Aircraft simulation model checking method

Abstract

The invention provides an aircraft simulation model checking method. The aircraft simulation model checking method comprises the following steps: S1, determining a flight mode of an aircraft operated by a pilot and recording flight data; s2, after the flight data are obtained, an airplane simulation model is established in combination with existing appearance parameters of an airplane; s3, carrying out correction and compatibility detection on the flight data and the simulation data; s4, determining and identifying an aerodynamic derivative needing to be corrected according to a compatibility detection result, and correcting the aerodynamic derivative; and S5, when the corrected data accords with the specification, exporting a final corrected aerodynamic model parameter table, and generating a debugging report. According to the airplane simulation model checking method, airplane digitization can be faster and more efficient, and the method has extremely high replicability and can be applied to various aircrafts such as manned aerial vehicles and unmanned aerial vehicles. The flight simulation model created by the method can also be applied to national defense digital software such as war simulation, flight training machines and the like, and has very practical contributions to national safety and military strength.

Description

Technical Field

[0001] The invention relates to an aircraft simulation model, in particular to an aircraft simulation model calibration method. Background Art

[0002] In recent years, the digital application of national defense and military industry has become more and more extensive. In this environment, flight simulation models have also been frequently used in training simulators, tactical training simulators, combat simulation simulations, and new aircraft model development. Then, the accuracy of the flight simulation model will determine whether the model application is authentic, reliable, and reproducible. In this regard, as far as the flight simulation modeling technology of domestic training simulators is concerned, there is no systematic model verification method, nor is there a qualitative method to assess whether the model is reliable at the digital level. In response to this problem, and in combination with some existing national simulator identification standards, a flight simulation model verification method combined with test flight data verification has been developed. Summary of the invention

[0003] The present invention provides an aircraft simulation model calibration method, which solves the problem of parameter optimization and accuracy calibration of the aircraft simulation model. The technical solution is as follows:

[0004] A method for checking an aircraft simulation model comprises the following steps:

[0005] S1: Determine the flight mode in which the pilot operates the aircraft and record the flight data;

[0006] S2: After obtaining the flight data, the aircraft simulation model is established in combination with the existing shape parameters of the aircraft;

[0007] S3: Correct and check the compatibility of flight data and simulation data;

[0008] S4: Determine and identify the aerodynamic derivatives that need to be corrected according to the compatibility test results and correct them;

[0009] S5: When the corrected data meets the specifications, the final corrected aerodynamic model parameter table is exported and a debugging report is generated.

[0010] Furthermore, in step S2, when establishing the aircraft simulation model, it is necessary to create a Simulin simulation model according to the aerodynamic shape and layout of the target aircraft, wherein the Simulin simulation model includes 12th-order motion equations, aerodynamic model, engine model, landing gear model, center of gravity model, and atmosphere model, and then the created simulation model is corrected using real aircraft flight data, including parameter verification of aerodynamic characteristics.

[0011] Furthermore, in step S3, the acquired real aircraft flight data is smoothed, filtered and time-shifted to form a model verification data packet, and the compatibility of the data is verified by checking the compatibility with the simulation data.

[0012] Furthermore, in step S4, the aerodynamic characteristics and control characteristics of the simulation model are verified using real aircraft flight data, and the aerodynamic derivatives of the aircraft simulation model are corrected, so that the aircraft simulation model finally meets the tolerance requirements according to the time history of the test data of each subject of the simulator certification standard.

[0013] Furthermore, in step S1, the flight data to be collected include: flight log time, flight altitude, deflection commands of three rudders, throttle commands, flap position commands, longitudinal, lateral and roll-on attitude angles and angular rates, true airspeed, indicated airspeed, speed and acceleration in the ground coordinate system, remaining fuel, air pressure and temperature, and engine speed.

[0014] Furthermore, the aircraft simulation model includes an aerodynamic module, an engine module, an aerodynamic force and torque calculation module, an aircraft mass and inertia moment module, an overall force and torque module, a kinematics module and a standard sea level atmospheric parameter module.

