Data-driven flight simulator simulation model correction method and device

Abstract

The application relates to the technical field of data driving, and provides a flight simulator simulation model correction method and device based on data driving, which solves the problems of poor consistency between a simulation model state and real data and low dynamic correction efficiency. The method comprises the following steps: identifying a current compound flight state of a flight target according to flight parameters of the flight target, wherein the flight parameters comprise counter-driving data; determining a deviation degree according to key parameters and secondary key parameters corresponding to the compound flight state; selecting a corresponding target state matching method according to the deviation degree and the compound flight state, and performing state matching according to the target state matching method; and after successful matching, calling a controller of a corresponding channel according to the compound flight state or a current flight stage, and correcting the state of a simulation model corresponding to the flight target through an existing control mode. The application improves the consistency between the simulation model state and the real data and the dynamic correction efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data driving, and in particular to a flight simulator simulation model correction method and device based on data driving. BACKGROUND

[0002] In the real-time simulation and state reproduction process of dynamic systems, there is a clear need for high-precision and high-consistency model correction technology. Especially in application scenarios that require reverse driving of real running data to reproduce a specific running state, the simulation model is required to quickly and automatically track real state changes and adjust in real time when there is a deviation, in order to ensure the effectiveness and reliability of the simulation results.

[0003] At present, a common targeted solution is to use a state monitoring and matching method based on fixed thresholds. According to the pre-defined running stage division, the allowed deviation range of the key running parameters is set; when the difference between the simulation output and the reverse driving data exceeds the set threshold, the pre-set trimming strategy based on the running stage is triggered, and the simulation model is reset or corrected to return to the expected state.

[0004] However, such a solution has certain limitations in responding to complex running states or rapid dynamic transition processes. Since it relies on pre-defined stages and fixed thresholds, when dealing with multi-dimensional coupling, non-steady-state or continuously changing states, the adjustment process may be delayed, and state oscillation may occur after multiple corrections, affecting the continuity and reproduction accuracy of the simulation process, and the adaptability to disturbances in the running environment and model errors also has room for improvement. SUMMARY

[0005] The present application provides a flight simulator simulation model correction method and device based on data driving, an electronic device and a storage medium, to solve the problem of poor consistency between the simulation model state and the real data and low dynamic correction efficiency in the prior art.

[0006] To solve the above technical problems, in a first aspect, the present application provides a flight simulator simulation model correction method based on data driving, comprising:

[0007] Identifying the current complex flight state of the flight target according to the flight parameters of the flight target, the flight parameters including reverse driving data;

[0008] Determining the deviation degree according to the key parameters and the secondary key parameters corresponding to the complex flight state;

[0009] Selecting a corresponding target state matching method according to the deviation degree and the complex flight state, and performing state matching according to the target state matching method;

[0010] After the matching succeeds, a controller of a corresponding channel is invoked according to the compound flight state or a current flight phase to correct a state of a simulation model corresponding to the flight target through an existing control mode.

[0011] Optionally, the method for selecting a corresponding target state matching method according to the deviation degree and the compound flight state comprises:

[0012] According to the deviation degree, it is determined whether the compound flight state is a stable state.

[0013] In a case where the compound flight state is a stable state, a state matching is performed through parameter setting and a first trimming mode corresponding to the stable state based on a flight parameter state for different compound flight states, or in a case where the compound flight state is an unstable state, a state matching is performed according to a second trimming mode corresponding to the unstable state.

[0014] In a second aspect, the present application provides a data-driven flight simulator simulation model correction device, comprising:

[0015] A recognition module is configured to recognize a current compound flight state of a flight target according to flight parameters of the flight target, wherein the flight parameters comprise anti-driving data.

[0016] A determination module is configured to determine a deviation degree according to key parameters and secondary key parameters corresponding to the compound flight state.

[0017] A selection module is configured to select a corresponding target state matching method according to the deviation degree and the compound flight state, and perform a state matching according to the target state matching method.

[0018] An invocation module is configured to invoke a controller of a corresponding channel according to the compound flight state or a current flight phase after the matching succeeds, and correct a state of a simulation model corresponding to the flight target through an existing control mode.

[0019] In a third aspect, the present application provides an electronic device, comprising:

[0020] A memory is configured to store a computer program.

[0021] A processor is configured to implement steps of the data-driven flight simulator simulation model correction method according to the first aspect when the computer program is executed.

[0022] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program, when executed by a processor, enables the steps of the data-driven flight simulator simulation model correction method according to the first aspect to be implemented.

[0023] The technical scheme provided by the present application has the following beneficial effects:

[0024] The present application realizes fine and multi-dimensional identification of the target operating state, provides accurate state basis for subsequent targeted correction, establishes a hierarchical deviation discrimination mechanism, improves the accuracy of state evaluation and the discrimination ability of the influence of different parameters, realizes intelligent adaptation and dynamic selection of the correction strategy, enhances the effectiveness and flexibility of the method under different states, ensures smooth transition and rapid convergence of the correction process, and improves the response speed of the overall system and the coherence of state reproduction.

[0025] Further, according to the determined deviation degree, the present application first judges whether the composite operating state belongs to a stable or unstable state, if it is a stable state, according to the specific state type and real-time parameters, matches through parameter setting and combining the first type of trimming mode specially used for stable state, if it is an unstable state, adopts the second type of trimming mode suitable for non-steady state to implement matching, thereby realizing differentiated and accurate correction under different states. Moreover, this mode can adaptively select the most suitable trimming matching strategy according to different state characteristics such as stability and instability, effectively improve the accuracy and adaptability of state correction, and ensure the response quality and reproduction consistency of the simulation model under various dynamic conditions.

[0026] These aspects or other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0027] In order to more clearly illustrate the technical schemes in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0028] Figure 1 A flowchart of a data-driven flight simulator simulation model correction method provided by an embodiment of the present application;

[0029] Figure 2 A simulation model real-time correction technology module principle diagram of a data-driven flight simulator simulation model correction method provided by an embodiment of the present application;

[0030] Figure 3A target state matching flowchart of a data-driven flight simulator simulation model correction method provided by an embodiment of the present application is shown in the figure.

[0031] Figure 4 A structural schematic diagram of a data-driven flight simulator simulation model correction device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0032] In the field of dynamic system simulation and state reproduction, existing correction methods mostly rely on state monitoring and matching strategies based on fixed thresholds. This method makes judgments and adjustments according to preset operating stages and static parameter tolerances, and although the structure is simple, it has obvious shortcomings in responding to complex, continuous changes or rapid transitions of operating states: the adjustment process is prone to lag, multiple corrections may cause state oscillation, and the adaptability to model errors and external disturbances is limited, affecting the accuracy and continuity of simulation reproduction.

[0033] In view of the above limitations, the present application proposes a data-driven flight simulator simulation model correction method, the core of which is to identify the complex operating state of the target according to real-time parameters, and to adaptively select the matching strategy according to the state characteristics and deviation. Specifically, this method distinguishes between stable and unstable states, and uses targeted trimming and matching methods respectively, to achieve rapid and accurate correction of the simulation state. This scheme effectively overcomes the adaptability and response lag problems of the fixed threshold strategy, improves the tracking accuracy and reproduction consistency of the simulation model under complex operating dynamics, and ensures the smoothness and reliability of the correction process.

[0034] In order to enable personnel in the technical field to better understand the present application scheme, the present application will be further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0035] The core of the present application is to provide a data-driven flight simulator simulation model correction method, and a flowchart of one specific embodiment thereof is shown in the figure. Figure 1 The method comprises:

[0036] Step 101: identifying the current complex flight state of the flight target according to the flight parameters of the flight target, the flight parameters including anti-driving data.

[0037] In step 101, the flight parameters refer to a series of data sets reflecting the motion and state of the flight target obtained in real time, mainly including position, speed, attitude, acceleration, and counter-driving data from external systems. Counter-driving data refers to a sequence of data reflecting the external action or control instruction of the target in the actual running environment obtained by inverse driving, such as manipulation input, environmental disturbance, etc. The composite flight state refers to a complex running mode formed by superimposing multiple basic running states (such as horizontal movement, climbing, descending, turning, etc.) in time and space, and its recognition depends on the coordinated analysis of multi-dimensional flight parameters.

