aerodynamics, and in particular, relates to an aircraft tail separation grid force measurement aerodynamic modeling sample compression method and system. More specifically, it is a tail separation grid force measurement aerodynamic modeling sample compression method based on separation trajectory range sampling. BACKGROUND

[0002] During the tail separation process of the aircraft, the separated body is disturbed by the high-speed flight platform, and its aerodynamic characteristics are quite different from those under free flow conditions.

[0003] Under free flow conditions, the aerodynamic characteristics of the separated body are only related to its own state, i.e. height, Mach number, angle of attack, side slip angle, and rudder deflection angle of the three channels. During tail separation, the aerodynamic characteristics of the separated body are not only related to its own state, but also related to the relative position and attitude between the separated body and the high-speed flight platform.

[0004] To describe the aerodynamic characteristics of the separated body during tail separation, a grid force measurement method based on complete orthogonal sampling is used for aerodynamic modeling. Due to the numerous influencing factors during tail separation, the number of samples required for modeling is at least 3 N where N is the number of influencing factors considered in modeling, and 3 represents the number of values for each factor; when N is 6, the number of samples is 729; when N is 7, the number of samples is 2187. Such a large number of experiments is a huge challenge in terms of cost and period.

[0005] Therefore, it is necessary to improve the tail separation aerodynamic modeling method to reduce the number of modeling samples as much as possible under the premise of ensuring accurate aerodynamic description and reduce the modeling cost.

[0006] Patent document CN118862428A discloses a tail separation process numerical simulation method with slip constraint, which includes: introducing a multi-body transfer matrix method into the numerical simulation process of tail separation, accurately simulating the multi-body dynamics motion process with slip constraint, improving the simulation accuracy of the attitude and speed of the carrier and the separated body during separation, and further providing more accurate initial conditions for the subsequent simulation of the process without geometric constraints, thereby improving the simulation accuracy and reliability of the whole tail separation process. Compared with the traditional numerical simulation method that ignores the out-of-cabin stage, the present application provides the numerical simulation capability of the whole tail separation process. Compared with the simulation method that ignores the slip hinge constraint during the out-of-cabin process and uses an approximate constraint, the transfer matrix method for solving the system total transfer matrix equation only increases a small amount of time and memory occupation, but can significantly improve the accuracy of motion simulation. However, this scheme cannot reduce the number of modeling samples and reduce the modeling cost.

[0007] This problem needs to be solved urgently. SUMMARY

[0008] In view of the defects in the prior art, the present application aims to provide an aircraft tail separation grid force measurement aerodynamic modeling sample reduction method and system.

[0009] According to the present application, an aircraft tail separation grid force measurement aerodynamic modeling sample compression method is provided, comprising:

[0010] Step S1: based on the separation state range given by the task, the separation trajectory of the separation body is obtained;

[0011] Step S2: based on the separation trajectory, the position bundle and attitude bundle of the separation body are determined, and then the sampling range of the aerodynamic modeling of the separation body is selected;

[0012] Step S3: based on the sampling range, sample points are arranged, and then data is completed for sample points not arranged, and an aerodynamic model of the separation body is constructed.

[0013] Preferably, in the step S1, the separation state range comprises separation Mach number, height, angle of attack, side slip angle and angular velocity;

[0014] The separation trajectory of the separation body is obtained by CFD software, and the mathematical expression is:

[0015] y t =f(Ma,H,alpha,beta,u0,v0,w0,wx0,wy0,wz0,x0,y0,z0,θ0,ψ0,γ0,t)

[0016] Wherein, y t represents the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of the separation body at time t; f represents a function; Ma is the separation Mach number, H is the separation height, alpha is the separation angle of attack, beta is the separation side slip angle, u0, v0 and w0 are the axial, vertical and lateral velocities of the separation body at the initial separation time, wx0, wy0 and wz0 are the axial, vertical and lateral angular velocities of the separation body at the initial separation time, x0, y0 and z0 are the position coordinates of the separation body at the initial separation time, θ0, ψ0 and γ0 are the pitch attitude angle, yaw attitude angle and roll attitude angle of the separation body at the initial separation time, and t is the time.

[0017] Preferably, in the step S2, the number of position variables is confirmed, the position bundle and attitude bundle obtained by the curve of the separation trajectory are used to obtain the maximum and minimum values of each position variable and the corresponding attitude angle, and a sampling range is formed; the sampling range is (K*y max , K*y min , K*z max , K*z min , K*θ max , K*θ minK * ψ max K * ψ min , wherein K is a safety factor, * represents multiplication, subscript max represents maximum value, min represents minimum value; θ represents pitch attitude angle, and ψ represents yaw attitude angle.

[0018] Preferably, in the step S3, based on the sampling range, sample points are arranged according to an “×” sample compression algorithm, then data of sample points not arranged are recovered and sample points are interpolated through a data completion algorithm, and an aerodynamic model of the separation body is constructed.

[0019] The sample points arranged according to the “×” sample compression algorithm are provided, wherein variables of the sample points include A and B; A is taken as A1, A2 and A3; B is taken as B1, B2 and B3, (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2) are obtained; wherein A and B are of the same variable type, and are any one of position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of the separation body.

