Rigid-elastic coupling flight simulation platform of large flexible aircraft and control method of rigid-elastic coupling flight simulation platform

Abstract

The invention discloses a rigid-elastic coupling flight simulation platform of a large flexible aircraft and a control method of the rigid-elastic coupling flight simulation platform. The rigid-elastic coupling flight simulation platform comprises an upper motion platform, a lower fixed base, six electric servo actuators forming a parallel structure, a control system and a main controller. The method is characterized in that hinge points of six actuators on a lower fixed base adopt a non-concyclic optimized layout, the distance between the hinge points in the pitching axis direction of the platform is remarkably increased, a base geometric configuration longer in the longitudinal direction is formed, then the platform is embedded into an advanced rigid-elastic coupling flight mechanical model, and the model is a three-dimensional model. According to the method, effective verification of key flight events which are difficult to accurately simulate by a traditional six-degree-of-freedom platform can be realized, such as wing flutter boundary exploration of a large-flexibility aircraft, structural load coupling analysis under large maneuver, flight control law performance evaluation under rigid-elastic coupling influence and the like. Physical motion of the simulation platform can accurately reproduce complex dynamic characteristics predicted by the model, so that credibility and engineering application value of flutter simulation results are greatly improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rigid-elastic coupling flight simulation, and in particular to a flight simulation platform for large flexible aircraft. BACKGROUND

[0002] The six-degree-of-freedom (6-DOF) motion platform is the core physical device of a flight simulator. However, in response to the simulation challenges of new-generation aircraft such as large flexible aircraft, the current mainstream technology faces a double bottleneck from the computing core and the physical execution level. First, at the computing core level, the existing platform generally uses a traditional rigid-body flight dynamics model, which fundamentally cannot describe the coupling between structural elastic deformation and aerodynamic force in flight, resulting in missing simulation data sources for key physical phenomena such as flutter and buffeting. Second, at the physical execution level, the classic symmetrical configuration of the Stewart platform, as a motion reproduction carrier, results in a very limited pitch motion range (usually within ±20°), so even with more accurate data, it is physically impossible to reproduce large-angle maneuvers such as high angles of attack and overspeed. This coexistence of "model inaccuracies" and "platform limitations" severely restricts the high-fidelity simulation of the flight characteristics of large flexible aircraft and is a technical barrier that needs to be broken through in this field. SUMMARY

[0003] In view of the two technical pain points that the existing technology generally lacks a rigid-elastic coupling flight dynamics model that can accurately describe the physical characteristics of large flexible aircraft, and the pitch angle range of the physical platform is severely limited due to geometric constraints, the present application provides a rigid-elastic coupling flight simulation platform for large flexible aircraft and a control method thereof, which expands the available pitch angle to more than ±30°, and introduces an advanced rigid-elastic coupling flight dynamics model and gives the platform a high dynamic response capability.

[0004] The core of the technical solution is the collaborative design of the hardware physical configuration and the software core model, which is embodied in the following two aspects:

[0005] A control method for a rigid-elastic coupling flight simulation platform for large flexible aircraft, comprising the following steps:

[0006] S101, obtain the set value of the throttle and the rocker, i.e. the control signal, according to the control signal, the main controller in the control system receives the large flexible aircraft full-state data generated by the rigid-elastic coupling flight dynamics model in real time, and the control system receives the flight control feedback signal;

[0007] S102, use a motion excitation algorithm to analyze, filter and map the full-state data, generate high-fidelity reproduced rigid-elastic coupling effect upper motion platform target motion instructions, and the upper motion platform target motion instructions conform to the physical travel and dynamic response limitations of the servo cylinder.

[0008] S103, calculating the required extension lengths of the six electric servo actuators based on the kinematic inverse solution model according to the target motion instruction of the upper motion platform;

[0009] S104, sending the extension length instruction to each servo driver by the control system, driving the actuator to move, and performing closed-loop control according to the feedback of the displacement sensor, so that the upper motion platform can reproduce the target motion driven by the rigid-elastic coupling flight dynamics model.

[0010] Further, the full state data includes rigid body motion information and elastic modal information of key structures.

