Method, device and medium for online estimation of aircraft aerodynamic parameters based on thrust deviation correction

Abstract

The present invention provides an online estimation method, device and medium for aircraft aerodynamic parameters based on thrust deviation correction, and relates to the technical field of online identification of aircraft parameters. The method comprises: using an extended Kalman filter to iteratively predict the augmented state of the aircraft system step by step; using a forgotten recursive least squares estimator to estimate the axial aerodynamic force and normal aerodynamic force coefficients and the thrust deviation proportional coefficient within a given time window, and then calculating the aerodynamic lift, drag coefficients and the corrected thrust estimation results; judging the convergence of the thrust deviation proportional coefficient estimation results; if it has not converged, aerodynamic derivative estimation is carried out based on the uncorrected thrust and lift and drag aerodynamic coefficients; if it has converged, the corrected thrust and the corrected lift and drag aerodynamic coefficients are brought into the general recursive least squares estimator to carry out aerodynamic derivative estimation. The online estimation method proposed by the present invention can effectively improve the accuracy of longitudinal aerodynamic derivative estimation under thrust measurement deviation.

Description

Technical Field

[0001] The present invention relates to the technical field of on-line identification of aircraft parameters. Specifically, it relates to a method, device and medium for on-line estimation of aircraft aerodynamic parameters based on thrust deviation correction. Background Technique

[0002] During the flight of an aircraft, on-line estimation of aerodynamic parameters refers to gradually obtaining the aerodynamic characteristic parameters of the aircraft in real time from flight measurement data through methods such as filtering or recursion. It is an important means to obtain the aerodynamic characteristics of the aircraft in the real environment. Obtaining aerodynamic parameters in real time is of great significance for flight safety and fault adaptability, etc.

[0003] In the research of on-line parameter estimation methods, the estimation accuracy has always been the focus of attention. Among them, the accuracy of measurement data is an important factor affecting the accuracy. During the actual flight of an aircraft, it is generally powered flight. Thrust, as the system input, needs to be measured. However, the engine thrust is usually difficult to directly measure and obtain, and often can only be estimated by some engineering methods. Therefore, there is inevitably a deviation between the thrust measurement data and the true value. Especially when the aircraft makes a maneuver, the credibility of the estimated thrust data is usually low, which also brings difficulties to the identification of aerodynamic characteristic parameters of powered flight data, and will greatly affect the accuracy and credibility of the aerodynamic force parameter identification results.

[0004] The thrust measurement deviation directly affects the flight state of the aircraft, bringing uncertainty to the flight state model. Different from the influencing factors such as wind field and atmospheric environment disturbance, the influence brought by the thrust measurement deviation belongs to the category of deterministic influence. Therefore, the processing methods of some random variables are not very applicable to the processing of thrust measurement deviation. At present, in the aerodynamic parameter identification, the treatment of thrust deviation mostly adopts an avoidance method, such as using data under unpowered flight, or only identifying aerodynamic parameters in the direction not affected by thrust. However, for some aircraft with the engine not shut down and axial aerodynamic parameters, the identified aerodynamic parameters will inevitably include the influence of the thrust measurement deviation amount. If the deviation is large, it will lead to the unbelievability of the identification result.

[0005] Therefore, there is an urgent need to develop an on-line identification method that can simultaneously estimate the thrust deviation and aerodynamic parameters for the problem of on-line estimation of aerodynamic parameters under the influence of the deterministic thrust deviation, so as to solve the accuracy problem caused by inaccurate thrust measurement that has long been faced in aerodynamic parameter identification. Summary of the Invention

[0006] The present invention aims to solve at least one of the above technical problems existing in the prior art.

[0007] To this end, the first aspect of the present invention provides a method for on-line estimation of aircraft aerodynamic parameters based on thrust deviation correction.

[0008] The second aspect of the present invention provides a computer device.

[0009] The third aspect of the present invention provides a computer-readable storage medium.