[0020] Preferably, in the step S3, the data completion algorithm is mathematically expressed as:

[0021] CN(θ1, ψ2) = (CN(θ1, ψ3) - CN(θ1, ψ1)) / (ψ3 - ψ1) * (ψ2 - ψ1) + CN(θ1, ψ1)

[0022] CN(θ2, ψ1) = (CN(θ3, ψ1) - CN(θ1, ψ1)) / (θ3 - θ1) * (θ2 - θ1) + CN(θ1, ψ1)

[0023] CN(θ3, ψ2) = (CN(θ3, ψ3) - CN(θ3, ψ1)) / (ψ3 - ψ1) * (ψ2 - ψ1) + CN(θ3, ψ1)

[0024] CN(θ2, ψ3) = (CN(θ3, ψ3) - CN(θ1, ψ3)) / (θ3 - θ1) * (θ2 - θ1) + CN(θ1, ψ3)

[0025] Wherein CN is a quantity to be completed; θ1, θ2 and θ3 respectively represent pitch attitude angles of sample point one, sample point two and sample point three; ψ1, ψ2 and ψ3 respectively represent yaw attitude angles of sample point one, sample point two and sample point three.

[0026] The interpolation of sample points comprises:

[0027] Step A1: judging whether a variable to be interpolated is out of a sample range on left and right end surfaces of a to-be-predicted x i position, and the result is yes for amplitude limiting processing, and the result is no for linear interpolation; wherein xi Position is the i-th position variable;

[0028] Step A2: judging x i whether the position is beyond the axial maximum sample range, if yes, linear interpolation is performed by using the linear decay transition to zero method, if no, linear interpolation is performed;

[0029] The mathematical expression of the limiting amplitude processing is:

[0030] CNleft (θ i , ψ i ) = CNleft (θ max / θ min , ψ max / ψ min )

[0031] CNright (θ i , ψ i ) = CNright (θ max / θ min , ψ max / ψ min )

[0032] Wherein, θ i is the i-th sample point's pitch attitude angle; ψ i is the i-th sample point's yaw attitude angle; CNleft represents the normal force coefficient on the left end face of the x i position to be predicted; CNright represents the normal force coefficient on the right end face of the x i position to be predicted;

[0033] The mathematical expression of the linear interpolation by using the linear decay transition to zero method is:

[0034] CN (x i ) = -CN (x max ) / x ∞ -x max ) * (x i -x max ) + CN (x max )

[0035] Wherein, x max is the transverse maximum sample point position; x ∞ is the position where the aerodynamic interference tends to zero, which is a preset threshold value set according to the task requirement.

[0036] According to the aircraft tail-retreating separation grid force-measuring aerodynamic modeling sample compression system provided by the application, the following steps are included:

[0037] Module M1: based on the separation state range given by the task, the separation trajectory of the separation body is obtained;

[0038] Module M2: determining the position bundle and attitude bundle of the separated body based on the separation trajectory, and then selecting the sampling range of the aerodynamic modeling of the separated body;

[0039] Module M3: arranging sample points based on the sampling range, then data completing the unarranged sample point data, and constructing the aerodynamic model of the separated body.

[0040] Preferably, in the module M1, the separation state range includes separation Mach number, height, angle of attack, side slip angle and angular velocity.

[0041] The separation trajectory of the separated body is obtained by CFD software, and the mathematical expression is:

[0042] y t =f(Ma,H,alpha,beta,u0,v0,w0,wx0,wy0,wz0,x0,y0,z0,θ0,ψ0,γ0,t)

[0043] Wherein, y t represents the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of the separated body at time t; f represents function; Ma is separation Mach number, H is separation height, alpha is separation angle of attack, beta is separation side slip angle, u0, v0 and w0 are respectively the axial, vertical and lateral velocities of the separated body at the initial time of separation, wx0, wy0 and wz0 are the axial, vertical and lateral angular velocities of the separated body at the initial time of separation, x0, y0 and z0 are the position coordinates of the separated body at the initial time of separation, θ0, ψ0 and γ0 are the pitch attitude angle, yaw attitude angle and roll attitude angle of the separated body at the initial time of separation, and t is time.

[0044] Preferably, in the module M2, the number of position variables is confirmed, the position bundle and attitude bundle obtained through the curve of the separation trajectory are used to obtain the maximum and minimum values of each position variable and the corresponding attitude angle, and the sampling range is formed; the sampling range is (K*y max , K*y min , K*z max , K*z min , K*θ max , K*θ min , K*ψ max , K*ψ min ), wherein K is safety factor, * represents product, max represents maximum value, and min represents minimum value; θ represents pitch attitude angle, and ψ represents yaw attitude angle.

[0045] Preferably, in the module M3, sample points are arranged based on the sampling range according to the "×" sample compression algorithm, and then the unarranged sample point data is recovered through data completion algorithm to construct the aerodynamic model of the separated body.

[0046] The sample points are arranged according to the "X" sample compression algorithm, and the variables of the sample points include A and B; A takes A1, A2, A3; B takes B1, B2, B3, and (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2) are obtained; wherein, A and B have the same variable type, which is any one of the position, attitude angle, speed, angular velocity, aerodynamic force or aerodynamic moment of a separated body.