[0038] In the embodiments of the present application, first, the system collects various flight parameters of the flight target in real time, including position, speed, attitude angle, angular velocity, and counter-driving data; then, using these parameters, the current running state is preliminarily classified through a state recognition algorithm (such as a rule-based state machine or a lightweight time sequence pattern recognition method), and basic states such as horizontal flight, climbing, descending, and turning are recognized; then, according to the continuity of these basic states in time and the coupling relationship between parameters, further fusion judgment is made to form the description of the composite flight state, such as "climbing turning" or "accelerating descending"; finally, the system outputs the composite flight state recognized at the current time, providing a state basis for the subsequent steps.

[0039] For example, assuming that a flight target A is executing a task, the system obtains its height, speed, heading angle, and manipulation instruction in the counter-driving data in real time. Through analysis, it is found that the current height is continuously increasing, the speed is basically stable, the heading angle is slowly changing, and there is a clear pitch manipulation input in the counter-driving data, so the system identifies that it is currently in the "climbing" state; at the same time, the continuous change of the heading combined with the rudder manipulation input judges that it also has a "turning" behavior; the system finally fuses the two to determine that the current composite flight state is "climbing turning". This recognition result provides a clear state label for the subsequent steps.

[0040] Step 102: determining the deviation degree according to the key parameters and secondary key parameters corresponding to the composite flight state.

[0041] In step 102, the key parameters refer to the core running indicators that have a decisive influence on the current composite flight state, for example, in the "climbing turning" state, the height change rate, speed, and roll angle can be used as key parameters. The secondary key parameters refer to other parameters that have auxiliary influence on state maintenance or transition, such as angular velocity, acceleration, etc. The deviation degree refers to the comprehensive measure of the difference between the simulation model output state and the real flight target state in each parameter dimension, which is used to quantify the mismatching degree of the current simulation.

[0042] In the embodiments of the present application, the system obtains the set of key parameters and secondary key parameters in the composite flight state from the pre-defined configuration according to the composite flight state identified in step 101; calculates the values of these parameters in the simulation model in real time and compares them with the corresponding parameters of the real flight target; calculates the deviation values of the key parameters and secondary key parameters respectively according to the difference tolerance range set for each type of parameter in advance; finally, the overall deviation degree index is generated by weighting or logical combination of these deviation values, which is used to represent the overall difference level between the current simulation state and the real state.

[0043] For example, based on the above example, after identifying the "climb turn" state, the system calls the key parameters (rate of altitude change, airspeed, roll angle) and secondary key parameters (pitch angle rate, yaw angle rate) in this state. Assuming that the rate of altitude change of the real target is 10 meters per second, the airspeed is 100 meters per second, and the roll angle is 15 degrees; while the corresponding output of the simulation model is 8 meters per second, 105 meters per second and 12 degrees respectively. The system calculates the difference of each parameter, and judges whether the key parameter deviation is according to the preset threshold. If the difference of the rate of altitude change exceeds the threshold, it is determined that the deviation degree is high, and the correction needs to be started.

[0044] Step 103: selecting a corresponding target state matching method according to the deviation degree and the composite flight state, and performing state matching according to the target state matching method.

[0045] In step 103, the target state matching method refers to the specific strategy or algorithm set adopted to adjust the simulation model state to be consistent with the real target state, including parameter resetting, model trimming, control law adjustment, etc. The selection of this method depends on the current deviation degree and the type of composite flight state.

[0046] In the embodiments of the present application, the system receives the deviation degree output by step 102 and the composite flight state result of step 101; if the deviation degree is low, a mild correction strategy is selected, such as fine-tuning the model parameters; if the deviation degree is high, the matching method is further selected according to the composite flight state: for stable state (such as level flight), the parameter trimming method based on balance condition is adopted; for non-stable state (such as maneuvering turn), the dynamic tracking or predictive control method is adopted; finally, the selected matching method is executed to adjust the internal parameters or control input of the simulation model, so that its output state tends to the real target state.

[0047] For example, continuing the previous example, the system determines that the current deviation is high and the compound flight state is "climbing turn" (a non-stable state), and therefore selects the dynamic tracking matching method. According to the height rate of change, airspeed and roll angle target values of the real target, the method generates control instructions for the simulation model, and adjusts the pitch, thrust and roll control surfaces of the simulation model through the internal controller, so that the output height rate of change gradually approaches 10 meters per second, the airspeed approaches 100 meters per second, and the roll angle approaches 15 degrees, thereby completing state matching.

[0048] Step 104: After successful matching, according to the compound flight state or the current flight phase, the controller of the corresponding channel is called to correct the state of the simulation model corresponding to the flight target through the existing control mode.

[0049] In step 104, the channel controller refers to the control module designed independently for different motion dimensions of the simulation model (such as pitch, roll, yaw, thrust, etc.). The control mode refers to the specific adjustment law adopted by the controller, such as proportional-integral-derivative control, fuzzy control, model tracking control, etc., which is used to ensure that the model state stably and accurately tracks the target.

[0050] In the embodiments of the present application, after completing state matching in step 103, the system activates the corresponding channel controller (such as the pitch controller, the roll controller) according to the current compound flight state or flight phase (such as climbing, cruising, landing); each controller continuously calculates the error between the simulation model state and the target state according to its built-in control mode, and generates control instructions; these instructions are applied to the simulation model to dynamically adjust its control surfaces or power output, achieve fine maintenance and continuous correction of the model state, and ensure that its output is consistent with the real flight target state.

[0051] For example, after matching is completed, the system confirms that it is still in the "climbing turn" state, and therefore calls the pitch controller and the roll controller. The pitch controller calculates the elevator deflection angle according to the height rate of change error, and the roll controller calculates the aileron deflection angle according to the roll angle error; these control instructions act on the simulation model in real time, so that the simulation model maintains a height rate of change of 10 meters per second and a roll angle of 15 degrees during the climbing turn, thereby realizing continuous consistency between the simulation state and the real target state.

[0052] The method realizes continuous and stable tracking of the simulation state through real-time identification of the compound operating state of the flight target, accurate judgment of the deviation of the simulation model from the real state, adaptive selection of matching and correction strategies, and finally multi-channel control. The method improves the response speed, state consistency and adaptive ability of the simulation model in complex operating scenarios, effectively guarantees the accuracy and reliability of simulation reproduction.

[0053] To solve how to adaptively select the most suitable matching correction method according to the deviation degree of the simulation state from the real target and the type of the running state, so as to improve the accuracy and efficiency of state reproduction, in some embodiments, step 103: selecting a corresponding target state matching method according to the deviation degree and the compound flight state, performing state matching according to the target state matching method, includes:

[0054] Step 201: judging whether the compound flight state is a stable state according to the deviation degree.

[0055] In step 201, the stable state refers to a running condition in which the running parameters of the flight target change smoothly, the force reaches balance, and there is no obvious acceleration or drastic attitude change, such as uniform horizontal flight or stable climbing. The judgment process depends on the overall deviation degree determined in step 102 and the time-varying characteristics of each key parameter in the current compound flight state.

[0056] In the embodiments of the present application, the system receives the deviation degree value and the current compound flight state information from step 102. First, the system checks whether the deviation degree is lower than the threshold value set for the stable state. If it is lower, it is preliminarily indicated that it may be in a stable state; then, the system further analyzes the change of each key parameter constituting the current compound flight state in a recent period of time, such as calculating the change rate or fluctuation amplitude. If the changes of these parameters are maintained within a small range, it is finally determined that the current compound flight state is a stable state; otherwise, it is determined to be an unstable state. The judgment result will be directly used to guide the selection of the matching method in subsequent step 202.

[0057] Step 202: in the case where the compound flight state is a stable state, for different compound flight states, performing state matching through parameter setting and the first trimming method corresponding to the stable state based on the flight parameter state, or in the case where the compound flight state is an unstable state, performing state matching according to the second trimming method corresponding to the unstable state.