[0011] Further, the motion excitation algorithm in step S102 is a multi-channel and multi-target signal processing and fusion framework, which can analyze the full state data stream transmitted in step S101 in real time, and decompose the full state data into two parallel signal channels according to the frequency and physical meaning:

[0012] The low-frequency large-amplitude rigid body motion channel includes the attitude, angular velocity and linear acceleration of the aircraft, and is used to reproduce the pilot's perception of the macro motion of the aircraft in the maneuver and overload;

[0013] The high-frequency fine structure vibration channel includes the displacement and velocity of each elastic modal, and is used to reproduce the pilot's perception of the micro structure dynamics of the aircraft in the flutter and buffeting under specific flight conditions.

[0014] Further, the kinematic inverse solution model in step S103 is:

[0015] A nonlinear equation set based on the geometric constraints of the non-circular hinge points on the lower fixed base is established, and a real-time numerical solution method is used to solve the nonlinear equation set to obtain a complex target motion instruction including low-frequency large dynamic and high-frequency micro-vibration. The kinematic inverse solution algorithm is used to map the large flexible aircraft motion instruction to the length value of the six electric servo actuators (2).

[0016] In a second aspect, the application also provides a rigid-elastic coupling flight simulation platform for a large flexible aircraft, which is controlled by the control method according to claims 1 to 4. The simulation platform comprises a lower fixed base, six electric servo actuators (2) connecting the upper motion platform and the lower fixed base, and an upper motion platform.

[0017] The six electric servo actuator hinge points on the lower fixed base are arranged in a non-circular layout, wherein the longitudinal spacing between the hinge points distributed along the pitch axis direction of the upper motion platform is significantly greater than the transverse spacing between the hinge points distributed along the roll axis direction.

[0018] Further, the six-branch electric servo actuator system comprises six independent electric servo actuators, each electric servo actuator comprising: an alternating current servo motor, a ball screw transmission mechanism connected with the alternating current servo motor, and a displacement sensor for monitoring the length of the electric servo actuator in real time.

[0019] Further, the simulation platform further comprises a control system and a main controller connected with the six-branch electric servo actuators (2).

[0020] The control system comprises:

[0021] a main controller for running a core algorithm, six servo drives corresponding to the electric servo actuators respectively, and an I / O interface module for data interaction with the main controller.

[0022] Further, the main controller adopts a real-time computing architecture based on a digital signal processor (DSP) or a field programmable gate array (FPGA).

[0023] The main controller and each servo drive communicate through a real-time industrial bus.

[0024] Further, the pitch angle range of the upper motion platform is not less than ±25°.

[0025] Advantages:

[0026] The present application has one or more of the following advantages through the cooperative design of software and hardware compared with the prior art:

[0027] 1. Cross-generation improvement of simulation capability: The present application not only expands the available pitch angle to more than ±30° through optimization of geometric layout, meeting the simulation needs of extreme rigid body maneuvering, but more importantly, by introducing an advanced rigid-elastic coupling flight dynamics model and giving the platform corresponding high dynamic response capability, it is for the first time possible to reproduce key aeroelastic phenomena such as flutter and buffeting on a physical platform with high fidelity. This makes the simulation capability cross from the traditional rigid body field to the more complex and advanced rigid-elastic coupling field.

[0028] 2. High cost performance and engineering feasibility: The present application realizes the cross-generation development of the above simulation capability through the ingenious combination of structural innovation and software upgrade without significantly increasing the cost of actuator stroke or changing the basic complexity of the platform (hexapod configuration), and the scheme is extremely economical and has great engineering promotion value.

[0029] 3. Excellent dynamic fidelity: The optimized configuration combined with high-speed control algorithm gives full play to the inherent advantages of high rigidity and high precision of the parallel electric platform. This ensures that the platform can accurately track and reproduce the complex wideband instructions generated by the rigid-elastic coupling model, including low-frequency large dynamics (such as high-angle maneuver) and high-frequency micro-vibration (such as flutter), providing pilots or test systems with unprecedented, highly consistent physical motion excitation. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 It is a schematic diagram of the overall structure of the flight simulation platform of the large flexible aircraft of the present application.

[0031] Figure 2 It is a schematic diagram of the geometric layout comparison before and after the optimization of the lower fixed base hinge point of the present application.

[0032] Figure 3 It is a performance comparison diagram of the pitch angle workspace before and after optimization of the present application.

[0033] Figure 4 It is a schematic diagram of the structure of the upper moving platform and the simulation driving seat of the present application.

[0034] Figure 5 It is a schematic diagram of the structure of the lower fixed base of the present application.

[0035] Figure 6 It is a schematic diagram of the electric servo actuator used in the present application.