[0047] Preferably, in the module M3, the data completion algorithm has a mathematical expression as follows:

[0048] CN(θ1,ψ2)=(CN(θ1,ψ3)-CN(θ1,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ1,ψ1)

[0049] CN(θ2,ψ1)=(CN(θ3,ψ1)-CN(θ1,ψ1)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ1)

[0050] CN(θ3,ψ2)=(CN(θ3,ψ3)-CN(θ3,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ3,ψ1)

[0051] CN(θ2,ψ3)=(CN(θ3,ψ3)-CN(θ1,ψ3)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ3)

[0052] Wherein, CN is the quantity to be completed; θ1, θ2 and θ3 respectively represent the pitch attitude angle of the sample point one, the sample point two and the sample point three; ψ1, ψ2 and ψ3 respectively represent the yaw attitude angle of the sample point one, the sample point two and the sample point three;

[0053] The interpolation sample point comprises:

[0054] Module A1: judging whether the variable to be interpolated is out of the sample range at the x i position, the result is yes, and the amplitude limiting processing is performed; the result is no, and the linear interpolation is performed; wherein, x i position is the i-th position variable;

[0055] Module A2: judging whether the x i position is out of the maximum sample range in the axial direction, the result is yes, and the linear interpolation is performed by using the linear decay transition to zero method; the result is no, and the linear interpolation is performed;

[0056] The mathematical expression of the amplitude limiting processing is as follows:

[0057] CNleft(θ i ,ψi )=CN left(θ max / θ min ,ψ max / ψ min )

[0058] CN right(θ) i ,ψ i ) = CN right (θ max / θ min ,ψ max / ψ min )

[0059] Where, θ i Let ψ be the pitch attitude angle of the i-th sample point; i Let be the yaw attitude angle of the i-th sample point; CN left represents the x to be predicted. i The normal force coefficient on the left end face; CN right represents the x to be predicted. i Normal force coefficient on the right end face;

[0060] The method for linear decay transitioning to zero involves linear interpolation, and its mathematical expression is:

[0061] CN(x i )=-CN(x max ) / x ∞ -x max )*(x i -x max )+CN(x max )

[0062] Where, x max x represents the location of the largest sample point in the horizontal direction. ∞ The position where aerodynamic interference approaches zero is a preset threshold set according to mission requirements.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. The sample point arrangement provided by this invention takes into account the distribution range constraints of the separation trajectory, thereby improving the effectiveness and relevance of the sample arrangement.

[0065] 2. The advantage of this invention is that the sample point arrangement method takes into account the distribution range constraint of the separation trajectory, which greatly improves the effectiveness and pertinence of the sample arrangement, and can significantly reduce the number of sample arrangements and reduce the modeling cost.

[0066] 3. This invention further reduces the number of orthogonal samples by using the "×" sample compression algorithm, thereby improving modeling efficiency. Attached Figure Description

[0067] Other features, objects, and advantages of the application will become more apparent from the following detailed description when read in conjunction with the accompanying drawings:

[0068] Figure 1 An effect diagram of drawing a trajectory bundle provided by the application, wherein Xa represents an axial displacement of a motion, and Ya represents a vertical displacement of the motion;

[0069] Figure 2 An "X" type sample arrangement diagram provided by the application;

[0070] Figure 3 A comparison diagram of a predicted result of a reduced modeling sample and a verification data curve of a tail separation aerodynamic model provided by the application, wherein CN_GM represents the predicted result, CN_test represents the verification data, and CN represents an aircraft normal force coefficient. DETAILED DESCRIPTION

[0071] The application will be described in detail below with specific embodiments. The following embodiments will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the application. These are within the scope of protection of the application.

[0072] In order to reduce the number of modeling samples, the application provides a kind of aircraft tail separation grid force aerodynamic modeling sample compression method based on separation trajectory range sampling. First, the sample interval of tail separation aerodynamic modeling is obtained from the separation trajectory obtained by numerical calculation by CFD method, i.e. Computational Fluid Dynamics method, to reduce the sampling range of grid force sample as a constraint, to improve the effectiveness of sample arrangement. Then, the "X" type sample compression algorithm is used to further reduce the number of orthogonal samples. The number of modeling samples is reduced through the above two steps.

[0073] According to the aircraft tail separation grid force aerodynamic modeling sample compression method provided by the application, comprising:

[0074] Step S1: based on the separation state range given by the task, the separation trajectory of the separation body is obtained;

[0075] Step S2: based on the separation trajectory, the position bundle and the attitude bundle of the separation body are determined, and then the sampling range of the aerodynamic modeling of the separation body is selected;

[0076] Step S3: based on the sampling range, sample points are arranged according to the "X" sample compression algorithm, then the data of the sample points not arranged are recovered through the data completion algorithm, and then the aerodynamic model of the separation body is constructed.

[0077] In the step S1, the separation state range includes: separation Mach number, height, angle of attack, side slip angle and angular velocity;

[0078] The separation trajectory of the separation body is obtained by CFD software, and the mathematical expression is:

[0079] y t =f(Ma,H,alpha,beta,u0,v0,w0,wx0,wy0,wz0,x0,y0,z0,θ0,ψ0,γ0,t)

[0080] Wherein, y t represents the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of the separation body at time t; f represents a function; Ma is the separation Mach number, H is the separation height, alpha is the separation angle of attack, beta is the separation side slip angle, u0, v0 and w0 are the axial, vertical and lateral velocities of the separation body at the initial separation time, wx0, wy0 and wz0 are the axial, vertical and lateral angular velocities of the separation body at the initial separation time, x0, y0 and z0 are the position coordinates of the separation body at the initial separation time, θ0, ψ0 and γ0 are the pitch attitude angle, yaw attitude angle and roll attitude angle of the separation body at the initial separation time, and t is the time.

[0081] Specifically, in the step S2, the positions (x t , y t , z t ), attitude angles (θ t , ψ t ) on the separation trajectory are plotted into curves according to time to obtain a position beam and an attitude beam.

[0082] Wherein, x t , y t , z t are the position coordinates of the separation body at time t; θ t , ψ t are the pitch attitude angle and yaw attitude angle of the separation body at time t, and t is the time.