[0058] In step 202, the first trimming method is a set of adjustment strategies specially designed for the stable state, the core of which is to make the model quickly enter and maintain a steady state running by directly setting the core parameters (such as position, attitude, speed) of the simulation model and supplemented by balance calculation. The second trimming method is a strategy designed to deal with non-stable states (such as acceleration, drastic turning), which focuses more on dynamic tracking and compensation of model force or motion trend, rather than pursuing instantaneous static balance.

[0059] In the embodiments of the present application, the system performs branch operations according to the steady state judgment result made in step 201. If it is determined that the steady state, the target parameter values are extracted from the real data according to the current specific compound flight state (such as "flat flight" or "steady climb"), and a first trimming method is adopted: the target parameter values are directly assigned to the simulation model first, and then the balance calculation program for the steady state is started to fine-tune the control variables in the model, so that the model output quickly converges and stabilizes in the target state. If it is determined that the unstable state in step 201, a second trimming method is used: the system no longer seeks static setting of parameters, but calculates the difference between the simulation model and the real target in the trend of motion and force in real time based on the current flight parameter state, and compensates for the difference by dynamically adjusting the control command, so that the state change process of the simulation model is as close as possible to the real target.

[0060] The following is a specific example: in the foregoing embodiment, the flight target A is continuously in the climb turning state, the system determines that the deviation is high and the compound flight state is unstable, and enters the target state matching method selection and execution process. The system first determines whether the compound flight state is a steady state according to the deviation, and since the difference in the rate of change of height exceeds the preset threshold and the parameters change dramatically, it is determined that it is an unstable state. Then the system performs state matching according to the second trimming method corresponding to the unstable state, which is a dynamic tracking compensation based on the current flight parameter state. The system acquires the rate of change of height of the real target A as 10 meters per second, the airspeed as 100 meters per second, and the roll angle as 15 degrees as target values, and continuously collects the corresponding output values of the simulation model as 8 meters per second, 105 meters per second, and 12 degrees. The system calculates the instantaneous errors of each parameter, wherein the height change rate error is 2 meters per second, the airspeed error is 5 meters per second, and the roll angle error is 3 degrees. These error values are input into a dynamic compensation controller, which generates control commands according to the error ratio, and the core calculation relationship is: the control command adjustment amount is equal to the error value multiplied by a proportional coefficient, wherein the proportional coefficient is pre-set according to the importance of the parameter, for example, the height change rate proportional coefficient is 0.5, the airspeed proportional coefficient is 0.2, and the roll angle proportional coefficient is 0.3. Then the height change rate control adjustment amount is 2 meters per second multiplied by 0.5, which is 1 meter per second, the airspeed control adjustment amount is 5 meters per second multiplied by 0.2, which is 1 meter per second, and the roll angle control adjustment amount is 3 degrees multiplied by 0.3, which is 0.9 degrees. The system generates additional thrust control commands and pitch and roll control commands according to this, and applies them to the simulation model, so that the internal parameters of the simulation model are dynamically adjusted. After several control periods, the output of the simulation model gradually approaches 10 meters per second, 100 meters per second, and 15 degrees, thereby completing the matching in the unstable state and laying a foundation for subsequent continuous correction by the channel controller.

[0061] In the embodiments of the present application, the complete step scheme described above ensures that the simulation model can obtain rapid, accurate and stable correction effect under stable or severe running conditions by intelligently judging the stability of the running state and selecting different but highly targeted matching strategies accordingly, thereby enhancing the adaptability and overall correction efficiency of the entire method to different running scenarios.

[0062] To solve the problem of how to adaptively select and execute accurate trimming strategies according to the stability of the composite flight state to efficiently complete the state matching of the simulation model, in some embodiments, step 202: in the case that the composite flight state is a stable state, for different composite flight states, state matching is performed based on the flight parameter state by parameter setting and the first trimming mode corresponding to the stable state, or in the case that the composite flight state is an unstable state, state matching is performed according to the second trimming mode corresponding to the unstable state, including:

[0063] Step 301: in the case that the composite flight state is a stable state, set the position parameter, attitude parameter and speed parameter of the simulation model corresponding to the flight target, and take the position parameter, attitude parameter and speed parameter as the initial conditions for state matching.

[0064] In step 301, the position parameter refers to coordinate data describing the specific point where the flight target is located in the three-dimensional space. The attitude parameter refers to angle data describing the orientation or inclination of the flight target. The speed parameter refers to vector data describing the speed and direction of the flight target. The initial conditions for state matching refer to a set of model initial state values preset for subsequent trimming operations, which are the starting point and basis for matching.

[0065] In the embodiments of the present application, when the system determines that the composite flight state is a stable state, it first extracts the position, attitude and speed data of the current real flight target from the real-time flight parameter state. Then, the system directly sets these data as the corresponding parameter values of the simulation model, so that the simulation model instantly enters an initial running state very close to the real target, thereby preparing conditions for subsequent fine trimming.

[0066] Step 302: based on the flight parameter state and the initial conditions, select a first trimming mode matching the composite flight state from a plurality of preset trimming modes.

[0067] In step 302, the plurality of preset trimming modes refer to a plurality of model balancing adjustment strategy libraries designed in advance for different stable flight states.

[0068] In the embodiments of the present application, the system queries a preset trim strategy mapping table according to the current specific compound flight state type and in combination with the set initial conditions. The table defines the correspondence between different compound flight states and the best trim mode. For example, one trim mode is selected for the steady flight state, and another trim mode is selected for the steady climb state. The system automatically selects the first trim mode that best matches the current state by looking up the table.

[0069] Step 303: Perform a trim operation on the simulation model in the initial condition using the selected first trim mode to complete the state matching.

[0070] In step 303, the trim operation refers to performing calculation and setting on the internal control variables of the simulation model according to the calculation rules and adjustment processes defined by the selected trim mode, so as to eliminate the residual unbalanced force or moment and make the model output state stable at the target value.

[0071] In the embodiments of the present application, the system starts the calculation program corresponding to the selected first trim mode. The program takes the current set initial conditions as input and calculates the control quantities such as the accurate deflection angle of each control surface or the thrust size required by the simulation model to maintain the stable state. After the calculation is completed, the system assigns these control quantities to the simulation model, and the model adjusts its internal state accordingly, reaches a stable equilibrium after a short transition, and thus realizes accurate matching with the real flight target state.

[0072] Step 304: In the case where the compound flight state is an unstable state, based on the flight parameter state, the dynamic change trend of the unstable shaft system is predicted using a long short-term memory network, and based on the dynamic change trend, the final force state under the corresponding shaft system is determined.

[0073] In step 304, the unstable shaft system refers to a specific shaft system whose force is unbalanced or whose motion state change rate exceeds the stable threshold in the current flight state, and the state of this shaft system cannot be matched stably through conventional trim operation. It is not specifically referred to as one of the longitudinal, lateral or normal shaft systems, but refers to any one or more shaft systems that are identified as unstable due to unbalanced force or dramatic changes in motion state under the current specific flight state, and the specific direction depends on the real-time flight conditions and dynamic characteristics. The corresponding shaft system refers to a reference coordinate system determined according to the attitude and motion direction of the aircraft, which is derived from the analysis of the force state and motion attitude of the aircraft in the air. It specifically includes the longitudinal, lateral and normal shaft systems, which correspond to the pitch, roll and yaw motion dimensions of the aircraft, respectively. The dynamic change trend refers to the predicted changes of the motion parameters of these shaft systems in a short period of time in the future. The final force state refers to the target force condition determined for trim calculation after considering the current measured force and the predicted future trend.

[0074] In the embodiments of the present application, when in an unstable state, the system inputs the flight parameter state of a recent time sequence into a pre-trained long short-term memory network model. The network model analyzes the time-dependent relationship of the parameters and outputs a prediction trend of the unstable shaft system movement change in the future short time. The system then combines the prediction trend with the currently measured flight parameter state, and calculates a more accurate and forward-looking final force state on each corresponding shaft system through a data fusion algorithm as the target basis for trimming.