[0036] Figure 7 It is a simulation visualization diagram of the kinematic inverse solution model of the present application.

[0037] Figure 8 It is a graph of the motion state of the large flexible aircraft changing with time.

[0038] Figure 9 It is a graph of the displacement of each mode of the large flexible aircraft changing with time.

[0039] Figure 10 It is a control implementation flowchart of the rigid-elastic coupling flight simulation platform of the large flexible aircraft.

[0040] In the figure, 1 is a lower fixed base, 1-1 is a step ladder, 1-2 is a base bottom plate, 1-3 is a hinged round hole, 1-4 is an aviation plug, 1-5 is a control cabinet cooling fan, 1-6 is a control cabinet shell, 1-7 is a cable channel, 1-8 is an indicator light, 1-9 is a lower hinged round hole for mounting an actuator; 2 is an electric servo actuator (electric cylinder), 2-1 is a servo driver, 2-2 is an actuator cylinder, 2-3 is a hooke joint, 2-4 is a servo driver support; 3 is an upper movement platform, 3-1 is a flight simulation seat, 3-2 is a bent pipe, 3-4 is a rubber plug, 3-5 is a display support, 3-6 is a driving rod, 3-7 is a throttle valve, 3-8 is a footrest, 3-9 is a display, 3-10 is a display support pipe, and 3-11 is an upper movement platform frame. DETAILED DESCRIPTION

[0041] In order to facilitate the understanding of those skilled in the art, the present application will be further described below in conjunction with the embodiments and the accompanying drawings. The content mentioned in the embodiments is not a limitation of the present application.

[0042] Embodiment 1:

[0043] The embodiment provides a rigid-flexible coupling flight simulation platform of a large flexible aircraft.

[0044] Referring to Figure 1 , the overall structure of the present application includes a lower fixed base 1, six electric servo actuators 2 connecting the upper and lower platforms, and an upper movement platform 3. The whole system is controlled by the main controller and control system integrated in the cabinet shown. Figure 5

[0045] Referring to Figure 4 , the upper movement platform 3 is integrated with a simulated cockpit, which mainly includes a bent pipe 3-2 and an upper movement platform frame 3-11 for support; a flight simulation seat 3-1, a driving rod 3-6, a throttle valve 3-7 and a footrest 3-8 for human-computer interaction; and a flight visualization system composed of a display 3-9, a display support pipe 3-10, a display support 3-5 and a display wall connecting plate 3-11.

[0046] Referring to Figure 5 , the lower fixed base 1 is a solid metal platform, which provides a stable installation and support foundation for the whole device. Its structure includes a base bottom plate 1-2, a lower hinged round hole 1-9 for mounting an actuator, and an integrated control cabinet shell 1-6. The cabinet body is provided with a control cabinet cooling port 1-5, an indicator light 1-8, an aviation plug 1-4 and a cable channel 1-7, and is equipped with a step ladder 1-1 for the pilot to go up and down.

[0047] Referring to Figure 6 ​The electric servo actuator leg system 2 is the core execution mechanism of the turntable. It is mainly composed of an actuator cylinder 2-2 and a hooke joint 2-3 for connecting upper and lower platforms. Each actuator is precisely controlled by an independent servo driver 2-1.

[0048] The technical scheme of the present application discards the mode of combining a simplified model and a general platform in traditional flight simulation, and proposes a new idea of driving an adaptive hardware structure by a rigid-elastic coupling flight mechanics model.

[0049] The core of the scheme is firstly embedding an advanced rigid-elastic coupling flight mechanics model specially developed for a large flexible aircraft. The model no longer regards the aircraft as a rigid body, but solves in real time as a multi-body elastic system, and can accurately capture key physical phenomena such as wing deformation and aeroelastic flutter in flight, which cannot be captured by traditional models.

[0050] In order to convert the high-dimensional data stream containing complex coupling effects output by the model into real physical motion, the present application then makes targeted innovations on the hardware: as shown in Figure 2 The six actuator hinge points of the lower fixed base 1 adopt an optimized non-circular geometric layout (solid line), which is in sharp contrast with the conventional circular design (dashed line). The base length along the platform pitch axis (X-axis direction in the figure) is significantly lengthened, while the width along the roll axis (Y-axis direction in the figure) is relatively shortened.

[0051] This hardware design is aimed at meeting the dual stringent requirements of the rigid-elastic coupling model for large pitch angle range and high frequency dynamic response, thereby providing a hardware foundation for high-fidelity physical simulation.