[0015] Furthermore, in step S3, the flight simulation data is obtained by giving the same excitation to the flight simulation model according to the pilot's operation, and then extracting the flight parameters required for comparison in the kinematics module. Let S be the test flight data and T be the simulation solution result data, and the following error analysis calculation can be performed:

[0016] Mean Square Error:

[0017] e=sqrt(|T1-S1| 2 +|T2+S2| 2 +…+|T n -S n | 2 ) / n

[0018] Here n represents the dimension of the two sets of data, and Tn and Sn are the components of vectors S and T respectively.

[0019] Furthermore, in step S4, a preliminary fitting is performed after the data processing, and the aerodynamic model is adjusted according to the fitting results, on the one hand by directly adding an offset to the aerodynamic coefficient, and on the other hand by modifying the aerodynamic derivatives corresponding to the aerodynamic parameters, so that the flight simulation data and the actual flight data finally meet the compatibility requirements of this type of aircraft, and the verification of the flight simulation model is completed.

[0020] Furthermore, the aerodynamic derivatives include a lift coefficient, a drag coefficient, a side force coefficient, a pitch moment coefficient, a rolling moment coefficient and a yaw moment coefficient.

[0021] In step S4, the correction method includes: the first step is to observe through macroscopic observation that a single excitation of a controllable parameter of the aircraft attitude change will lead to a response change of the speed and attitude angle. Under the same excitation condition, by comparing the curves of the response values ​​of the flight data and the simulation data, the deviation direction and the approximate deviation range of the force and moment coefficients are first determined;

[0022] The second step is to determine the specific aerodynamic derivative or derivatives that need to be adjusted through the control variable method, and then observe the curve fitting after adjustment by pulling the deviation value by 5%, 10% or 15%, and finally determine the value through multiple adjustments.

[0023] The aircraft simulation model verification method can make the aircraft simulation model modeling more accurate and more efficient, and the method is replicable and can be applied to various aircraft such as manned aircraft and unmanned aircraft. The flight simulation model created by the method can also be applied to tactical training simulators and combat simulations, flight training devices and other national defense digital software, making very practical contributions to national security and military strength. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flow chart of the aircraft simulation model calibration method. DETAILED DESCRIPTION

[0025] In order to verify the accuracy of the aircraft simulation model, we first obtain test flight data through real aircraft flight. The pilot first trims the aircraft at a certain speed during fixed-altitude flight. After the aircraft is stable, the pilot manipulates the control surface according to a fixed excitation law (pulse, sweep frequency, square wave, etc.) to allow the aircraft to enter a preset flight mode, including the following modes: short-period oscillation, long-period oscillation, roll mode, Dutch roll, and spiral mode. The flight data of the aircraft is recorded from the time the trim is stable until a certain mode of motion is completed. The next step is to identify the parameters of the test flight data to obtain the aerodynamic characteristics data of the simulation model and perform model verification. The overall steps are as follows:

[0026] Step 1: Use parameter identification method to identify the test flight data and obtain aerodynamic parameters. These aerodynamic parameter data will be used to establish the aerodynamic part of the flight simulation model and combined with the aircraft kinematic part to finally build the entire aircraft simulation model.

[0027] Step 2: By inputting the same wing and control surface operation excitations (wing and elevator deflection) as the test flight operation into the flight simulation model, the simulation data obtained will be used to compare with the actual test flight data, so as to determine the accuracy of the aerodynamic parameters.

[0028] Step 3: Process and check the compatibility of the two sets of data, the test flight data and the simulation data. Different processing can be done here according to the actual situation. If the data collected by the aircraft has high-frequency noise, a low-pass filter can be designed; if the flight data has many outliers, the outliers can be corrected by Lagrange interpolation; if the pilot's operation causes a time delay in the change of the aircraft's attitude, the time delay can also be corrected by the calculus relationship of the flight parameters, etc.