[0075] Step 305: Based on the final force state, a second trimming mode is selected which does not perform trimming operation on the unstable shaft system.

[0076] In the embodiments of the present application, the system analyzes the final force state obtained in step 304 to identify which shaft system is unstable. Then, the system selects a strategy from the preset trimming mode library for unstable state, which explicitly instructs to ignore or not directly perform the regular balance calculation on the identified unstable shaft system, but to adopt a special adjustment mode that bypasses or compensates for the instability, and determines this mode as the second trimming mode to be used currently.

[0077] Step 306: Perform trimming operation on the simulation model using the selected second trimming mode to complete state matching.

[0078] In step 306, the trimming operation here refers to performing the specific calculation and adjustment process defined by the second trimming mode, and the process may be different from the trimming of the stable state, which focuses more on dynamic tracking and compensation.

[0079] In the embodiments of the present application, the system runs the selected second trimming mode. The mode takes the final force state determined in the foregoing as the main input, and calculates a set of control instructions. These instructions may include adjustment instructions for other stable shaft systems, or directly generate compensation signals for dynamically tracking the change trend of the unstable shaft system. The system applies these control instructions to the simulation model, and the model adjusts its behavior according to the instructions, so that the overall output state effectively converges to the real target state without directly forcing the unstable shaft system to balance, thereby completing the matching.

[0080] The following is a specific example: in the preceding embodiment, flight target A continues to be in an unstable compound flight state of climbing turn, and the system has completed preliminary matching based on dynamic compensation, making its height rate of change approach 10 meters per second, airspeed approach 100 meters per second, and roll angle approach 15 degrees. To further achieve precise and adaptive state matching to future changes, the system performs a deep trim process in an unstable state. The system first predicts the dynamic change trend of the unstable axis system based on the current flight parameter state using a long short-term memory network. Specifically, the height rate of change, airspeed, and roll angle data [8.2 meters per second, 99 meters per second, and 14.8 degrees], [8.5 meters per second, 99.5 meters per second, and 14.9 degrees], and [8.8 meters per second, 100 meters per second, and 15 degrees] of a recent time sequence are input into the network, and the network outputs a prediction result that the roll axis system angular velocity will increase from the current 1.5 degrees per second to 2.1 degrees per second in the next 2 seconds, indicating that the roll axis system is an unstable axis system and has an unstable aggravation trend. Based on this dynamic change trend, the system determines the final force state under the corresponding axis system by weighted fusion of the current measured force and the predicted trend, where the roll axis system target moment value is calculated by the formula , where represents the final target moment, with units of Newton meters, represents the weight of the current measured value, which can be 0.6, represents the current measured roll moment, which can be 150 Newton meters, represents the weight of the predicted value, which can be 0.4, represents the expected moment calculated according to the predicted angular velocity change, which can be 160 Newton meters, and the calculation is as follows: Based on this final force state, the system selects a second trim mode that does not directly trim the unstable roll axis system, i.e., a strategy that focuses on indirect compensation by coordinating the motion of other axis systems. Finally, the system trims the simulation model using the selected second trim mode, which calculates that the aileron deflection angle of the simulation model needs to be increased by 2 degrees and the rudder deflection angle needs to be decreased by 1 degree to produce a coordinated yaw moment of 154 Newton meters for compensation, rather than directly adjusting the roll control surface. After applying these instructions to the model, the roll moment output stabilizes around 154 Newton meters within a few control periods, thereby completing high-precision state matching without directly intervening in the unstable axis system, laying a solid foundation for subsequent control.

[0081] In the embodiments of the present application, the above complete step scheme improves the adaptability and matching accuracy of complex and variable flight states by intelligently distinguishing between stable and unstable states and using different strategies of combining static presetting and dynamic prediction for accurate trimming, thereby ensuring that the simulation model can quickly and reliably reproduce the state of the real flight target under various working conditions.

[0082] To solve the problem of how to accurately predict the dynamic changes of the unstable shaft system and determine the accurate stress state, in some embodiments, step 304: based on the flight parameter state, the long short-term memory network is used to predict the dynamic change trend of the unstable shaft system, and based on the dynamic change trend, the final stress state under the corresponding shaft system is determined, comprising:

[0083] Step 401: extracting the aerodynamic force state and the engine thrust state from the flight parameter state.

[0084] In step 401, the aerodynamic force state refers to the comprehensive description of various forces generated when the flight target moves in the air, mainly including lift, drag and side force. The engine thrust state refers to the size and direction information of the propulsive force generated by the engine of the flight target. The extraction process is to separate these two specific data from the overall flight parameter state.

[0085] In the embodiments of the present application, the system first receives the real-time collected flight parameter state, which is a data set containing multiple parameters. Then, the system identifies and separates all parameter data related to air dynamics from this set according to the predetermined data classification rules, and merges them into the aerodynamic force state. At the same time, the system also identifies and separates all parameter data related to engine output, and merges them into the engine thrust state. The two parts of the extracted state data will be used for subsequent sequence construction and analysis.

[0086] Step 402: arranging the force vectors contained in the aerodynamic force state and the engine thrust state in time sequence to form a parameter sequence.

[0087] In step 402, the force vector refers to the information representation of both the size and direction of the force. The parameter sequence refers to an ordered data list formed by arranging the force vector data at multiple consecutive time points in chronological order, used to reflect the change of force over time. The force vector refers to the force vector contained in the two states. The aerodynamic force state contains aerodynamic force vectors such as lift, drag and side force, and the engine thrust state contains thrust vectors and their direction components. These force vectors together constitute the input of the dynamics analysis.

[0088] In the embodiments of the present application, the system obtains the aerodynamic force state and the engine thrust state obtained in step 401, both of which contain force vector information. The system extracts the aerodynamic force vector data and the engine thrust vector data at each sampling time in a recent period of time according to the time stamp of the data. Then, the force vector data is arranged in sequence according to the time sequence to form a continuous time parameter sequence. The sequence describes the change process of the resultant force acting on the flight target in the recent period.

[0089] Step 403: inputting the parameter sequence into a long short-term memory network, extracting a time-dependent relationship from the parameter sequence by the long short-term memory network, and outputting a dynamic change trend of the unstable shaft system in a future preset period of time based on the time-dependent relationship.

[0090] In step 403, the long short-term memory network is a special computing model capable of learning and memorizing the dependent relationship in long time sequence data. The time-dependent relationship specifically refers to the change rule and mutual influence relationship of the force vectors of the aerodynamic force and the engine thrust at continuous time points in the present application. This relationship reflects the time sequence evolution characteristics of the aircraft dynamics state and is the basis for prediction by the long short-term memory network.

[0091] In the embodiments of the present application, the system inputs the parameter sequence constructed in step 402 into a long short-term memory network model pre-trained by a large amount of data. The analysis unit inside the network model processes each data point in the sequence and learns the historical rule and pattern of the force vector change, that is, extracts the time-dependent relationship. Then, the network uses the learned relationship to deduce the force condition in a future preset period of time, and finally outputs a prediction result. The result describes how the unstable shaft system will change in the future, for example, the change trend of the angular velocity or acceleration thereof.

[0092] Step 404: decomposing the force vectors contained in the aerodynamic force state and the engine thrust state to corresponding longitudinal shaft systems, transverse shaft systems and normal shaft systems according to a preset attitude decomposition manner, to obtain a preliminary force state under each shaft system.

[0093] In step 404, the preset attitude decomposition manner is a set of defined mathematical rules for decomposing a force vector in a space to three mutually perpendicular reference directions. The longitudinal shaft system can refer to the direction from the head to the tail of the flight target. The transverse shaft system can refer to the direction from the left side to the right side of the flight target. The normal shaft system can refer to the direction from the top to the bottom of the flight target. The preliminary force state refers to the component force size in the direction of each shaft system after decomposition.