[0083] For aerodynamic modeling, extrapolation is generally not allowed, therefore, the range of modeling states needs to be determined first, and the position beam and the attitude beam reflect the possible range of each state in the separation motion process of the separation body; specifically, the sampling range of the separation body is selected based on the position beam and the attitude beam:

[0084] Firstly, the number N (x1, x2, … x i ..., x N ) of x variables is determined, then the maximum and minimum values of the position and attitude angle corresponding to each x i are obtained by using the position beam and the trajectory beam curve (y max , ymin ,z max ,z min ,θ max ,θ min ,ψ max ,ψ min ), i.e. the sampling range. x i corresponds to the x position to be predicted. The x variable is the position.

[0085] In particular, considering that the CFD calculation results have errors and the finiteness of the calculation state, each x i The corresponding position and attitude angle range is enlarged to ensure that the trajectory under other unknown states does not exceed this range, i.e. (K*y max , K*y min , K*z max , K*z min , K*θ max , K*θ min , K*ψ max , K*ψ min ), where K is a safety factor, * represents multiplication, max represents maximum, and min represents minimum; θ represents the pitch attitude angle, and ψ represents the yaw attitude angle.

[0086] In the step S3, the "×" sample compression algorithm, specifically:

[0087] When arranging sample points by the complete orthogonal method, taking the attitude angle as an example, the θ and ψ attitude variables take θ1, θ2, θ3 and ψ1, ψ2, ψ3, and a total of 9 (3x3) sample points are needed to be arranged.

[0088] The "×" sample compression algorithm only needs to arrange 4 corner points and 1 center point, i.e. (θ1, ψ1), (θ1, ψ3), (θ3, ψ1), (θ3, ψ3), (θ2, ψ2).

[0089] Compared with the complete orthogonal method, the number of samples is reduced by 4 / 9, and the more modeling variables, the more obvious the sample reduction.

[0090] The data completion algorithm, taking the normal force coefficient as an example, needs to restore the data of the following four sample points, and the mathematical expression is:

[0091] CN(θ1, ψ2) = (CN(θ1, ψ3) - CN(θ1, ψ1)) / (ψ3 - ψ1) * (ψ2 - ψ1) + CN(θ1, ψ1)

[0092] CN(θ2, ψ1) = (CN(θ3, ψ1) - CN(θ1, ψ1)) / (θ3 - θ1) * (θ2 - θ1) + CN(θ1, ψ1)

[0093] CN(θ3, ψ2) = (CN(θ3, ψ3) - CN(θ3, ψ1)) / (ψ3-ψ1)*(ψ2-ψ1) + CN(θ3, ψ1)

[0094] CN(θ2, ψ3) = (CN(θ3, ψ3) - CN(θ1, ψ3)) / (θ3-θ1)*(θ2-θ1) + CN(θ1, ψ3)

[0095] wherein CN is the to-be-completed quantity; θ1, θ2 and θ3 respectively represent the pitch attitude angle of the sample point one, the sample point two and the sample point three; ψ1, ψ2 and ψ3 respectively represent the yaw attitude angle of the sample point one, the sample point two and the sample point three;

[0096] In the step S3, the x-direction interpolation sample arrangement point comprises:

[0097] Step A1: judging whether the to-be-interpolated variable is out of the sample range at the to-be-predicted x i position left and right end faces, the result is yes, and the amplitude limiting processing is performed; the result is no, and the normal linear interpolation is performed;

[0098] CNleft(θ i , ψ i ) = CNleft(θ max / θ min , ψ max / ψ min )

[0099] CNright(θ i , ψ i ) = CNright(θ max / θ min , ψ max / ψ min )

[0100] wherein θ i is the pitch attitude angle of the i-th sample point; ψ i is the yaw attitude angle of the i-th sample point; CNleft represents the normal force coefficient on the left end face of the to-be-predicted x i position; CNright represents the normal force coefficient on the right end face of the to-be-predicted x i position;

[0101] Step A2: judging whether the x i position is out of the axial maximum sample range, the result is yes, and the linear interpolation is performed by using the linear decay transition to zero method; the result is no, and the normal linear interpolation is performed;

[0102] In the step A2, the linear interpolation is performed by using the linear decay transition to zero method, and the mathematical expression is:

[0103] CN(x i)=-CN(x max ) / x ∞ -x max )*(x i -x max )+CN(x max )

[0104] wherein, x max is the maximum sample point position in the lateral direction; x ∞ is the position at which the aerodynamic interference tends to zero, and is a preset threshold value set according to the task requirement.

[0105] In other words, x max is the maximum sample point position in the x direction, i.e. the lateral direction; x i is the i-th position variable; in other words, x i is the x position to be predicted; and CN is the normal force coefficient.

[0106] According to the aircraft tail-retreating separation grid aerodynamic modeling sample reduction method based on separation trajectory range sampling provided by the application, the method comprises the following steps:

[0107] Step 1, calculating a separation trajectory.

[0108] According to the separation state range given by the whole, CFD numerical calculation is carried out to obtain the separation trajectory of the separation body with time, i.e. the position and attitude, under different separation states.

[0109] Step 2, selecting modeling variables.

[0110] The angle of attack, the sideslip angle, the position (x, y, z) and the attitude angle (θ, ψ) are selected as the modeling variables.

[0111] Step 3, drawing a trajectory bundle.

[0112] The position (x, y, z) and the attitude angle (θ, ψ) on the separation trajectory are plotted into curves according to time to obtain a position bundle and an attitude bundle.

[0113] Step 4, determining variable sampling range.