[0052] Embodiment 2:

[0053] The control method of the rigid-elastic coupling flight simulation platform for the large flexible aircraft in embodiment 1 is essentially to convert the complex thinking process of the rigid-elastic coupling flight mechanics model, which is the simulation brain, into the precise action of the physical platform "strong torso", and its process closely revolves around the model:

[0054] S101, the main controller runs the flight simulation software, the core of which is an advanced rigid-elastic coupling flight mechanics model. According to the pilot's operation, the model performs real-time multi-body system dynamics calculation and outputs not only rigid body motion parameters (attitude, angular velocity, linear acceleration), but also high-dimensional, full-state data stream of elastic modes such as wing deformation, structural chattering and aeroelastic flutter, which are crucial to simulation.

[0055] S102, the control system located in the 1-6 cabinet receives the high-dimensional data stream, and intelligently analyzes and fuses through a multi-channel motion excitation algorithm. The algorithm translates the rigid-elastic coupling effect output by the model into a unified target motion instruction that the platform can execute, which simultaneously contains high-frequency vibration information such as nonlinear body sensation and flutter at large angles of attack.

[0056] S103, the control system located in the 1-6 cabinet calls a high-speed real-time inverse model (as shown in Figure 7 ) optimized for the specific non-circular hinge point layout of the invention, and accurately inversely solves the length instruction generated in the previous step, which contains complex coupling effects, into the extension length that each electric servo actuator 2-2 needs to achieve.

[0057] S104, the control system located in the 1-6 cabinet sends the length instruction to the servo driver 2-1 of each actuator 2, drives the upper motion platform 3 to perform high-dynamic and high-precision physical motion, and ensures that the final action is consistent with the original calculation of the rigid-elastic coupling model through closed-loop control.

[0058] To verify the effectiveness of the optimized layout of the invention, we simulated and compared the working space of the platform before and after optimization.

[0059] Referring to Figure 2 , the two schemes compared are: the conventional co-circular hinge point layout and the longitudinal elongated non-circular hinge point layout proposed by the invention.

[0060] Referring to Figure 3 , the chart shows the comparison results of the pitch angle working range of the two layouts at different platform heights. It can be seen that the pitch angle working space (light area in the figure) before optimization is relatively small. After using the optimized layout of the invention, the optimized pitch angle working space (dark area in the figure) is greatly expanded.

[0061] Taking the height of the platform as an example, the pitch angle range of the pre-optimization layout is about -16° to +21°. The pitch angle range of the optimized layout of the invention at this height is expanded to about -12° to +28°, with an increase of about 33% in positive pitch angle capability and a significant increase in total pitch angle travel.

[0062] The simulation results clearly confirm that the optimized layout of the invention can effectively expand the pitch motion range of the platform without increasing the cost and complexity of the actuator, thereby being able to meet the simulation needs of large-angle flight attitude with high fidelity.

[0063] In S101, the flight simulation platform embeds an advanced rigid-elastic coupling flight mechanics model. Its built-in flight mechanics state equation is as follows:

[0064] ;

[0065] In the examples of the present application, is a 48-dimensional vector, and the physical meaning of each element from the first row to the last row is as follows: linear displacement in the direction, linear displacement in the direction, linear displacement in the direction, angular displacement in the direction, angular displacement in the direction, angular displacement in the direction, first to eighth order modal displacement, rudder deflection angle of the first to tenth rudders at the current time, linear velocity in the direction, linear velocity in the direction, linear velocity in the direction, angular velocity in the direction, angular velocity in the direction, angular velocity in the direction, first to eighth order modal velocity, rudder deflection angle velocity of the first to tenth rudders at the current time: is a 11-dimensional vector, and the physical meaning of each element from the first row to the last row is as follows: engine thrust, rudder deflection angle command of the first to tenth rudders; is a constant matrix in the state space equation, is a 48-dimensional matrix, is a 48x11 matrix. The mutual relationship between the above state vector, control vector and matrix can be seen from the figure of the motion state of the large flexible aircraft changing with time shown in Figure 8 reflects the real-time response of the state variables of the large flexible aircraft in the system to the control command input. Figure 9 reflects the change of the modal displacement of the large flexible aircraft with time.

[0066] Further, in the examples of the present application, a p-k method is built in the computer to solve the body flutter frequency, and then the obtained frequency and amplitude are input in real time to the servo driver (2-1), and after the kinematic inverse solution algorithm, they are transmitted to the electric servo actuator (2-2) for actuation.