[0029] Step 4: After data processing, preliminary fitting can be performed. The aerodynamic model can be adjusted according to the fitting results. On the one hand, an offset can be directly added to the aerodynamic coefficient, and on the other hand, the aerodynamic derivatives corresponding to the aerodynamic parameters can be modified. Ultimately, the flight simulation data and the actual flight data meet the compatibility requirements of this type of aircraft, and the verification of the flight simulation model is completed.

[0030] like Figure 1 As shown, the aircraft simulation model verification method includes the following specific steps:

[0031] S1: Determine the flight mode in which the pilot operates the aircraft and record the flight data;

[0032] For the dynamic operation response of a 6-DOF aircraft, there are five main flight modes, including: short-period pitch motion, long-period pitch motion, Dutch roll, roll, and spiral.

[0033] Taking short-period pitch motion as an example, this flight mode is the short-period vibration of the aircraft on the pitch axis, which can be expressed as:

[0034] s 2 +2ζ s ω s s+ω s 2 =0

[0035] Among them, s is the point of the motion trajectory, ζ s represents the damping coefficient, ω s Indicates the natural frequency.

[0036] The numerical changes of short-period pitch motion are mainly reflected in the pitch angle θ, the angle of attack (angle of attack) α and the pitch rate q. The frequency of this mode is generally 0.5-2Hz. For the pilot, the aircraft needs to quickly return to the trimmed angle of attack (AoA) before the pitch vibration. In this mode, the pilot will clearly feel the short-period oscillation by pulse excitation of the elevator, and record the flight data of the entire stage through the test flight data acquisition sensor system.

[0037] For the recorded flight data, the flight data that needs to be collected include: flight log time, flight altitude, three rudder deflection instructions, throttle instructions, flap position instructions, longitudinal, lateral and roll-on attitude angles and angular rates, true airspeed, indicated airspeed, speed and acceleration in the ground coordinate system, remaining fuel, air pressure and temperature, engine speed, etc., and the unit of each parameter must be clearly indicated.

[0038] S2: Establishing a flight simulation model: After obtaining the flight data, the aircraft simulation model is established in combination with the existing shape parameters of the aircraft;

[0039] When establishing an aircraft simulation model, it is necessary to create a Simulin simulation model according to the aerodynamic shape and layout of the target aircraft. The Simulin simulation model includes 12th-order motion equations, aerodynamic model, engine model, landing gear model, center of gravity model, and atmosphere model. The created simulation model is then corrected using real aircraft flight data, including parameter verification of aerodynamic characteristics. By inputting the same wing and rudder operation excitations (wing and elevator deflection) as the test flight operation into the flight simulation model, the obtained simulation data will be used for comparison with the real test flight data, so that the accuracy of the aerodynamic parameters can be determined.

[0040] The flight simulation model mainly includes the following key parts:

[0041] 1. Aerodynamic Model

[0042] The aerodynamic module is the most important part of the entire flight simulation model. The main parameters that can be adjusted for model verification are also in this part. This part is for the calculation of the force coefficients and torque coefficients of the three coordinate axes, where the force coefficients include the lift coefficient, the drag coefficient and the sideslip force coefficient, and the torque coefficients include the pitch moment coefficient, the rolling moment coefficient and the yaw moment coefficient. Each coefficient is calculated by combining multiple sub-coefficient values. The sub-coefficient refers to the influence factor of the change caused by a certain action of the aircraft on the change of the coefficient. Different aircraft have a different set of aerodynamic coefficients. The accuracy of this module will directly affect the reliability of the entire flight simulation model.

[0043] 2. Propulsion System Model

[0044] The engine mainly provides thrust to the aircraft, driving it forward and generating lift for flight. The thrust and torque of the engine are calculated by this module, and the obtained values ​​will be combined with the aerodynamic force and torque to calculate the total force and total torque of the aircraft. The initialization values ​​of the engine module mainly depend on the different types of engines, such as the turbojet, turbofan and propfan engines of large aircraft, the thrust tail rotor of small aircraft or drones, etc. The calculation of the engine module will rely on the following values: engine control parameter variables, atmospheric environment parameters (mainly air pressure), flight altitude and airspeed. The output variables obtained from these parameters are engine thrust, torque, engine speed and real-time fuel consumption.