[0114] 5-9 time points are selected according to the requirement, i.e. 5-9 x positions, and then the sampling range of the separation body at different time points is determined through the position bundle and the attitude bundle; in addition, considering the error of the CFD numerical calculation, the sampling range is appropriately enlarged by 1.2-1.5 times to ensure that the separation trajectory under the new state still falls within the range of the trajectory bundle.

[0115] Step 5, arranging a starting point sample.

[0116] For the separation starting point, i.e. the first time point, since the relative position and the relative attitude of the child and the parent are the same at each separation initial time point, 1 modeling sample point is arranged.

[0117] Step 6, "X" sample compression algorithm.

[0118] For each subsequent time, the y, z position variables and the θ, ψ attitude variables, the "X" sample compression algorithm is used to arrange the modeling sample points. Specifically, when arranging sample points using the full orthogonal method, the y, z position variables take y1, y2, y3 and z1, z2, z3, which, when combined, require the arrangement of 9 (3x3) sample points; the θ, ψ attitude variables take θ1, θ2, θ3 and ψ1, ψ2, ψ3, which, when combined, also require the arrangement of 9 (3x3) sample points. The "X" sample compression algorithm only arranges 4 corner points and 1 center point, i.e. (y1, z1), (y1, z3), (y3, z1), (y3, z3), (y2, z2), compared with the full orthogonal method, the number of samples is reduced by 4 / 9, and the more modeling variables, the more obvious the sample reduction.

[0119] In other words, the sample points arranged according to the "X" sample compression algorithm, assuming that the variables of the sample points include A and B; let A take A1, A2, A3; let B take B1, B2, B3, to get (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2); wherein A and B are of the same type of variable, being any one of the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of a separated body.

[0120] Step 7, data completion.

[0121] For states that are not arranged with samples, linear interpolation is used to complete the data. Specifically, for (y2, z2), linear interpolation is used to complete (y1, z1) and (y1, z3). Other states that are not arranged with samples are completed using the same method.

[0122] After completion, bilinear interpolation can be used for interpolation prediction, which is the same as the traditional grid force measurement modeling method based on full orthogonal sampling.

[0123] Step 8, x-direction interpolation.

[0124] Because the sample arrangement ranges of the previous and subsequent time are different, the sample range size gradually increases along the x-direction, forming a horn shape, which cannot be directly interpolated in multiple dimensions.

[0125] Therefore, for any x-direction position, such as x0, first determine the x-direction interval to which the position x0 belongs, then separately interpolate on both sides, obtain the interpolation results on both sides, and then perform linear interpolation in the x-direction to obtain the aerodynamic parameters of the separated body that need to be predicted.

[0126] Specifically, in step 1, the separation state range refers to the separation Mach number, altitude, angle of attack, side slip angle, and angular velocity, etc.

[0127] Specifically, in step 2, the selection of modeling variables is obtained by analyzing the influence of a certain variable on the separation interference aerodynamic characteristics, and for the variable with significant influence, the variable should be selected as the modeling variable, otherwise, the variable is not selected as the modeling variable.

[0128] Specifically, the direct logical association of selecting the modeling variable and drawing the time curve, solving the position beam and the attitude beam is that the trajectory is the change of the position and the attitude of the separation body with time, and the position and the attitude determine the force and the torque borne by the separation body. Obtaining the position beam and the attitude beam means obtaining the force and the torque borne by the separation body, and the force and the torque can be used to calculate the trajectory.

[0129] Specifically, the significant influence is that the influence is greater than 10% to 20%; in practice, further adjustment can be made according to the specific task requirement.

[0130] Specifically, in step 3, the position beam and the trajectory beam refer to drawing the displacement and the attitude angle of the separation body in the x, y and z directions along time into a curve under different separation initial conditions.

[0131] Specifically, in step 4, 5 to 9 time points are selected according to the requirement, that is, 5 to 9 x-direction positions, and the 5 to 9 positions should cover the starting point and the ending point of the separation test to avoid model extrapolation; the sampling range is appropriately enlarged by 1.2 to 1.5 times, that is, when determining the sampling boundary of each variable, a certain margin is reserved to ensure that the separation trajectory under the new state still falls within the range of the trajectory beam. In particular, when only enlarging by 1.0 times, it indicates that the sampling range is the same as the upper and lower boundary range formed by the trajectory beam.

[0132] Specifically, in step 5, the more the modeling variables, the more obvious the sample reduction, for example, when considering the attitude angle, the number of samples is reduced from 81 to 25.

[0133] Specifically, in step 8, when the variable to be interpolated exceeds the sample range on a certain end face, an amplitude limiting method is used for processing to avoid model extrapolation. It should be noted that in the x direction, when the variable exceeds the maximum x sample range, no amplitude limiting processing is performed, but a linear decay transition to zero method is used for linear interpolation to meet the objective law that the farther the separation distance, the smaller the aerodynamic interference.

[0134] The advantage of the present application lies in that the sample point arrangement of the method considers the distribution range constraint of the separation trajectory, greatly improves the effectiveness and pertinence of the sample arrangement, and the number of sample arrangements can be significantly reduced by using the method, and the modeling cost is reduced.

[0135] The application further provides an aircraft tail-retreating separation grid force measurement aerodynamic modeling sample compression system, which can be realized by performing the flow steps of the aircraft tail-retreating separation grid force measurement aerodynamic modeling sample compression method, i.e., the aircraft tail-retreating separation grid force measurement aerodynamic modeling sample compression method can be understood as the preferred implementation of the aircraft tail-retreating separation grid force measurement aerodynamic modeling sample compression system by those skilled in the art.