[0010] The present invention provides an online estimation method for aircraft aerodynamic parameters based on thrust deviation correction, including:

[0011] S1. Initialize the parameters of the extended Kalman filter, the forgetting recursive least squares estimator, and the general recursive least squares estimator;

[0012] S2. Use the extended Kalman filter to predict the augmented state of the aircraft system at the i (i + 1)-th step from the augmented state of the aircraft system at the i i-th step; wherein, the state variables in the extended Kalman filter include the aerodynamic coefficients of the aircraft;

[0013] S3. Use the forgetting recursive least squares estimator to estimate the axial aerodynamic coefficient, the normal aerodynamic coefficient, and the thrust deviation proportionality coefficient within a given time window and including the (i + 1)-th step according to the thrust measurement data and the overload measurement data;

[0014] S4. Obtain the coordinate transformation matrix using the state estimation result at the i (i + 1)-th step, and convert the axial aerodynamic coefficient and the normal aerodynamic coefficient in the body coordinate system into the drag coefficient and the lift coefficient as the corrected aerodynamic coefficients;

[0015] S5. Calculate the corrected thrust estimation result according to the estimated thrust deviation proportionality coefficient;

[0016] S6. Judge the convergence of the estimation result of the thrust deviation proportionality coefficient within a given data window;

[0017] S7. If the estimation result of the thrust deviation proportionality coefficient does not converge, perform RLS estimation based on the uncorrected thrust and aerodynamic coefficients to obtain the aerodynamic derivative estimation result; if the estimation result of the thrust deviation proportionality coefficient has converged, start from the convergence moment, substitute the corrected thrust, the corrected lift coefficient, and the drag coefficient into the general recursive least squares estimator to obtain the aerodynamic derivative estimation result.

[0018] According to the online estimation method for aircraft aerodynamic parameters based on thrust deviation correction of the above technical solution of the present invention, the following additional technical features may also be included:

[0019] In the above technical solution, the initialization of the parameters of the extended Kalman filter, the forgetting recursive least squares estimator, and the general recursive least squares estimator includes:

[0020] Set the initial values, initial covariance matrix, noise covariance matrix, window length, and forgetting factor size of the extended Kalman filter, forgetting recursive least squares estimator, and general recursive least squares estimator.

[0021] In the above technical solution, the aerodynamic coefficients of the aircraft include six-component aerodynamic coefficients.

[0022] In the above technical solution, step S3 includes:

[0023] Input the thrust measurement data and overload measurement data with a data window length of N w and including the i +1 step into the forgetting recursive least squares estimator to calculate the axial aerodynamic coefficient, normal aerodynamic coefficient, and thrust deviation proportional coefficient. Among them, the overload measurement data includes the measured values of the axial linear acceleration and normal linear acceleration of the aircraft, and the calculation method includes:

[0024]

[0025]

[0026] Among them, represents the dynamic pressure; represents the characteristic area of the aircraft; represents the thrust measurement data; represents the axial aerodynamic coefficient; represents the thrust deviation proportional coefficient; represents the mass of the aircraft; represents the measured value of the axial linear acceleration of the aircraft; represents the normal aerodynamic coefficient; represents the measured value of the normal linear acceleration of the aircraft.

[0027] In the above technical solution, the method for obtaining the coordinate transformation matrix using the state estimation result of the i +1 step in step S4 includes calculating the coordinate transformation matrix using the estimated data of the angle of attack and sideslip angle.

[0028] In the above technical solution, the method for calculating the corrected thrust estimation result according to the estimated thrust deviation proportional coefficient in step S5 includes:

[0029]

[0030] Among them, represents the corrected thrust estimation result.

[0031] In the above technical solution, the method for judging the convergence of the thrust deviation proportional coefficient estimation result within a given data window in step S6 includes:

[0032] For each thrust deviation proportional coefficient estimation result reaching a given data window length, the mean and standard deviation of the data segment are counted;

[0033] Judge whether the absolute deviation of the current mean value relative to the mean value of the previous data segment and the current standard deviation exceed the convergence judgment threshold. If both are less than the threshold value, it is considered that the thrust deviation proportional coefficient estimation result at the current moment has converged, that is, the thrust estimation result has converged; otherwise, it has not converged.