[0067] First, the dynamic aeroelastic equation is established in the modal coordinate system, and the equation is as follows:

[0068] ;

[0069] In the examples of the present application, are the modal mass, modal damping and modal stiffness matrices, respectively. is a modal coordinate column vector. is the air density, is the flight speed. is the modal aerodynamic matrix, which is a function of the non-dimensional reduced frequency and Mach number .

[0070] Assuming the system does harmonic oscillation when it is in flutter, the solution of the equation is in the form of:

[0071] ;

[0072] In the examples of the present invention, is a complex number, . is the transient decay rate coefficient, representing the damping of the system. When , the system is in a sustained equal amplitude oscillation state, i.e. the flutter critical state. is the circular frequency of oscillation.

[0073] In order to make the equation dimensionless, the parameter is introduced:

[0074] ;

[0075] In the examples of the present invention, is the reduced frequency (or the scaled frequency), , is the reference half chord length.

[0076] Substituting the assumed solution into the governing equation, a quadratic eigenvalue problem in the form of:

[0077] ;

[0078] is obtained, where is divided into real and imaginary parts, as follows

[0079] ;

[0080] Substituting it into the original equation, we get:

[0081]

[0082] When , the flutter critical speed can be obtained, i.e. there is:

[0083] ;

[0084] Substituting it back, we get:

[0085]

[0086] where,

[0087] ​

[0088] The problem can be solved by using the MATLAB library functions polyeig and eig to transform it into a standard eigenvalue problem and a generalized eigenvalue problem.

[0089] Considering All are reduced frequencies The function, therefore Iterative calculations are required for each given flight speed. and Mach number The iterative process is as follows:

[0090] Step 1: Guessing Value: First, the reduced frequency. Set an initial guess value .

[0091] Step 2: Calculate the aerodynamic matrix Based on speculation Given a Mach number, calculate the aerodynamic matrix. .

[0092] Step 3: Solve for eigenvalues : Calculated Substituting these values ​​into the governing equation, the equation now becomes a standard quadratic eigenvalue problem. Solving this problem yields a set of complex eigenvalues. .

[0093] Step 4: Comparison Value iteration: from the obtained eigenvalues In the middle, extract its imaginary part as the new calculation result. .Compare and If the two are not equal or the difference exceeds the tolerance, then use... As a new Repeat steps two and three until... .

[0094] The above iterative process can be performed at multiple different flight speeds. Repeat the process below.

[0095] Step 1: For each flight speed, find a set of matching feature values ​​through iteration. .

[0096] Step 2: Take each feature value The real part, i.e., the dimensionless damping of the system. .

[0097] Step 3: Plot the damping for all modes. With flight speed The curve of the change, i.e. Fig. (or Fig.) and Fig.

[0098] Step four, solving the flutter speed, flutter frequency, when a certain mode of damping curve from the negative value across the horizontal axis to the positive value (i.e. ), the corresponding flight speed is the flutter speed, the corresponding frequency is the flutter frequency. Wherein, the structural damping can be taken

[0099] ;

[0100] The flutter frequency is

[0101] ;

[0102] The initial flow rate is generally selected as a low flow rate. Since the fluid-structure coupling effect is weak at this time, the iteration initial value of the equivalent frequency at this flow rate can be determined by the structural natural frequency information, and then the iteration is started.

[0103] The first-order flutter frequency of the system is calculated to be about 5Hz.

[0104] In S102, the motion excitation algorithm used by the application is a multi-channel, multi-target signal processing and fusion framework specially designed for high-dimensional, wide-band rigid-elastic coupling flight data. Its fundamental task is no longer the traditional "motion simulation", but to intelligently "translate" and "map" a complex data stream containing 48 state dimensions, which simultaneously contains low-frequency large-amplitude rigid body motion and high-frequency small structure vibration, into 6-dimensional motion instructions that can be reproduced by a six-degree-of-freedom motion platform in a limited physical space with high fidelity.

[0105] The theoretical basis of this algorithm integrates the classical cleaning filter theory and the state observation and signal fusion ideas in modern control theory, and its specific implementation can be decomposed into the following core modules:

[0106] 1. State vector parsing and weighting module

[0107] This module is the entrance of the algorithm. It first receives the 48-dimensional full state vector calculated by the rigid-elastic coupling model on the main controller.