[0045] 3. Aerodynamic force and torque calculation module

[0046] This module will convert the force coefficients and moment coefficients calculated by the aerodynamic module into forces and moments respectively. This will be combined with the center position and pressure center position of the aircraft. The calculation results will be combined with the forces and moments calculated by the engine module to obtain the overall force and moment of the aircraft.

[0047] 4. Aircraft mass and moment of inertia module

[0048] The consumption of aviation fuel will cause the aircraft's moment of inertia and center of gravity to change. However, these two parameters will participate in the calculation of flight kinematics, so this module will calculate the aircraft's mass, moment of inertia, moment of inertia change rate, center of gravity position, and center of gravity position change according to the real-time fuel consumption. The final output will be combined with the aircraft's total force and total torque to calculate the aircraft's kinematic equations.

[0049] 5. Overall force and torque module

[0050] This part will combine the aerodynamic force and torque calculation module, the forces and torques calculated by the engine module, and the center of gravity position calculated in real time to calculate the overall forces and torques of the aircraft. These values ​​will be used to calculate the kinematic equations of the aircraft.

[0051] 6. Kinematics Module (Equation of Motion)

[0052] The kinematics module in the aircraft simulation model is mainly based on the 6-DOF center of mass motion equation. The 12th-order motion parameters of the aircraft will undergo three main calculation steps in this module. The first is the rotational dynamics equation (Rotational Dynamic). This step will calculate the first-order derivatives of the three components uvw of the aircraft velocity along the body coordinates and the angular velocity component pqr. The angular velocity component pqr obtained by the integrator will then be used for the next step of calculation, namely dynamics (Kinematics). The Euler angle obtained after this step of calculation will eventually be obtained through the third step of the direction cosine matrix (Direction Cosine Matrix) to finally obtain the position of the aircraft. The parameters of this module will be used for compatibility testing to assess the accuracy of the aerodynamic derivatives and the force and torque coefficients.

[0053] 7. Standard sea level atmospheric parameters module

[0054] This module will calculate the changes in atmospheric parameters caused by altitude changes in real time, and will participate in the entire aircraft simulation model in a closed loop. This module will recalculate the atmospheric temperature, pressure, air density at the altitude, and air velocity based on the final altitude value calculated by the system.

[0055] After establishing the flight simulation model as described above, it is also necessary to establish a corresponding numerical monitoring system that can display the flight parameters, flight animation and flight trajectory of the aircraft model in real time.

[0056] S3: Correct and check the compatibility of flight data and simulation data;

[0057] In order to verify the accuracy and effectiveness of the model, it is necessary to compare the flight data and flight simulation data, and then verify whether the two sets of data meet the compatibility requirements. The flight simulation data comes from giving the flight simulation model the same excitation according to the pilot's operation, and then extracting the flight parameters that need to be compared in the kinematic module. Let S be the test flight data and T be the simulation solution result data, and the following error analysis calculation can be performed:

[0058] Mean Square Error:

[0059] e=sqrt(|T1-S1| 2 +|T2+S2| 2 +…+|T n -S n | 2 ) / n

[0060] Here n represents the dimension of the two sets of data, and Tn and Sn are the components of vectors S and T respectively.

[0061] According to the tolerance range of the Aircraft Flight Training Device Identification Performance Standard 2019 (AC-60-FS-2019-007), important test indicators, such as the time history curves of altitude, speed, and pitch angle, should be compared with the flight parameter data to ensure that they are within the allowable tolerance range. Different processing can be done according to the actual situation. If the data collected by the aircraft has high-frequency noise, a low-pass filter can be designed; if the flight data has many wild value points, the wild value points can be corrected by Lagrange interpolation; if the pilot's operation causes a time delay in the change of the aircraft's attitude, the time delay can also be corrected by the calculus relationship of the flight parameters, etc.