[0034] In the above technical solution, in step S7, in the process of carrying out RLS estimation based on uncorrected thrust and aerodynamic coefficients, the thrust and aerodynamic coefficients output by the extended Kalman filter are input into the general recursive least squares estimator to obtain the aerodynamic derivative estimation result.

[0035] The present invention provides a computer device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is loaded and executed by the processor, an online estimation method of aerodynamic parameters of an aircraft based on thrust deviation correction as described in any one of the above technical solutions is implemented.

[0036] The present invention provides a computer-readable storage medium storing a program, which, when loaded by a processor, implements an online estimation method for aerodynamic parameters of an aircraft based on thrust deviation correction as described in any one of the above technical solutions.

[0037] In summary, due to the adoption of the above technical features, the beneficial effects of the present invention are:

[0038] Based on the idea of ​​multi-source measurement data fusion, the present invention uses the extended Kalman filter, the forgotten recursive least squares estimator and the general recursive least squares estimator to jointly realize the online estimation of aerodynamic parameters based on thrust deviation correction, and designs a convergence judgment link for the thrust estimation results obtained by recursion. The overall algorithm framework can run on the airborne computing platform to realize online processing of measurement data.

[0039] The aerodynamic parameter online estimation algorithm based on thrust deviation correction established by the present invention is based on system input and output measurement data, overload measurement data and thrust measurement data, and can simultaneously perform online estimation of the aircraft state, thrust deviation proportional coefficient, aerodynamic coefficient and aerodynamic derivative.

[0040] ​The online estimation method proposed by the present invention can effectively improve the accuracy of aerodynamic derivative estimation under thrust measurement deviation. For the level flight simulation example of the same aircraft, for the identification of the drag stability derivative under four different thrust measurement deviations, the relative root mean square error of the identification result can be reduced by about 40%-55% compared with the uncorrected case.

[0041] The additional aspects and advantages of the present invention will become obvious in the following description part, or be understood through the practice of the present invention. Brief Description of the Drawings

[0042] The above and / or additional aspects and advantages of the present invention will become obvious and easy to understand from the description of the embodiments in conjunction with the following drawings, where:

[0043] Figure 1 is a flowchart of an online estimation method for aircraft aerodynamic parameters based on thrust deviation correction according to an embodiment of the present invention;

[0044] Figure 2 is a comparison diagram of the thrust estimation results of the aircraft corrected by using the method of the present disclosure under four deviation conditions;

[0045] Figure 3 is a comparison diagram of the uncorrected and corrected aerodynamic derivative identification results when the thrust deviation proportionality coefficient is 0.5;

[0046] Figure 4 is a comparison diagram of the uncorrected and corrected aerodynamic derivative identification results when the thrust deviation proportionality coefficient is 0.8;

[0047] Figure 5 is a comparison diagram of the uncorrected and corrected aerodynamic derivative identification results when the thrust deviation proportionality coefficient is 1.2;

[0048] Figure 6 is a comparison diagram of the uncorrected and corrected aerodynamic derivative identification results when the thrust deviation proportionality coefficient is 1.5. Detailed Description of the Embodiments

[0049] In order to be able to more clearly understand the above objects, features and advantages of the present invention, the present invention will be further described in detail below in conjunction with the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0050] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0051] The following refers toFigures 1 to 6 To describe an on-line estimation method of aircraft aerodynamic parameters based on thrust deviation correction provided according to some embodiments of the present invention.

[0052] Some embodiments of the present application provide an on-line estimation method of aircraft aerodynamic parameters based on thrust deviation correction.

[0053] As Figure 1 shown, the first embodiment of the present invention proposes an on-line estimation method of aircraft aerodynamic parameters based on thrust deviation correction, including steps S1 to S8.