[0094] In the embodiments of the present application, the system again utilizes the force vectors contained in the aerodynamic force state and the engine thrust state extracted in step 401. According to the pre-set decomposition rules, the system decomposes each force vector into the longitudinal, lateral and normal axes by mathematical calculation, and respectively calculates the component force in each axis. After the calculation, the system obtains the current force state of each axis, which is referred to as the preliminary force state of each axis. The specific decomposition process is as follows: the current aerodynamic force state (containing lift, drag, side force, etc.) and the engine thrust state (containing thrust size and direction) are respectively mapped to the longitudinal axis (calculating the front and rear direction forces such as drag and thrust components), the lateral axis (calculating the left and right direction forces such as side force and thrust lateral components), and the normal axis (calculating the up and down direction forces such as lift and thrust vertical components) according to the aircraft body coordinate system through vector projection and decomposition calculation. A specific example is as follows: if the aircraft is in a climbing turning state and the engine thrust is along the body longitudinal axis forward, the engine thrust is decomposed to the longitudinal axis as forward thrust, and the lift in the aerodynamic force is decomposed to the normal axis, and the side force and the lateral component of the thrust generated in the turning are decomposed to the lateral axis.

[0095] Step 405: combining the dynamic change trend, the preliminary force state of each axis is corrected to obtain the final force state of each axis.

[0096] In step 405, correction refers to the process of adjusting and improving the original data according to new information.

[0097] In the embodiments of the present application, the system receives the dynamic change trend prediction information of the unstable axis from step 403 and the preliminary force state of each axis obtained from step 404. The system first identifies which axis is predicted to be an unstable axis. Then, for this specific axis, the system adjusts the preliminary force state value according to the predicted change trend. For example, if the prediction shows that the force of the axis will increase, the corresponding force value will be increased. For other stable axes, the preliminary force state remains basically unchanged. After this adjustment, the system outputs a set of corrected final force states of each axis which are closer to the real situation in the future.

[0098] The following is a specific example: in the embodiment where the preceding flight target A is in a climbing turn unstable state and the system has predicted that the roll axis angular velocity will increase from 1.5 degrees per second to 2.1 degrees per second in the next 2 seconds, the system performs the following process for precise determination of the final force state. The system first extracts the aerodynamic force state from the current flight parameter state, including lift 12000 N, drag 2000 N, and side force 300 N, and the engine thrust state, including thrust 15000 N in the positive direction along the body longitudinal axis. Then the system arranges the force vectors contained in the aerodynamic force state and the engine thrust state in chronological order to form a parameter sequence, which contains the combined force vector data at the last three time points: [thrust 14800 N, lift 11900 N, drag 1950 N, side force 280 N], [thrust 14900 N, lift 11950 N, drag 1980 N, side force 290 N], [thrust 15000 N, lift 12000 N, drag 2000 N, side force 300 N]. The system inputs the parameter sequence into the long short-term memory network, which outputs a confirmation of the dynamic change trend of the unstable axis, i.e. the roll axis, by analyzing the time-dependent relationship in the sequence, and predicts that the angular velocity change trend is consistent with the foregoing. Then the system decomposes the force vectors contained in the current aerodynamic force state and the engine thrust state into the corresponding longitudinal axis, lateral axis, and normal axis according to the preset attitude decomposition method, where the preliminary force state of the longitudinal axis is thrust 15000 N minus drag 2000 N, equal to 13000 N, the preliminary force state of the normal axis is lift 12000 N, and the preliminary force state of the lateral axis is side force 300 N. Finally, the system corrects the preliminary force state under each axis in combination with the dynamic change trend of the increasing roll axis angular velocity, where the normal axis force is strongly coupled with the roll motion and needs to be corrected, and the correction formula is where represents the final force of the normal axis, in N, represents the preliminary force of the normal axis, which is 12000 N, represents the coupling coefficient, which is 0.02 units per degree per second, represents the predicted angular velocity change, which is 2.1-1.5=0.6 degrees per second , and the calculation is as follows: , the longitudinal and lateral axis forces remain unchanged, so the final force state under each axis is longitudinal axis 13000 N, normal axis 12144 N, and lateral axis 300 N, which provides accurate force input for precise positioning of the unstable influence and subsequent trim.

[0099] In the embodiments of the present application, the above complete step scheme forms a set of processes capable of accurately judging the force conditions of each shaft system in a forward-looking manner by finely extracting force information from original data, constructing time series, predicting changes using intelligent networks, performing scientific force decomposition, and finally completing data correction, greatly improving the accuracy and reliability of matching correction in unstable states.

[0100] To solve how to accurately call control channels according to flight states and effectively correct simulation models to achieve high-precision state tracking, in some embodiments, step 104: according to the composite flight state or the current flight phase, a corresponding channel controller is called to correct the state of the simulation model corresponding to the flight target through an existing control mode, including:

[0101] Step 501: according to the composite flight state or the current flight phase, at least one controller to be activated is selected from the pitch channel controller, the roll channel controller, and the yaw channel controller, and the controller is taken as a target controller.

[0102] In step 501, the pitch channel controller is a control module specially responsible for regulating and controlling the pitch attitude of the simulation model, i.e., the up-and-down swinging movement of the nose. The roll channel controller is a control module specially responsible for regulating and controlling the roll attitude of the simulation model, i.e., the left-and-right tilting movement of the fuselage. The yaw channel controller is a control module specially responsible for regulating and controlling the yaw attitude of the simulation model, i.e., the left-and-right turning movement of the nose. The target controller refers to one or more specific controllers that need to be started to perform correction tasks according to the current state.

[0103] In the embodiments of the present application, the system first receives the composite flight state or flight phase information identified at present. Then, the system queries the control channels that need to participate in work under the state or phase according to a preset rule mapping table. For example, the pitch channel controller is needed in the climbing state, and the roll and yaw channel controllers are needed in the turning state. The system activates the corresponding pitch channel controller, roll channel controller, or yaw channel controller from the available controller library according to the query result, and marks it as the target controller that needs to be used at present.

[0104] Step 502: configuring a control mode for the target controller, the control mode including a steady-state control for maintaining state stability or a tracking control for tracking changes in corresponding parameters in anti-driving data.

[0105] In step 502, the control mode refers to the specific adjustment rules and strategies followed by the controller when working. The steady-state control is a control mode whose goal is to eliminate errors and maintain the controlled parameter at a fixed value. The tracking control is another control mode whose goal is to allow the controlled parameter to change in real time to follow the changes of an external input signal.

[0106] In the embodiments of the present application, the system configures appropriate control modes for each activated target controller. The system determines which mode to use again according to the current composite flight state or flight phase. If the state requires to maintain a stable flight attitude, the controller is configured with a steady-state control mode. If the state requires to closely track changes in the anti-driving data, the controller is configured with a tracking control mode. The configuration process is to set the control law and parameters inside the controller, so that it works according to the selected mode.

[0107] Step 503: Based on the anti-driving data, combined with the current state of the simulation model, drive the target controller configured with the control mode to generate corresponding control instructions.

[0108] In step 503, driving refers to providing input data to the controller and starting its calculation process. The control instruction refers to the control signal output by the controller after calculation, which is used to directly operate the control surface of the simulation model, such as rudder deflection angle or throttle size.

[0109] In the embodiments of the present application, the system extracts the parameters related to the target controller in the anti-driving data, such as pitch angle data for the pitch channel. At the same time, the system also obtains the current state feedback of the simulation model in real time. Then, the system inputs these two parts of data into the target controller configured with the control mode. The controller compares the difference between the target value (from the anti-driving data) and the current value (from the model feedback) according to its internal algorithm, and performs calculation, and finally generates specific control instructions, such as how many degrees the elevator deflects.

[0110] Step 504: Based on the control instructions, correct the attitude, trajectory or speed state of the simulation model.

[0111] In step 504, correction refers to actually applying control instructions to the simulation model, so that its internal state changes, thereby reducing the difference between the target state.

[0112] In the embodiments of the present application, the system sends all control instructions generated in step 503 to the simulation model. After receiving these instructions, the simulation model converts them into actual actions on the corresponding control surface, such as adjusting the rudder angle or changing the engine thrust. These actions will change the forces and moments acting on the model, and then cause the attitude, flight path or speed of the model to change accordingly. Through this continuous instruction application and state adjustment, the output of the simulation model is continuously corrected, and finally reaches a high degree of consistency with the real flight target state described by the anti-driving data.