[0062] S4: Determine and identify the aerodynamic derivatives that need to be corrected according to the compatibility test results and correct them;

[0063] After data processing, preliminary fitting can be performed, and the aerodynamic model can be adjusted according to the fitting results. On the one hand, the aerodynamic coefficient can be directly added with an offset, and on the other hand, the aerodynamic derivatives corresponding to the aerodynamic parameters can be modified. Finally, the flight simulation data and the real flight data meet the compatibility requirements of this type of aircraft, and the verification of the flight simulation model is completed. For the flight simulation model, one of the core steps is to output the overall force and torque of the aircraft by inputting the adjusted force and torque coefficients, and finally simulate the flight of the aircraft through the kinematic module. Each of the identified parameters is calculated jointly by the aerodynamic derivatives of the aircraft and the controllable variables of the aircraft. All identifiable parameters and numerical calculation formulas are listed below:

[0064] 1. Lift Coefficient

[0065] Lift coefficient C L It is calculated by the following formula:

[0066]

[0067]

[0068] 2. Drag Coefficient

[0069] Drag coefficient C D It is calculated by the following formula:

[0070]

[0071]

[0072]

[0073] 3. Side Force Coefficient Side Force Coefficient C Y It is calculated by the following formula:

[0074]

[0075]

[0076]

[0077] 4. Pitch Moment Coefficient Pitch Moment Coefficient C m It is calculated by the following formula:

[0078]

[0079]

[0080]

[0081] 5. Roll Moment Coefficient Roll Moment Coefficient C l It is calculated by the following formula:

[0082]

[0083]

[0084] 6. Yaw Moment Coefficient Yaw Moment Coefficient C n It is calculated by the following formula:

[0085]

[0086]

[0087] The correction method is mainly divided into two steps: the first step is macro observation. A single excitation of the controllable parameters of the aircraft attitude change (elevator angle, rudder angle, aileron angle) will lead to response changes in speed and attitude angle. Under the same excitation conditions, by comparing the curves of the response values ​​of the flight data and the simulation data, the deviation direction and approximate deviation range of the force and moment coefficients can be determined first.

[0088] The second step is to determine the specific aerodynamic derivative or derivatives that need to be adjusted through the control variable method, and then observe the curve fitting after adjustment by pulling the value by 5%, 10% or 15%, and finally determine the value through multiple adjustments.

[0089] The aerodynamic and control characteristics of the simulation model are verified using real aircraft flight data, and the aerodynamic derivatives of the aircraft simulation model are corrected, ultimately making the aircraft simulation model meet the tolerance requirements of the test data time history of each subject in the simulator certification standards.

[0090] The table below lists the model calibration example values ​​for a certain type of aircraft.

[0091] 1. Lift coefficient identification and correction method

[0092] 2. Drag coefficient identification and correction method

[0093]

[0094]

[0095] 3. Identification and correction method of lateral force coefficient

[0096]

[0097] 4. Identification and correction method of pitch moment coefficient

[0098]

[0099] 5. Rolling moment coefficient identification and correction method

[0100]

[0101]

[0102] 6. Identification and correction method of yaw moment coefficient

[0103]

[0104] S5: When the corrected data meets the specifications, the final corrected aerodynamic model parameter table is exported and a debugging report is generated.

[0105] The aircraft simulation model verification method can make the aircraft simulation model modeling more accurate and more efficient, and the method is replicable and can be applied to various aircraft such as manned aircraft and unmanned aircraft. The flight simulation model created by the method can also be applied to tactical training simulators and combat simulations, flight training devices and other national defense digital software, making very practical contributions to national security and military strength.