[0136] According to the application, an aircraft tail-retreating separation grid force measurement aerodynamic modeling sample compression system is provided, comprising:

[0137] Module M1: based on the separation state range given by the task, the separation trajectory of the separation body is obtained;

[0138] Module M2: based on the separation trajectory, the position bundle and the attitude bundle of the separation body are determined, and then the sampling range of the aerodynamic modeling of the separation body is selected;

[0139] Module M3: based on the sampling range, sample points are arranged, then the data of the sample points not arranged is completed, and the aerodynamic model of the separation body is constructed.

[0140] Specifically, in the module M1, the separation state range comprises a separation Mach number, a height, an attack angle, a sideslip angle and an angular velocity.

[0141] The separation trajectory of the separation body is obtained by CFD software, and the mathematical expression is:

[0142] y t =f(Ma, H, alpha, beta, u0, v0, w0, wx0, wy0, wz0, x0, y0, z0, theta0, psi0, gamma0, t)

[0143] wherein y t represents the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic moment of the separation body at time t; f represents a function; Ma is the separation Mach number, H is the separation height, alpha is the separation attack angle, beta is the separation sideslip angle, u0, v0 and w0 are respectively the axial, vertical and lateral velocities of the separation body at the initial separation time, wx0, wy0 and wz0 are respectively the axial, vertical and lateral angular velocities of the separation body at the initial separation time, x0, y0 and z0 are the position coordinates of the separation body at the initial separation time, theta0, psi0 and gamma0 are respectively the pitch attitude angle, yaw attitude angle and roll attitude angle of the separation body at the initial separation time, and t is the time.

[0144] Specifically, in the module M2, the number of position variables is confirmed, the position bundle and the attitude bundle obtained by the curve of the separated trajectory are acquired, the maximum value and the minimum value of each position variable and the corresponding attitude angle are obtained, and a sampling range is formed; the sampling range is (K*y max ,K*y min ,K*z max ,K*z min ,K*θ max ,K*θ min ,K*ψ max ,K*ψ min ), wherein K is a safety coefficient, * represents multiplication, max represents the maximum value, min represents the minimum value, θ represents the pitch attitude angle, and ψ represents the yaw attitude angle.

[0145] Specifically, in the module M3, based on the sampling range, sample points are arranged according to an “×” sample compression algorithm, and then data of sample points not arranged are recovered through a data completion algorithm, so as to construct an aerodynamic model of the separated body.

[0146] The sample points are arranged according to the “×” sample compression algorithm, wherein variables of the sample points include A and B; A is taken as A1, A2 and A3; B is taken as B1, B2 and B3, so as to obtain (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2); wherein A and B have the same variable type, which is any one of the position, the attitude angle, the speed, the angular velocity, the aerodynamic force or the aerodynamic moment of the separated body.

[0147] Specifically, in the module M3, the data completion algorithm has a mathematical expression as follows:

[0148] CN (θ1, ψ2) = (CN (θ1, ψ3) - CN (θ1, ψ1)) / (ψ3-ψ1) * (ψ2-ψ1) + CN (θ1, ψ1)

[0149] CN (θ2, ψ1) = (CN (θ3, ψ1) - CN (θ1, ψ1)) / (θ3-θ1) * (θ2-θ1) + CN (θ1, ψ1)

[0150] CN (θ3, ψ2) = (CN (θ3, ψ3) - CN (θ3, ψ1)) / (ψ3-ψ1) * (ψ2-ψ1) + CN (θ3, ψ1)

[0151] CN (θ2, ψ3) = (CN (θ3, ψ3) - CN (θ1, ψ3)) / (θ3-θ1) * (θ2-θ1) + CN (θ1, ψ3)

[0152] Wherein, CN is the to-be-completed quantity; θ1, θ2 and θ3 respectively represent the pitch attitude angles of the sample point one, the sample point two and the sample point three; ψ1, ψ2 and ψ3 respectively represent the yaw attitude angles of the sample point one, the sample point two and the sample point three;

[0153] The interpolation sample point comprises:

[0154] Module A1: judging whether the variable to be interpolated is out of the sample range on the left end face of the to-be-predicted x i position, and the result is yes, and the limiting processing is performed; the result is no, and the linear interpolation is performed; wherein, x i position is the i-th position variable;

[0155] Module A2: judging whether the x i position is out of the maximum sample range, and the result is yes, and the linear interpolation is performed by using the linear decay transition to zero method; the result is no, and the linear interpolation is performed;

[0156] The mathematical expression of the limiting processing is:

[0157] CNleft(θ i , ψ i ) = CNleft(θ max / θ min , ψ max / ψ min )

[0158] CNright(θ i , ψ i ) = CNright(θ max / θ min , ψ max / ψ min )

[0159] Wherein, θ i is the pitch attitude angle of the i-th sample point; ψ i is the yaw attitude angle of the i-th sample point; CNleft represents the normal force coefficient on the left end face of the to-be-predicted x i position; CNright represents the normal force coefficient on the right end face of the to-be-predicted x i position;

[0160] The mathematical expression of the linear interpolation performed by using the linear decay transition to zero method is:

[0161] CN(x i ) = -CN(x max ) / x ∞ -x max )*(x i -x max )+CN(x max )

[0162] wherein x max is the position of the maximum sample point in the lateral direction; x ∞ is the position where the aerodynamic interference tends to zero, and is a preset threshold value set according to the task requirement.