[0113] The following is a specific example: in the embodiment where the preceding flight target A is in a climbing turn state and the height rate of change of the simulation model has approached 10 meters per second and the roll angle has approached 15 degrees after matching, the system performs controller call flow for continuous and accurate correction. The system first selects the pitch channel controller and the roll channel controller to be activated as target controllers from the pitch channel controller, the roll channel controller and the yaw channel controller according to the current composite flight state of climbing turn, because the state needs to control the pitch and roll attitude at the same time. Then the system configures the control mode for the two target controllers, because it needs to continuously track the changes of the height rate of change and the roll angle in the counterforce data, so it configures the tracking control mode for them. Subsequently, the system drives the pitch channel controller configured with the tracking control mode based on the target height rate of change of 10 meters per second and the target roll angle of 15 degrees extracted from the counterforce data, combined with the current output of the simulation model, the height rate of change of 9.8 meters per second and the roll angle of 14.7 meters, the pitch channel controller calculates the height rate of change error of 0.2 meters per second and generates the elevator deflection angle adjustment command of 0.1 degrees according to the internal proportional coefficient of 0.5, and the roll channel controller calculates the roll angle error of 0.3 degrees and generates the aileron deflection angle adjustment command of 0.12 degrees according to the proportional coefficient of 0.4. Finally, the system corrects the simulation model based on these control commands, increases the elevator deflection angle by 0.1 degrees and the aileron deflection angle by 0.12 degrees, which further stabilizes the height rate of change at 10 meters per second and the roll angle at 15 degrees after acting on the model, thereby realizing continuous high-precision correction of the simulation model attitude, ensuring complete consistency with the real target state.

[0114] In the embodiments of the present application, the above complete step scheme forms a set of efficient and reliable state correction closed loop by intelligently selecting control channels, flexibly configuring control modes, accurately generating control commands and effectively acting on the simulation model, which improves the tracking accuracy and response speed of the simulation model to the real target in various complex flight states.

[0115] In order to solve how to accurately identify the current running state of the flight target from multiple source data, in some embodiments, step 101: the current composite flight state of the flight target is identified according to the flight parameters of the flight target, comprising:

[0116] Step 601: according to the EICAS information and the counterforce data in the flight parameters, and according to the existing flight state system, the longitudinal flight state corresponding to the current flight phase is identified, and the lateral flight state is identified according to the yaw angle rate and the roll angle in the flight parameters, combined with the control command in the counterforce data.

[0117] In step 601, the engine indicating and crew alerting system (EICAS) information is a series of data about the state of the aircraft system and power system obtained from the engine indicating and crew alerting system. The existing flight state system is a pre-defined standard flight phase classification framework. The longitudinal flight state is a type describing the state of the aircraft motion in the vertical plane. The yaw rate is the angular velocity of the aircraft head turning left and right. The roll angle is the angle of the aircraft body tilting around the longitudinal axis. The control command is the control signal issued by the pilot or the automatic pilot system. The lateral flight state is a type describing the state of the aircraft lateral motion in the horizontal plane.

[0118] In the embodiment of the present application, the system first extracts the EICAS information such as the engine thrust parameter and the flap position from the flight parameters, and obtains the control input in the counterforce data. The system compares these data with the standards in the existing flight state system, judges which flight phase is currently in, and determines the corresponding longitudinal flight state, for example, climbing or descending. At the same time, the system reads the yaw rate and roll angle data from the flight parameters, and analyzes the lateral motion characteristics in combination with the rudder and aileron control commands in the counterforce data, judges the lateral flight state, for example, straight flight or turning. Through the two parallel analysis processes, the system respectively obtains the longitudinal and lateral state recognition results.

[0119] Step 602: combine the longitudinal flight state and the lateral flight state into a composite flight state.

[0120] In the embodiment of the present application, the system takes the longitudinal flight state and the lateral flight state recognized in step 601 as input, and fuses the two states according to the preset state combination rule. The system queries the state combination mapping table to find the corresponding composite flight state type, for example, combines the longitudinal climbing state and the lateral turning state into a composite flight state of climbing turning. Finally, the system outputs this combined composite flight state result, providing complete state information for subsequent processing.

[0121] Here is a specific example: Suppose a flight target A is performing a mission. The system acquires its flight parameters in real time, including engine indications and crew warning system information showing engine thrust at 85% and flaps retracted. The anti-drive data contains clear throttle push and stick control commands. Simultaneously, the flight parameters show altitude increasing continuously at 5 m / s, airspeed maintained at 100 m / s, and heading angle changing steadily at 0.5 degrees / s. Based on the existing flight state system, the system identifies the current flight phase as a climb phase, corresponding to a climb in the longitudinal flight state. Simultaneously, based on the yaw rate of 0.5 degrees / s and roll angle of 15 degrees in the flight parameters, combined with the rudder push and aileron control commands in the anti-drive data, the system identifies the lateral flight state as a stable turn. Finally, the system combines the longitudinal climb and lateral stable turn into a composite climb-turn flight state. This identification result provides a clear state label for subsequent steps.

[0122] In the embodiments of this application, the above-described complete step scheme achieves a comprehensive and accurate identification of the current operating state of the flight target by analyzing the longitudinal and lateral motion characteristics separately and intelligently integrating the two, providing a reliable state basis for subsequent deviation judgment and correction, and improving the pertinence and effectiveness of the entire correction process.

[0123] To address the challenge of accurately quantifying the state differences between simulation models and real flight targets, in some embodiments, step 102, determining the degree of deviation based on the key and secondary key parameters corresponding to the composite flight state, includes:

[0124] Step 701: Obtain the key parameters and secondary key parameters corresponding to the composite flight state from the preset mapping table of flight states and parameters.

[0125] In step 701, the preset mapping table between flight states and parameters is a predefined database or list that specifies which parameters are the most important key parameters and which are the secondary key parameters for each composite flight state.

[0126] In this embodiment, the system first receives the composite flight state result identified in step 101. Then, the system queries a preset mapping table to find the entry corresponding to the composite flight state. From the found entry, the system reads and obtains the specific names or identifiers of all key and secondary key parameters defined for this state. This parameter information will be used for subsequent deviation calculations and judgments.

[0127] Step 702: Set thresholds for key parameters and secondary key parameters respectively, and determine the degree of deviation according to the thresholds.

[0128] In step 702, the threshold is a threshold value of a permitted difference range set for each parameter.

[0129] In the embodiments of the present application, the system sets a corresponding threshold value for each key parameter and secondary key parameter obtained in step 701. These threshold values are usually set according to the importance and sensitivity of the parameters, and the threshold value of the key parameter is usually more stringent. Then, the system calculates the difference between the simulation model output value and the corresponding parameter value of the real flight target. The system compares these difference values with the respective threshold values. According to the comparison result, the system comprehensively judges the overall deviation degree of the current state, which can be defined as no deviation, slight deviation or deviation, etc. at different levels, to provide a decision basis for whether to start and how to execute the correction subsequently.

[0130] The following is a specific example: in the embodiment of the preceding flight target A in the climb turn state and the system has completed state recognition, the system performs the following process to determine the deviation degree. The system first queries the parameters corresponding to the climb turn state from the pre-set mapping relationship table of flight states and parameters, and obtains the key parameters including the height change rate, airspeed and roll angle, and the secondary key parameters including the pitch angle rate and yaw angle rate. Then the system sets a corresponding threshold value for each parameter, wherein the key parameter threshold value is set to height change rate ± 2 m / s, airspeed ± 4 m / s, roll angle ± 3 degrees, the secondary key parameter threshold value is set to pitch angle rate ± 1 degree / s, and yaw angle rate ± 1.5 degrees / s. The system calculates the difference between the simulation model output value and the real target value, and obtains the height change rate difference of 2 m / s, the airspeed difference of 5 m / s, the roll angle difference of 3 degrees, the pitch angle rate difference of 0.8 degrees / s, and the yaw angle rate difference of 1 degree / s. Comparing these difference values with the set threshold values, wherein the height change rate difference is equal to the threshold value, the airspeed difference exceeds the threshold value, the roll angle difference is equal to the threshold value, the pitch angle rate difference is within the threshold value, and the yaw angle rate difference is within the threshold value. According to the comparison result, the system determines that two items of key parameters reach or exceed the threshold value, and therefore determines that the deviation degree is high, and the correction program needs to be started immediately. This determination result provides a clear decision basis for selecting the dynamic tracking matching method subsequently.