Claims

1. A method for verifying an aircraft simulation model, comprising the following steps: S1: Determine the flight mode in which the pilot operates the aircraft and record the flight data; S2: After obtaining the flight data, the aircraft simulation model is established in combination with the existing shape parameters of the aircraft; S3: Correct and check the compatibility of flight data and simulation data; S4: Determine and identify the aerodynamic derivatives that need to be corrected according to the compatibility test results and correct them; S5: When the corrected data meets the specifications, the final corrected aerodynamic model parameter table is exported and a debugging report is generated.

2. The aircraft simulation model verification method according to claim 1, characterized in that: In step S2, when establishing an aircraft simulation model, it is necessary to create a Simulin simulation model according to the aerodynamic shape and layout of the target aircraft. The Simulin simulation model includes 12th-order motion equations, aerodynamic model, engine model, landing gear model, center of gravity model, and atmosphere model. The created simulation model is then corrected using real aircraft flight data, including parameter verification of aerodynamic characteristics.

3. The aircraft simulation model verification method according to claim 1, characterized in that: In step S3, the acquired real aircraft flight data is smoothed, filtered and time-shifted to form a model verification data package, and the compatibility of the data is verified by checking the compatibility with the simulation data.

4. The aircraft simulation model verification method according to claim 1, characterized in that: In step S4, the aerodynamic characteristics and control characteristics of the simulation model are verified using real aircraft flight data, and the aerodynamic derivatives of the aircraft simulation model are corrected, so that the aircraft simulation model finally meets the tolerance requirements of the test data time history of each subject according to the simulator identification standard.

5. The aircraft simulation model verification method according to claim 1, characterized in that: In step S1, the flight data to be collected include: flight log time, flight altitude, deflection commands of three rudders, throttle commands, flap position commands, longitudinal, lateral and roll-on attitude angles and angular rates, true airspeed, indicated airspeed, speed and acceleration in the ground coordinate system, remaining fuel, air pressure and temperature, and engine speed.

6. The aircraft simulation model verification method according to claim 1, characterized in that: The aircraft simulation model comprises an aerodynamic module, an engine module, an aerodynamic force and moment calculation module, an aircraft mass and inertia moment module, an overall force and moment module, a kinematics module and a standard sea level atmospheric parameter module.

7. The aircraft simulation model verification method according to claim 1, characterized in that: In step S3, the flight simulation data comes from giving the same excitation to the flight simulation model according to the pilot's operation, and then extracting the flight parameters required for comparison in the kinematics module. Let S be the test flight data and T be the simulation solution result data, and the following error analysis calculation can be performed: Mean Square Error: e=sqrt(|T1-S1| 2 +|T2+S2| 2 +…+|T n -S n | 2 ) / n Here n represents the dimension of the two sets of data, and Tn and Sn are the components of vectors S and T respectively.

8. The aircraft simulation model verification method according to claim 1, characterized in that: In step S4, a preliminary fitting is performed after data processing, and the aerodynamic model is adjusted according to the fitting results. On the one hand, an offset is directly added to the aerodynamic coefficient, and on the other hand, the aerodynamic derivatives corresponding to the aerodynamic parameters are modified. Finally, the flight simulation data and the actual flight data meet the compatibility requirements of this type of aircraft, and the verification of the flight simulation model is completed.

9. The aircraft simulation model calibration method according to claim 1, characterized in that: In step S4, the aerodynamic derivatives include a lift coefficient, a drag coefficient, a side force coefficient, a pitch moment coefficient, a rolling moment coefficient and a yaw moment coefficient.

10. The aircraft simulation model verification method according to claim 9, characterized in that: In step S4, the correction method includes: the first step is to observe through macroscopic observation that a single excitation of a controllable parameter of the aircraft attitude change will lead to a response change of the speed and attitude angle. Under the same excitation condition, by comparing the curves of the response values ​​of the flight data and the simulation data, the deviation direction and the approximate deviation range of the force and moment coefficients are first determined; The second step is to determine the specific aerodynamic derivative or derivatives that need to be adjusted through the control variable method, and then observe the curve fitting after adjustment by pulling the deviation value by 5%, 10% or 15%, and finally determine the value through multiple adjustments.

Images