[0054] S1. Initialize the parameters of the extended Kalman filter, the forgetting recursive least squares estimator, and the general recursive least squares estimator.

[0055] It should be noted that the extended Kalman filter (EKF), the forgetting recursive least squares estimator (FRLS), and the general recursive least squares estimator (RLS) are all common algorithms in aircraft parameter identification. The EKF can transform the parameter estimation problem into a state estimation problem. According to the actual flight situation of the aircraft, a continuous estimation model is adopted, and real-time effects are achieved through discrete-time measurements and discrete filtering algorithms. The RLS is one of the commonly used identification algorithms in system identification, which is suitable for identifying multi-input multi-output linear motion models; it updates parameter estimates in a recursive manner and is suitable for real-time system parameter identification. The FRLS combines the advantages of the RLS algorithm and can effectively identify the time-varying characteristics of system parameters; by introducing a forgetting factor, the algorithm assigns decreasing weights to past data, thereby adapting to the time-varying characteristics of parameters. The specific algorithm content will not be elaborated here.

[0056] In some embodiments, the initialization of the parameters of the extended Kalman filter, the forgetting recursive least squares estimator, and the general recursive least squares estimator includes:

[0057] Setting the initial values, initial covariance matrices, noise covariance matrices, window lengths, and forgetting factor sizes of the extended Kalman filter, the forgetting recursive least squares estimator, and the general recursive least squares estimator.

[0058] S2. Use the extended Kalman filter to predict the augmented state of the aircraft system at the i th step of the aircraft system augmented state at the i +1th step.

[0059] In the Figure 1 shown embodiment, the system input variable u, the system output measurement variable y, and the thrust measurement value F are input into the extended Kalman filter to obtain the system state variable x, the aerodynamic drag coefficient and the aerodynamic lift coefficient .

[0060] System state variables usually include the position information and velocity information of the aircraft. In the present disclosure, the aerodynamic coefficients of the aircraft are also used as state variables in the extended Kalman filter. In a specific embodiment, the aerodynamic coefficients of the aircraft include six-component aerodynamic coefficients, and the six-component aerodynamic coefficients usually include lift coefficient, drag coefficient, pitch moment coefficient, roll moment coefficient, yaw moment coefficient, and side force coefficient. In the present disclosure, the lift coefficient and drag coefficient are mainly taken as examples for illustration. That is to say, in the present disclosure, through step S2, the six-component aerodynamic coefficients are extended as state variables, and the EKF filter is used to calculate the estimated value and covariance matrix of the (i + 1)-th step from the estimated value and covariance matrix of the i-th step.

[0061] S3. Use a forgetting recursive least squares estimator to estimate the axial aerodynamic coefficient, normal aerodynamic coefficient, and thrust deviation proportional coefficient based on the thrust measurement data and overload measurement data.

[0062] As Figure 1 shown, the system state variable x output by the extended Kalman filter, the thrust measurement value F, and the overload measurement data are all input into the forgetting recursive least squares estimator FRLS.

[0063] Specifically, step S3 includes:

[0064] Input the thrust measurement data and overload measurement data with a data window length of N w and including the i +1-th step into the forgetting recursive least squares estimator to calculate the estimated axial aerodynamic coefficient, normal aerodynamic coefficient, and thrust deviation proportional coefficient. Among them, the overload measurement data includes the measured value of the axial linear acceleration of the aircraft and the measured value of the normal linear acceleration of the aircraft, and the calculation method includes:

[0065]

[0066]

[0067] Among them, represents the dynamic pressure; represents the characteristic area of the aircraft; represents the thrust measurement data; represents the axial aerodynamic coefficient; represents the thrust deviation proportional coefficient; represents the mass of the aircraft; represents the measured value of the axial linear acceleration of the aircraft; represents the normal aerodynamic coefficient; represents the measured value of the normal linear acceleration of the aircraft.