[0163] Those skilled in the art understand that, in addition to implementing the system provided by the present application and each device, module, unit thereof in the form of pure computer readable program code, the system provided by the present application and each device, module, unit thereof can also be implemented in the form of logic gate, switch, application specific integrated circuit, programmable logic controller and embedded microcontroller, etc. to achieve the same function by logically programming the method steps. Therefore, the system provided by the present application and each device, module, unit thereof can be considered as a hardware component, and the devices, modules, units included therein for achieving various functions can also be considered as structures within the hardware component; the devices, modules, units for achieving various functions can also be considered as both software modules for achieving methods and structures within the hardware component.

[0164] The specific embodiments of the present application are described above. It needs to be understood that the present application is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essential content of the present application. The embodiments of the present application and the features in the embodiments can be arbitrarily combined with each other without conflict.

Claims

1. A method for compressing samples in aerodynamic modeling of a tail retraction separation mesh, characterized in that, include: Step S1: Based on the separation state range given by the task, calculate the separation trajectory of the separated body; Step S2: Based on the separation trajectory, determine the position bundle and attitude bundle of the separated body, and then select the sampling range for aerodynamic modeling of the separated body; Step S3: Arrange sample points based on the sampling range, and then complete the data of the unarranged sample points to construct the aerodynamic model of the separator.

2. The method for compressing aerodynamic modeling samples of a tail-end separation mesh according to claim 1, characterized in that, In step S1, the separation state range includes: separation Mach number, altitude, angle of attack, sideslip angle, and angular velocity; The separation trajectory of the separated body was obtained using CFD software, and the mathematical expression is as follows: y t =f(Ma,H,alpha,beta,u0,v0,w0,wx0,wy0,wz0,x0,y0,z0,θ0,ψ0,γ0,t) Among them, y t The values ​​represent the position, attitude angles, velocities, angular velocities, aerodynamic forces, or aerodynamic torques of the separated body at time t; f represents a function; Ma is the separation Mach number, H is the separation altitude, alpha is the separation angle of attack, beta is the separation sideslip angle, u0, v0, and w0 are the axial, vertical, and lateral velocities of the separated body at the initial separation moment, wx0, wy0, and wz0 are the axial, vertical, and lateral angular velocities of the separated body at the initial separation moment, x0, y0, and z0 are the position coordinates of the separated body at the initial separation moment, θ0, ψ0, and γ0 are the pitch, yaw, and roll attitude angles of the separated body at the initial separation moment, and t is the time.

3. The method for compressing sample data for aircraft tail retraction separation mesh force measurement aerodynamic modeling according to claim 1, characterized in that, In step S2, the number of position variables is confirmed. Using the position and attitude bundles obtained from the curves of the separated trajectories, the maximum and minimum values ​​of each position variable and its corresponding attitude angle are acquired to form a sampling range. The sampling range is (K*y max ,K*y min ,K*z max ,K*z min ,K*θ max ,K*θ min ,K*ψ max ,K*ψ min ), where K is the safety factor, * represents the product, max represents the maximum value, min represents the minimum value; θ represents the pitch attitude angle, and ψ represents the yaw attitude angle.

4. The method for compressing aerodynamic modeling samples of a tail-end separation mesh according to claim 1, characterized in that, In step S3, based on the sampling range, sample points are arranged according to the "×" sample compression algorithm, and then the data of the unarranged sample points are recovered and the sample points are interpolated through the data completion algorithm to construct the aerodynamic model of the separated body. The sample points are arranged according to the "×" sample compression algorithm. Let the variables of the sample points include A and B. Let A take A1, A2, A3; let B take B1, B2, B3, to obtain (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2); where the variable types of A and B are the same, which is any one of the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic torque of the separated body.

5. The method for compressing aerodynamic modeling samples of a tail-end separation mesh according to claim 4, characterized in that, In step S3, the data completion algorithm is mathematically expressed as follows: CN(θ1,ψ2)=(CN(θ1,ψ3)-CN(θ1,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ1,ψ1) CN(θ2,ψ1)=(CN(θ3,ψ1)-CN(θ1,ψ1)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ1) CN(θ3,ψ2)=(CN(θ3,ψ3)-CN(θ3,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ3,ψ1) CN(θ2,ψ3)=(CN(θ3,ψ3)-CN(θ1,ψ3)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ3) Where CN is the complete parameter to be filled; θ1, θ2 and θ3 represent the pitch attitude angles of sample point 1, sample point 2 and sample point 3 respectively; ψ1, ψ2 and ψ3 represent the yaw attitude angles of sample point 1, sample point 2 and sample point 3 respectively. The interpolation sample points include: Step A1: Determine the value of the variable to be interpolated in the predicted x. i If the position exceeds the sample range on the left and right end faces, and the result is yes, amplitude limiting is applied; otherwise, linear interpolation is performed; where x i The position is the i-th position variable; Step A2: Determine x i If the position exceeds the maximum sample range along the axis, and the result is yes, then linear interpolation is performed using a linear decay transition to zero method; if the result is no, then linear interpolation is performed. The mathematical expression for the amplitude limiting process is: CN left(θ i , ψ i ) = CN left(θ max / θ min , ψ max / ψ min ) CN ≡ (θ i ,ψ i )=CNｳ(θ max / θ min ,ψ max / ψ min ) Where, θ i Let ψ be the pitch attitude angle of the i-th sample point; i Let be the yaw attitude angle of the i-th sample point; CN left represents the x to be predicted. i The normal force coefficient on the left end face; CN right represents the x to be predicted. i Normal force coefficient on the right end face; The method for linear decay transitioning to zero involves linear interpolation, and its mathematical expression is: CN(x i )=-CN(x max ) / x ∞ -x max )*(x i -x max )+CN(x max ) Where, x max x represents the location of the largest sample point in the horizontal direction. ∞ The position where aerodynamic interference approaches zero is a preset threshold set according to mission requirements.