[0108] Data separation: the algorithm decomposes the state vector into three subsets with clear physical meaning:

[0109] Rigid body state set: contains linear / angular displacement and linear / angular velocity of three axes.

[0110] Structural elastic modal state set: contains modal displacement and modal velocity of up to eight orders.

[0111] Control surface state set: contains deflection angle and angular velocity of ten control surfaces.

[0112] Dynamic weighting: one of the innovations of this module is that it can dynamically weight different state components according to the current flight state (such as Mach number, dynamic pressure, angle of attack, etc.). For example, when performing high-angle-of-attack maneuvers, the weight of the rigid body acceleration component is increased to highlight the sensation of overload; when approaching the flutter boundary, the weight of the key elastic modal component is highlighted to enhance the perception of structural vibration.

[0113] 2. Low-frequency motion excitation channel (macroscopic somatosensory simulation)

[0114] This channel is responsible for processing the rigid body state set, aiming to reproduce the pilot's macroscopic perception of the aircraft's overall maneuver.

[0115] Classic washout filter: the algorithm uses an optimized third-order or higher-order washout filter structure.

[0116] Translation channel: by high-pass filtering linear acceleration, it accurately reproduces the instantaneous sensation of pushback, sideslip, etc. At the same time, it uses the "impulse response back to home" feature to slowly "wash" the platform back to the center of the workspace, thereby saving and recycling limited physical travel.

[0117] Rotation channel: by high-pass filtering angular velocity, it reproduces the instantaneous rotation, while using tilt coordination (Tilt-Coordination) technology to convert the continuous linear acceleration signal into a slow tilt of the platform through low-pass filtering, and using the gravity component in the human body coordinate system to generate an equivalent continuous acceleration sensation.

[0118] Nonlinear optimization: the parameters of the filter (such as cutoff frequency, damping ratio) are specially optimized and scheduled for the nonlinear aerodynamic characteristics of high-angle-of-attack, stall, etc. flight, ensuring that the somatosensory simulation is not distorted and does not produce false sensations in extreme attitudes.

[0119] 3. High-frequency vibration excitation channel (microscopic somatosensory simulation)

[0120] This channel is the key innovation of the invention and is specifically used to process the structural elastic modal state set, aiming to reproduce the pilot's microscopic perception of the aircraft's structural vibration.

[0121] Modal contribution synthesis: the algorithm linearly superimposes modal displacements and velocities up to the 8th order, weighted by the state-dependent weights output by the weighting module, to synthesize an equivalent vibration signal that represents the main structural vibration characteristics at the moment.

[0122] Vibration signal mapping: the synthesized high-frequency vibration signal is directly mapped to specific degrees of freedom of the motion platform after reasonable scaling and amplitude limiting. For example, the first-order symmetric bending mode of the wing is mainly mapped to the vertical (Z-axis) motion of the platform; the anti-symmetric torsional mode of the wing may be mapped to the roll (Roll-axis) motion of the platform.

[0123] Application purpose: this channel enables the pilot to truly "feel" the initiation of aeroelastic flutter, the intensity changes of buffeting, or the response of the aircraft when crossing a gust, which is irreplaceable for verifying the aeroelastic characteristics of the aircraft and developing active control laws such as flutter active suppression or gust load alleviation.

[0124] 4. Multi-channel instruction fusion and safety limiting module

[0125] This module is the output of the algorithm, responsible for integrating the outputs of each channel into the final platform driving instructions.

[0126] Instruction vector fusion: the 6-dimensional motion instructions output by the low-frequency channel are vector superimposed with the 6-dimensional vibration instructions output by the high-frequency channel to form a unified expected motion instruction that contains complete rigid-elastic coupling effects.

[0127] State prediction and limiting: before outputting the final instructions, the algorithm will make predictive judgments based on the actuator model of the platform (including stroke, speed, and acceleration limits) to the fused instructions. If it is predicted that the instructions may cause the platform to exceed any physical limits in the future several control periods, the limiter will smoothly adjust the instructions, giving priority to the macro-motion instructions with larger amplitude and lower frequency, while appropriately attenuating the high-frequency vibration instructions, thereby maximizing the simulation fidelity while ensuring safety.

[0128] In summary, this motion excitation algorithm successfully converts high-dimensional rigid-elastic coupling simulation data that traditional models cannot reach into physical motions that pilots can perceive and have high fidelity through its ingenious multi-channel separation, processing, and fusion architecture, and is a key core technology that connects advanced simulation models and high-performance physical platforms.