[0068] According to the above calculation method, the state estimation result of the i +1-th step can be calculated, that is, the i axial aerodynamic coefficient, normal aerodynamic coefficient and thrust deviation ratio coefficient of the

[0069] S4. Obtain the coordinate transformation matrix by using the state estimation result of the i +1-th step, and convert the axial aerodynamic force and normal aerodynamic coefficient in the body coordinate system into the drag and lift coefficients as the corrected aerodynamic coefficients.

[0070] Specifically, the axial aerodynamic coefficient and normal aerodynamic coefficient obtained in step S3 are parameters in the body coordinate system, and they need to be converted to the velocity coordinate system. The axial aerodynamic coefficient is converted to the drag coefficient, and the normal aerodynamic coefficient is converted to the lift coefficient.

[0071] In a specific embodiment, the method for obtaining the coordinate transformation matrix by using the state estimation result of the i +1-th step in step S4 includes calculating the coordinate transformation matrix by using the estimated data of the angle of attack and sideslip angle. Convert the axial aerodynamic coefficient into the corrected drag coefficient through the coordinate transformation matrix , and convert the normal aerodynamic coefficient into the corrected lift coefficient through the coordinate transformation matrix .

[0072] S5. Calculate the corrected thrust estimation result according to the estimated thrust deviation ratio coefficient.

[0073] The method for calculating the corrected thrust estimation result according to the estimated thrust deviation ratio coefficient in step S5 includes:

[0074]

[0075] wherein, represents the corrected thrust estimation result.

[0076] S6. Judge the convergence of the estimated result of the thrust deviation ratio coefficient within the given data window.

[0077] In some embodiments, the method for judging the convergence of the estimated result of the thrust deviation ratio coefficient within the given data window in step S6 includes:

[0078] For each estimated result of the thrust deviation ratio coefficient that reaches the length of the given data window, statistically calculate the mean and standard deviation of this data segment;

[0079] Judge whether the absolute deviation of the current mean value relative to the mean value of the previous data segment and the current standard deviation exceed the convergence discrimination threshold. If both are less than the threshold value, it is considered that the estimation result of the thrust deviation proportional coefficient at the current moment has converged, that is, the thrust estimation result has converged; otherwise, it has not converged.

[0080] S7. If the estimation result of the thrust deviation proportional coefficient has not converged, perform RLS estimation based on the uncorrected thrust and aerodynamic coefficients to obtain the estimation result of the aerodynamic derivatives; if the estimation result of the thrust deviation proportional coefficient has converged, starting from the convergence moment, use the corrected thrust, corrected drag coefficient, and lift coefficient to substitute into the general recursive least squares estimator to obtain the estimation result of the aerodynamic derivatives.

[0081] Specifically, in step S7, during the process of performing RLS estimation based on the uncorrected thrust and aerodynamic coefficients, directly input the thrust and aerodynamic coefficients output by the extended Kalman filter into the general recursive least squares estimator to obtain the estimation result of the aerodynamic derivatives.

[0082] S8. Repeat steps S1 - S7 until all flight data processing is completed or until all sampled data filtering processing is completed.

[0083] In a specific embodiment, taking the longitudinal flight data identification of a certain civil aircraft standard model aircraft in level flight as an example, the specific implementation manner of the present disclosure is described. The flight speed of the aircraft is 0.2 Mach and the flight altitude is 100 m. Considering the influence of measurement noise, between the simulation time of 0.5 s and 2.5 s, a single - cycle "3211" small perturbation signal with an amplitude of 3 degrees is added to the trim elevator deflection input to stimulate the dynamic response of the system, improve the aircraft parameter identification accuracy, and carry out flight simulation to obtain the system input - output measurement data as the data input of the online estimation method, where it is assumed that the thrust deviation proportional coefficient of the thrust measurement value relative to the true situation are 0.5, 0.8, 1.2, and 1.5 times respectively.