[0131] In the embodiments of the present application, the complete step scheme described above realizes rapid, objective and quantitative evaluation of the state deviation degree through intelligent query of parameter mapping relationship and difference comparison based on threshold value, provides accurate and reliable input for subsequent selection of correct correction strategy, and effectively improves the efficiency and accuracy of the state matching process.

[0132] The present application also provides a flight data counter-driving simulator reproduction system based on real-time correction of a simulation model, which comprises Figure 2As shown, the system includes the following modules: flight phase / state identification module, state deviation degree determination module, target state matching module, and rapid tracking calibration module. The flight phase / state identification module is used to identify and classify the current flight state according to the real-time acquired aircraft parameters. The state deviation degree determination module is used to define key parameters and secondary key parameters according to different flight states, and set threshold values for deviation degree judgment. The target state matching module is used to select different target state matching methods according to the deviation degree and flight state, and perform override activation / inhibition setting. The rapid tracking calibration module is used to quickly calibrate the simulation model state after trimming, so that it is consistent with the real aircraft state.

[0133] First, the flight phase / state identification module divides the flight state into two dimensions of longitudinal and lateral directions. The two-dimensional states can be combined into multiple composite flight states. The current aircraft state is identified and classified according to the real-time acquired aircraft parameters. For different flight states, the state deviation degree determination module defines different key parameters and secondary key parameters, and sets different parameter thresholds for deviation degree judgment. According to the deviation degree of the key parameters and secondary key parameters, it is judged whether to perform target state matching. Then, according to the deviation parameters and flight state, different target state matching methods are selected. After target state matching, it is determined whether to perform rapid tracking control according to whether there is state deviation after counteracting the long-time data, and the matching method can be selected by three channels.

[0134] For flight phase / state identification, the following is explained: According to the engine indication and EICAS information of the crew warning system, and counteracting data, and according to the longitudinal flight state, the current flight phase (such as power-on, taxi-out, take-off 1, take-off 2, take-off 3-ground, take-off 3-climb, cruise, approach, landing, taxi-in, and power-off) is identified. The flight phase division is shown in Table 1:

[0135] Table 1 EICAS information flight phase definition example

[0136]

[0137] For the above EICAS definition, some flight phases are different in phase division for the entire flight mission profile, but the main purpose of this module is to classify the state deviation judgment and target state matching in advance. Therefore, some phases can be considered to be merged and split, and the aircraft state is divided into six states: taxi, take-off, climb, level flight, descent, and landing. There is no matching state requirement for the power-on and power-off phases, so no identification processing is performed. The specific content is shown in Table 2:

[0138] Table 2 EICAS information flight phase definition and flight state matching comparison

[0139]

[0140] In addition, the above stages are basically divided according to the longitudinal and pitch state of flight, and the special state in the horizontal direction during flight also needs to be identified for subsequent selection of different target matching methods, such as turning. For such a state, flight parameters need to be used for identification. The identification of turning and non-turning states can be performed by using the yaw rate and roll angle, assisted by the control command. The identification of the state is independent of the longitudinal flight profile, and can be combined with the longitudinal flight state to form a composite flight state, which can be divided into three working conditions of turning / non-turning, single engine / dual engine, and no or small sideslip / large sideslip, to form a composite flight state (such as ground sliding turning, climbing turning, level flight turning, descending turning, single-engine large sideslip climbing turning, etc.).

[0141] The following is described for state deviation degree discrimination: general judgment criterion: for different flight states, define key parameters (airspeed, altitude, attitude angle, flight phase) and secondary key parameters (pitch angle rate, roll angle rate, yaw angle rate, acceleration). Non-key parameters: other aircraft information. Set threshold values of key parameters and secondary key parameters to judge the deviation degree of the simulation model state from the real aircraft state. The specific content is shown in Table 3:

[0142] Table 3 Design of key parameters and secondary key parameters in each flight phase

[0143]

[0144] The threshold values of the key parameters include consistent flight phase requirements, airspeed ± 3 kts, altitude ± 100 ft, and attitude angle ± 2°. The threshold values of the secondary key parameters include pitch angle rate, roll angle rate, yaw angle rate, and acceleration of 1×10E-2 in international units. The specific content is shown in Table 4:

[0145] Table 4 Setting of threshold values of key parameters

[0146]

[0147] For target state matching, the following is described: for the matching of the target state, override activation and inhibition settings can be performed. The override activation setting application scenario is that when the recognition and deviation degree discrimination in the flight phase appear to be deviated, the user can activate the matching of the target state according to the manual judgment to select a specific matching mode. The override inhibition setting application scenario is that when performing verification matching of a specific subject, if the model itself has a large difference with the real aircraft, and during the subject action process, the super-tolerance condition may appear, at this time, in order to reflect the real state of the model, the matching of the target state should not be performed, and therefore during the subject action process, the matching function of the target state can be inhibited. When the override activation is activated, and the key parameters exceed the threshold value, the target state matching is started; according to the flight phase / state recognition, it can be known whether the aircraft state is a stable state, if it is a stable state, the stable state matching is adopted, and at the same time, according to the flight phase / state recognition, the trimming method in Table 5 is selected, otherwise the non-stable state matching is adopted. Stable state matching: by setting the position, attitude, speed and other parameters of the simulation model, and performing trimming (such as level flight trimming, constant climb angle trimming), the target matching in the stable state is realized. Usually, the matching of the target state is considered in the relative stable state. For the state matching in the stable state, the parameter setting and trimming are realized. The position, attitude, speed and other parameters of the simulation model are set in turn, and then the trimming in this state is performed to realize the stable flight in the target state, and the specific process is as shown in Figure 3 The trimming stage provides multiple trimming methods for selection according to different flight states. In different flight states, different trimming methods are selected according to the specific flight parameter state. For example, level flight trimming, constant climb angle trimming, constant thrust trimming, single engine failure trimming and six degrees of freedom rapid trimming methods. In addition to the above rapid six degrees of freedom trimming, single or multiple degrees of freedom combination trimming can also be performed according to the current aircraft state. Independent trimming methods are provided for the longitudinal, lateral, normal, pitch, roll and yaw six degrees of freedom, and the six degrees of freedom can be combined in different trimming methods to meet the different flight states of the aircraft.

[0148] Table 5 Trimming method

[0149]

[0150] Non-stable state matching: a combination trimming method is adopted, and for the unstable shaft system, trimming is not performed, according to the aerodynamic force and engine thrust state, the attitude is decomposed according to the setting, and the real force state under the corresponding shaft system is reflected (such as climbing with acceleration).

[0151] For fast tracking calibration, the following is described: after trimming, according to the flight phase / state recognition result, the corresponding controller (such as roll, pitch, yaw channel) is called to quickly calibrate the simulation model state. The control mode includes steady state control and tracking control, wherein the steady state control means controlling the tracking parameter to 0, which is suitable for maintaining the lateral stability. The tracking control means that the target is the anti-driving data, and the tracking parameter is controlled to the anti-driving data value in the control process, which is suitable for take-off, climbing and other performance related subjects. The controller framework: the pitch channel (elevator / thrust) includes control parameters: pitch angle, pitch angle velocity, pitch angle acceleration, barometric height (radio altitude), climb rate, normal acceleration, calibrated airspeed, longitudinal acceleration. Control mode: steady state control, tracking control. Roll channel (aileron): control parameters: roll angle, roll angle velocity, roll angle acceleration, yaw angle. Control mode: steady state control, tracking control. Yaw channel (rudder): control parameters: yaw angle, yaw angle velocity, yaw angle acceleration, runway lateral distance. Control mode: steady state control, tracking control. Specific content is shown in Table 6 and Table 7:

[0152] Table 6 Fast calibration tracking controller framework

[0153]

[0154] Table 7 Schematic diagram of relationship between fast calibration tracking parameters and flight states

[0155]

[0156] In the embodiments of the present application, the flight state of the real aircraft can be continuously tracked, ensuring that the flight parameters reproduced by the simulator are consistent with the anti-driving data, thereby improving the accuracy of accident investigation and simulator reproduction. Aircraft state recognition is adopted, which can quickly achieve the fast trimming matching of the target. At the same time, the continuity tracking control after trimming is considered, and the process fast tracking control is adopted to ensure the continuity of the anti-driving simulator process.