[0129] The kinematic inverse solution model described in step S103 of the present application has a core task of calculating the length of each of the six electric servo actuators (2) required to reach the target pose of the upper moving platform (3) according to the platform target motion instruction generated by the control system in real time and accurately. The model is established based on a specific non-circular geometry configuration, and the solving process does not involve iteration, so the calculation efficiency is high and meets the real-time control requirements.

[0130] 1. Model establishment premise and known parameters:

[0131] To solve, first define the following coordinate systems and known geometric parameters:

[0132] Lower fixed base coordinate system : A fixed world coordinate system, the origin is usually set at the geometric center of the lower fixed base (1), and the coordinate system follows the Cartesian rule.

[0133] Upper moving platform coordinate system : A moving body coordinate system, the origin is set at the geometric center of the upper moving platform (3).

[0134] Lower hinge point vector : The position vector of the six lower hinge points relative to the base coordinate system origin . The coordinates of the six vectors are known and fixed, and constitute the unique non-circular geometric layout of the present application.

[0135] Upper hinge point vector : The position vector of the six upper hinge points relative to the platform coordinate system origin . The coordinates of the six vectors are also known and fixed.

[0136] 2. For the target motion instruction generated by the motion excitation algorithm (S102) at each control cycle, the inverse solution model calculates in real time by the following steps:

[0137] First step: Determine the mathematical description of the target pose.

[0138] Extract the target pose of the upper moving platform (3) in the lower fixed base coordinate system from the target motion instruction, and describe it as:

[0139] A translation vector : represents the target position of the upper moving platform coordinate system origin relative to the base coordinate system origin , and the formula is as follows:

[0140] .

[0141] A rotation matrix : represents the target pose of the upper moving platform coordinate system relative to the base coordinate system . This 3x3 matrix is calculated from the target pose angles (such as roll angle , pitch angle , yaw angle ) as follows:

[0142] .

[0143] Step 2: Calculate the global position of each upper hinge point in the base coordinate system.

[0144] For each upper hinge point vector fixed in the upper moving platform coordinate system , its instantaneous global position vector in the base coordinate system is calculated through the above translation and rotation transformation.

[0145] .

[0146] Step 3: Calculate the instantaneous length vector of each actuator.

[0147] For each actuator, its instantaneous length vector is a directed line segment connecting its lower hinge point (defined by vector ) and the instantaneous global position of the upper hinge point (defined by vector ), as follows:

[0148] .

[0149] Step 4: Solve the target extension length of each actuator.

[0150] The target extension length that the i-th actuator needs to achieve is the Euclidean norm (i.e., the module length) of its instantaneous length vector .

[0151] .

[0152] Through the analysis and calculation of the above four steps, the control system can accurately and efficiently back-solve the six one-dimensional actuator length commands from any six-degree-of-freedom platform target pose command. These length commands are then sent to the servo driver (S104) to complete the closed-loop control of the platform.​

[0153] Step S104 is a high-fidelity closed loop execution and physical reproduction stage. In this step, the control system will issue the six-way target length command calculated in S103 to the servo drive (2-1) of each electric servo actuator (2) through a real-time industrial bus (such as EtherCAT) at a high speed.

[0154] The high-precision displacement sensor (such as a grating ruler or a magnetostrictive sensor) built in each actuator will input its instantaneous extension length as a feedback signal to the controller together with the command signal, forming a high-bandwidth, high-refresh-rate closed-loop servo loop. The optimized PID (proportional-integral-derivative) control law runs inside the loop, which calculates the position error in real time and dynamically adjusts the drive output to the servo motor.

[0155] The closed-loop control loop is specially designed to ensure that the actuator can accurately track the complex command generated by the motion excitation algorithm, which contains both low-frequency large-amplitude maneuvers and high-frequency small vibrations, and ultimately ensure that the actual physical motion of the upper motion platform (3) is a high-fidelity, delay-free final presentation of the original calculation intention of the rigid-elastic coupling model.

[0156] In summary, the present application not only improves the traditional six-degree-of-freedom platform locally, but also systematically restructures from the core algorithm to the physical structure. Its essence lies in using an advanced rigid-elastic coupling flight mechanics model that can accurately describe the physical characteristics of a large flexible aircraft as the "brain" to drive and define a suitable "robust torso" with non-circular optimization geometry. This deep hardware-software collaboration design fundamentally breaks the dual bottleneck of model accuracy and physical motion capability in traditional simulation technology. It enables the high-fidelity unified reproduction of extreme complex flight phenomena such as high-angle-of-attack high-dynamic maneuver and aeroelastic flutter on the same physical platform. Ultimately, the present application provides a wideband, high-fidelity high-performance flight simulator for the research, testing, and pilot training of the new generation of aircraft.