[0084] The specific process is as follows:

[0085] Initialization: Set the initial state estimation value of the EKF filter according to the flight data at the start of the simulation. The initial values of the aerodynamic coefficients and derivatives to be estimated are all 0, the initial value of the measurement noise covariance matrix is taken as 10 -4 I, where I represents the identity matrix; the estimation window length of the RLS estimator is set to 100 data points, the estimation window of the FRLS estimator is set to 1000, the convergence discrimination window length is 500, the forgetting factor is taken as 0.99, the parameter estimation convergence threshold value is taken as 0.1, the initial covariance matrix values of the state variables, aerodynamic coefficients, and thrust deviation proportional coefficient are taken as the identity matrix, and the initial covariance matrix of the longitudinal aerodynamic derivatives is taken as a diagonal matrix, and the diagonal elements are [1, 1e - 2, 1e - 2, 1, 1].

[0086] State estimation: The augmented state of the system at the (i + 1)-th step is predicted from the augmented state of the system at the i-th step using the EKF filter.

[0087] Estimation of thrust deviation proportional coefficient and aerodynamic coefficients: Based on the calculation method in step S3 and the FRLS estimator, the axial aerodynamic coefficient, normal aerodynamic coefficient, and thrust deviation proportional coefficient are estimated from the state estimation results, thrust, and overload measurement data from the (i + 2)- N w th to (i + 1)-th steps. The coordinate transformation matrix is obtained using the estimated data of the angle of attack and sideslip angle, and the corresponding drag coefficient and lift coefficient are obtained through coordinate transformation, thereby obtaining the correction results of the aerodynamic and thrust coefficients at the (i + 1)-th step.

[0088] Convergence judgment: For each estimated result of the thrust deviation proportional coefficient that reaches a given window length, the mean and standard deviation of this data segment are statistically calculated, and it is judged whether the absolute deviation of the current mean relative to the mean of the previous data segment and the current standard deviation exceed the convergence discrimination threshold. If both are less than the threshold value, it is considered that the estimated result at the current moment has converged; otherwise, it has not converged. Estimation of aerodynamic derivatives: If

[0089] the estimated result has converged, the corrected thrust, corrected drag coefficient, and lift coefficient obtained from the (i + 2)- th to (i + 1)-th steps are substituted into the RLS estimator to estimate the aerodynamic derivatives at the (i + 1)-th step; if N w the estimated result has not converged, the uncorrected thrust F, aerodynamic drag coefficient and aerodynamic lift coefficient are used to estimate the aerodynamic derivatives at the (i + 1)-th step. For this flight data identification example, the above steps are adopted. Finally, the thrust estimation results of the aircraft under four deviation conditions are calculated as

[0090] shown. It is found that the thrust variation curve obtained by FRLS estimation can converge to the reference value for simulation, that is, near the trim thrust. As Figure 2 shown, under different thrust deviation proportional coefficients, the steady aerodynamic drag derivatives estimated from the correction results of thrust and aerodynamic coefficients can gradually approach the reference value; the root mean square errors of the identification results with and without correction relative to the reference value are calculated respectively, and it is found that for the four deviation conditions, the improvement in the identification accuracy of the steady aerodynamic drag derivatives is about 40% - 55%. Figures 3 to 6 As

[0091] Another computer device provided by some embodiments of the present invention includes a processor and a memory. A computer program is stored in the memory. When the computer program is loaded and executed by the processor, it implements the online estimation method of aircraft aerodynamic parameters based on thrust deviation correction as described in any of the above embodiments.

[0092] A computer-readable storage medium provided by some other embodiments of the present invention stores a program. When the program is loaded by a processor, it implements the online estimation method of aircraft aerodynamic parameters based on thrust deviation correction as described in any of the above embodiments.