6. A sample compression system for aerodynamic modeling of a tail retraction separation grid, characterized in that, include: Module M1: Based on the separation state range given by the task, calculate the separation trajectory of the separated body; Module M2: Based on the separation trajectory, determine the position and attitude bundles of the separated body, and then select the sampling range for aerodynamic modeling of the separated body; Module M3: Based on the sampling range, sample points are arranged, and then the data of the unarranged sample points are supplemented to construct the aerodynamic model of the separator.

7. The aircraft tail retraction separation mesh force measurement aerodynamic modeling sample compression system according to claim 6, characterized in that, In module M1, the separation state range includes: separation Mach number, altitude, angle of attack, sideslip angle, and angular velocity; The separation trajectory of the separated body was obtained using CFD software, and the mathematical expression is as follows: y t =f(Ma,H,alpha,beta,u0,v0,w0,wx0,wy0,wz0,x0,y0,z0,θ0,ψ0,γ0,t) Among them, y t The values ​​represent the position, attitude angles, velocities, angular velocities, aerodynamic forces, or aerodynamic torques of the separated body at time t; f represents a function; Ma is the separation Mach number, H is the separation altitude, alpha is the separation angle of attack, beta is the separation sideslip angle, u0, v0, and w0 are the axial, vertical, and lateral velocities of the separated body at the initial separation moment, wx0, wy0, and wz0 are the axial, vertical, and lateral angular velocities of the separated body at the initial separation moment, x0, y0, and z0 are the position coordinates of the separated body at the initial separation moment, θ0, ψ0, and γ0 are the pitch, yaw, and roll attitude angles of the separated body at the initial separation moment, and t is the time.

8. The aircraft tail retraction separation grid force measurement aerodynamic modeling sample compression system according to claim 6, characterized in that, In module M2, the number of position variables is confirmed. Using the position and attitude bundles obtained from the curves of the separated trajectories, the maximum and minimum values ​​of each position variable and its corresponding attitude angle are acquired to form a sampling range; the sampling range is (K*y). max ,K*y min ,K*z max ,K*z min ,K*θ max ,K*θ min ,K*ψ max ,K*ψ min ), where K is the safety factor, * represents the product, max represents the maximum value, min represents the minimum value; θ represents the pitch attitude angle, and ψ represents the yaw attitude angle.

9. The aircraft tail retraction separation grid force measurement aerodynamic modeling sample compression system according to claim 6, characterized in that, In module M3, based on the sampling range, sample points are arranged according to the "×" sample compression algorithm, and then the data of the unarranged sample points are recovered through the data completion algorithm to construct the aerodynamic model of the separator. The sample points are arranged according to the "×" sample compression algorithm. Let the variables of the sample points include A and B. Let A take A1, A2, A3; let B take B1, B2, B3, to obtain (A1, B1), (A1, B3), (A3, B1), (A3, B3) and (A2, B2); where the variable types of A and B are the same, which is any one of the position, attitude angle, velocity, angular velocity, aerodynamic force or aerodynamic torque of the separated body.

10. The aircraft tail retraction separation grid force measurement aerodynamic modeling sample compression system according to claim 9, characterized in that, In module M3, the data completion algorithm has the following mathematical expression: CN(θ1,ψ2)=(CN(θ1,ψ3)-CN(θ1,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ1,ψ1) CN(θ2,ψ1)=(CN(θ3,ψ1)-CN(θ1,ψ1)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ1) CN(θ3,ψ2)=(CN(θ3,ψ3)-CN(θ3,ψ1)) / (ψ3-ψ1)*(ψ2-ψ1)+CN(θ3,ψ1) CN(θ2,ψ3)=(CN(θ3,ψ3)-CN(θ1,ψ3)) / (θ3-θ1)*(θ2-θ1)+CN(θ1,ψ3) Where CN is the complete parameter to be filled; θ1, θ2 and θ3 represent the pitch attitude angles of sample point 1, sample point 2 and sample point 3 respectively; ψ1, ψ2 and ψ3 represent the yaw attitude angles of sample point 1, sample point 2 and sample point 3 respectively. The interpolation sample points include: Module A1: Determines the value of the variable to be interpolated in the predicted x. i If the position exceeds the sample range on the left and right end faces, and the result is yes, amplitude limiting is applied; otherwise, linear interpolation is performed; where x i The position is the i-th position variable; Module A2: Determine x i If the position exceeds the maximum sample range along the axis, and the result is yes, then linear interpolation is performed using a linear decay transition to zero method; if the result is no, then linear interpolation is performed. The mathematical expression for the amplitude limiting process is: CN left(θ i , ψ i ) = CN left(θ max / θ min , ψ max / ψ min ) CN ≡ (θ i ,ψ i )=CNｳ(θ max / θ min ,ψ max / ψ min ) Where, θ i Let ψ be the pitch attitude angle of the i-th sample point; i Let be the yaw attitude angle of the i-th sample point; CN left represents the x to be predicted. i The normal force coefficient on the left end face; CN right represents the x to be predicted. i Normal force coefficient on the right end face; The method for linear decay transitioning to zero involves linear interpolation, and its mathematical expression is: CN(x i )=-CN(x max ) / x ∞ -x max )*(x i -x max )+CN(x max ) Where, x max x represents the location of the largest sample point in the horizontal direction. ∞ The position where aerodynamic interference approaches zero is a preset threshold set according to mission requirements.

Images