Claims

1. A control method for a rigid-elastic coupling flight simulation platform for a large flexible aircraft, characterized in that, Includes the following steps: S101. Obtain the throttle and joystick settings, i.e., the control signals. Based on the control signals, the main controller in the control system receives the full state data of the large flexible aircraft generated by the rigid-elastic coupling flight mechanics model in real time, and the control system receives the flight control feedback signal. S102. The motion excitation algorithm is used to analyze, filter and map the full state data to generate a high-fidelity reproduction of the upper motion platform target motion command of the rigid-elastic coupling effect. The upper motion platform target motion command conforms to the physical stroke and dynamic response limitations of the servo electric cylinder. S103. Based on the target motion command of the upper motion platform, calculate the extension length that each of the six electric servo actuators needs to achieve based on the inverse kinematics model. S104. The control system sends the instruction for the extension length to each servo driver, drives the actuator to move, and performs closed-loop control based on the feedback from the displacement sensor, so that the upper motion platform can reproduce the target motion driven by the rigid-elastic coupling flight mechanics model.

2. The control method as described in claim 1, characterized in that, The full-state data includes rigid body motion information and elastic modal information of key structures.

3. The control method as described in claim 1, characterized in that, The motion excitation algorithm described in step S102 is a multi-channel, multi-target signal processing and fusion framework. It performs real-time analysis on the full-state data stream input in step S101, decomposing the full-state data into two parallel signal channels according to frequency and physical meaning: Low-frequency, large-amplitude rigid body motion channel: includes the aircraft's attitude, angular velocity, and linear acceleration, used to reproduce the pilot's perception of macroscopic motion during maneuvers and overloads; High-frequency microstructure vibration channel: contains displacement and velocity of various elastic modes, used to reproduce the pilot's perception of the microstructure dynamics of flutter and buffeting of the aircraft under specific flight conditions.

4. The control method as described in claim 1, characterized in that, The inverse kinematics model mentioned in step S103 is specifically as follows: A set of nonlinear equations based on geometric constraints set at non-circular hinge points on a lower fixed base is established. The nonlinear equations are solved using a real-time numerical solution method to obtain complex target motion commands that include low-frequency large dynamics and high-frequency micro-vibrations. Based on the inverse kinematics algorithm, the motion commands of the highly flexible aircraft are mapped to the length values ​​of six electric servo actuators.

5. A rigid-elastic coupling flight simulation platform for a large flexible aircraft, characterized in that, The simulation platform is controlled by the control method described in claims 1 to 4. The simulation platform includes: a lower fixed base, six electric servo actuators connecting the upper motion platform and the lower fixed base, and the upper motion platform. The hinge points of the six electric servo actuators on the lower fixed base are arranged in a non-circular layout. The longitudinal spacing between the hinge points distributed along the pitch axis of the upper motion platform (3) is significantly greater than the lateral spacing between the hinge points distributed along the roll axis.

6. The rigid-elastic coupling flight simulation platform for large flexible aircraft as described in claim 5, characterized in that, The six-electric servo actuator system comprises six independent electric servo actuators, each of which includes: an AC servo motor, a ball screw drive mechanism connected to the AC servo motor, and a displacement sensor for real-time monitoring of the shortening length of the electric servo actuator.

7. The rigid-elastic coupling flight simulation platform for a large flexible aircraft as described in claim 5, characterized in that, The simulation platform also includes a control system and a main controller connected to the six electric servo actuators; The control system includes: The system consists of a main controller for running the core algorithm, six servo drives corresponding to electric servo actuators, and I / O interface modules for data interaction with the main controller.

8. The rigid-elastic coupling flight simulation platform for a large flexible aircraft as described in claim 7, characterized in that, The main controller adopts a real-time computing architecture based on a digital signal processor (DSP) or a field-programmable gate array (FPGA). The main controller communicates with each servo driver via a real-time industrial bus.

9. The rigid-elastic coupling flight simulation platform for a large flexible aircraft as described in claim 5, characterized in that, The pitch angle range of the upper motion platform is not less than ±25°.

Images