[0093] In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

[0094] Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction, characterized in that: include: S1, initializing the parameters of the extended Kalman filter, the forgotten recursive least squares estimator and the general recursive least squares estimator; S2, using the extended Kalman filter, i The augmented state prediction of the aircraft system i +1 step augmented state of the aircraft system; wherein the state variables in the extended Kalman filter include the aerodynamic coefficients of the aircraft; S3, using a forgotten recursive least squares estimator to estimate the axial aerodynamic coefficient, the normal aerodynamic coefficient, and the thrust deviation proportional coefficient within a given time window and including the i+1th step according to the thrust measurement data and the overload measurement data; S4. Using i The state estimation result of step +1 obtains the coordinate transformation matrix, which transforms the axial aerodynamic coefficient and normal aerodynamic coefficient in body coordinates into drag coefficient and lift coefficient as the corrected aerodynamic coefficient of step i+1; S5. Calculate the corrected thrust estimation result of step i+1 according to the estimated thrust deviation proportional coefficient; S6. judging the convergence of the thrust deviation proportional coefficient estimation result within a given data window; the judging method includes: For each thrust deviation proportional coefficient estimation result reaching a given data window length, the mean and standard deviation of the data segment are calculated; Determine whether the absolute deviation of the current mean value relative to the mean value of the previous data segment and the current standard deviation exceed the convergence judgment threshold. If both are less than the threshold value, it is considered that the thrust deviation proportional coefficient estimation result at the current moment has converged, that is, the thrust estimation result has converged; otherwise, it has not converged; S7. If the estimation result of the thrust deviation proportional coefficient has not converged, a general recursive least squares estimation is performed based on the uncorrected thrust and aerodynamic force coefficient to obtain the aerodynamic derivative estimation result of the i+1th step; if the estimation result of the thrust deviation proportional coefficient has converged, the corrected thrust, the corrected lift coefficient and the drag coefficient are introduced into the general recursive least squares estimator from the moment of convergence to obtain the aerodynamic derivative estimation result of the i+1th step.

2. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 1, characterized in that: The parameters of the initialization extended Kalman filter, the forgotten recursive least squares estimator and the general recursive least squares estimator include: Set the initial values, initial covariance matrix, noise covariance matrix, window length, and forgetting factor size for the extended Kalman filter, forgotten recursive least squares estimator, and general recursive least squares estimator.

3. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 1, characterized in that: The aerodynamic coefficients of the aircraft include six-component aerodynamic coefficients.

4. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 1, characterized in that: Step S3 includes: The data window length is N w And including i The thrust measurement data and overload measurement data of step +1 are input into the forgotten recursive least squares estimator to calculate the axial aerodynamic coefficient, the normal aerodynamic coefficient and the thrust deviation proportional coefficient, wherein the overload measurement data includes the measurement value of the axial acceleration and the normal acceleration of the aircraft, and the calculation method includes: in, Indicates dynamic pressure; Indicates the characteristic area of ​​the aircraft; Indicates thrust measurement data; represents the axial aerodynamic coefficient; It represents the thrust deviation proportionality coefficient; Indicates the mass of the aircraft; It represents the measured value of the axial linear acceleration of the aircraft; represents the normal aerodynamic coefficient; Represents the measured value of the normal acceleration of the aircraft.

5. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 1, characterized in that: In step S4, the i The method for obtaining the coordinate transformation matrix from the state estimation result of step +1 includes calculating the coordinate transformation matrix using the estimated data of the angle of attack and the sideslip angle.

6. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 4, characterized in that: The method for calculating the corrected thrust estimation result according to the estimated thrust deviation proportional coefficient in step S5 includes: in, represents the corrected thrust estimation result.

7. The method for online estimation of aircraft aerodynamic parameters based on thrust deviation correction according to claim 1, characterized in that: In step S7, in the general recursive least squares estimation process based on the uncorrected thrust and aerodynamic force coefficients, the thrust and aerodynamic force coefficients output by the extended Kalman filter are input into the general recursive least squares estimator to obtain the aerodynamic derivative estimation result.

8. A computer device, characterized in that: The invention comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded and executed by the processor, an online estimation method for aerodynamic parameters of an aircraft based on thrust deviation correction as described in any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium, characterized in that: A program is stored, and when the program is loaded by a processor, an online estimation method of aircraft aerodynamic parameters based on thrust deviation correction as described in any one of claims 1 to 7 is implemented.

Images