[0157] Figure 4 The structure diagram of the flight simulator simulation model correction device provided by the embodiments of the present application is described in the specific implementation part:

[0158] The recognition module 41 is configured to recognize the current compound flight state of the flight target according to flight parameters of the flight target, wherein the flight parameters include anti-driving data.

[0159] The determination module 42 is configured to determine the deviation degree according to the key parameters and the secondary key parameters corresponding to the compound flight state.

[0160] The selection module 43 is configured to select a corresponding target state matching method according to the deviation degree and the composite flight state, and perform state matching according to the target state matching method.

[0161] The calling module 44 is configured to call a controller of a corresponding channel according to the composite flight state or a current flight phase after the matching is successful, and correct a state of a simulation model corresponding to the flight target through an existing control mode.

[0162] The data-driven flight simulator simulation model correction device according to the embodiments of the present application is used to implement the foregoing data-driven flight simulator simulation model correction method, and therefore the specific embodiments in the data-driven flight simulator simulation model correction device can refer to the foregoing embodiment part of the data-driven flight simulator simulation model correction method, and the specific embodiments can refer to the description of the corresponding embodiment part, which will not be described herein again.

[0163] The present application further provides an electronic device, comprising a memory for storing a computer program, and a processor for executing the computer program to implement the steps of the foregoing data-driven flight simulator simulation model correction method.

[0164] The present application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the foregoing data-driven flight simulator simulation model correction method.

[0165] In an exemplary embodiment, the foregoing computer readable storage medium can include but is not limited to a U disk, a read-only memory, a random access memory, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store a computer program.

[0166] The embodiments of the present application further provide a computer program product, wherein the computer program product comprises a computer program, and the computer program is executed by a processor to implement the steps in the foregoing data-driven flight simulator simulation model correction method embodiments.

[0167] The skilled person can further realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the foregoing description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0168] The above describes in detail a data-driven flight simulator simulation model correction method and device, an electronic device and a storage medium provided by the present application. In this paper, specific examples are used to explain the principles and implementation modes of the present application. The above description of the embodiments is only used to help understand the method of the present application and its core idea. It should be pointed out that for ordinary skilled persons in the technical field, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A data-driven based flight simulator simulation model correction method, characterized in that, The method comprises the following steps: According to the flight parameters of the flight target, the current compound flight state of the flight target is identified, and the flight parameters include anti-driving data; According to the key parameters and secondary key parameters corresponding to the compound flight state, the deviation degree is determined; According to the deviation degree and the compound flight state, the corresponding target state matching method is selected, and the state matching is performed according to the target state matching method; After successful matching, according to the compound flight state or the current flight stage, the controller of the corresponding channel is called, and the state of the simulation model corresponding to the flight target is corrected through the existing control mode; According to the deviation degree and the compound flight state, the corresponding target state matching method is selected, and the state matching is performed according to the target state matching method, which comprises: According to the deviation degree, it is judged whether the compound flight state is a stable state; In the case that the compound flight state is a stable state, for different compound flight states, the state matching is performed through parameter setting and the first trimming mode corresponding to the stable state based on the flight parameter state, or in the case that the compound flight state is an unstable state, the state matching is performed according to the second trimming mode corresponding to the unstable state; In the case that the compound flight state is a stable state, the position parameters, attitude parameters and velocity parameters of the simulation model corresponding to the flight target are set, and the position parameters, attitude parameters and velocity parameters are taken as initial conditions for state matching; Based on the flight parameter state and the initial condition, a first trimming mode matched with the compound flight state is selected from a plurality of preset trimming modes; The selected first trimming mode is used to trim the simulation model under the initial condition to complete state matching; In the case that the compound flight state is an unstable state, based on the flight parameter state, the dynamic change trend of the unstable shaft system is predicted by using a long short-term memory network, and based on the dynamic change trend, the final stress state under the corresponding shaft system is determined; Based on the final stress state, a second trimming mode which does not trim the unstable shaft system is selected; The selected second trimming mode is used to trim the simulation model to complete state matching. The method comprises the following steps:

2. The data-driven flight simulator model correction method of claim 1, wherein, The aerodynamic force state and the engine thrust state are extracted from the flight parameter state; The force vectors contained in the aerodynamic force state and the engine thrust state are arranged in time sequence to form a parameter sequence; ​ input the parameter sequence into a long short-term memory network, extract time dependence from the parameter sequence through the long short-term memory network, and output a dynamic change trend of the unstable shaft system in a future preset time period based on the time dependence; decompose force vectors contained in the aerodynamic force state and the engine thrust state into corresponding longitudinal shaft systems, lateral shaft systems, and normal shaft systems according to a preset attitude decomposition mode, to obtain preliminary force states under each shaft system; correct the preliminary force states under each shaft system in combination with the dynamic change trend, to obtain final force states under each shaft system.

3. The data-driven flight simulator model correction method of claim 1, wherein, According to the composite flight state or the current flight phase, the controller of the corresponding channel is called, and the state of the simulation model corresponding to the flight target is corrected through the existing control mode, including: According to the composite flight state or the current flight phase, at least one controller to be activated is selected from the pitch channel controller, the roll channel controller, and the yaw channel controller, and the controller is taken as a target controller; The control mode is configured for the target controller, and the control mode includes a steady-state control for maintaining the state stability or a tracking control for tracking the corresponding parameter change in the anti-driving data; Based on the anti-driving data, the target controller of the configured control mode is driven in combination with the current state of the simulation model, to generate a corresponding control instruction; Based on the control instruction, the attitude, the track, or the speed state of the simulation model is corrected.

4. The data-driven flight simulator model correction method of claim 1, wherein, According to the flight parameters of the flight target, the current composite flight state of the flight target is identified, including: According to the EICAS information and the anti-driving data in the flight parameters, and according to the existing flight state system, the longitudinal flight state corresponding to the current flight phase is identified, and according to the yaw angle rate and the roll angle in the flight parameters, in combination with the control instruction in the anti-driving data, the lateral flight state is identified; The longitudinal flight state and the lateral flight state are combined into a composite flight state.

5. The data-driven flight simulator model correction method of claim 1, wherein, According to the key parameters and the secondary key parameters corresponding to the composite flight state, the deviation degree is determined, including: The key parameters and the secondary key parameters corresponding to the composite flight state are obtained from a preset mapping relationship table of flight states and parameters; The threshold values corresponding to the key parameters and the secondary key parameters are set, and the deviation degree is determined according to the threshold values.

6. A data-driven based flight simulator simulation model correction apparatus, characterized by, including: The identification module is configured to identify the current composite flight state of the flight target according to flight parameters of the flight target, and the flight parameters include anti-driving data; The determination module is configured to determine the deviation degree according to the key parameters and the secondary key parameters corresponding to the composite flight state; The selection module is configured to select a corresponding target state matching method according to the deviation degree and the composite flight state, and perform state matching according to the target state matching method; The calling module is configured to call the controller of the corresponding channel according to the composite flight state or the current flight phase after successful matching, and correct the state of the simulation model corresponding to the flight target through the existing control mode; The selection module is configured to select a corresponding target state matching method according to the deviation degree and the composite flight state, and perform state matching according to the target state matching method. According to the deviation degree, it is judged whether the compound flight state is a stable state or not; In the case where the compound flight state is a stable state, for different compound flight states, state matching is performed based on a flight parameter state, through parameter setting and a first trimming mode corresponding to the stable state, or in the case where the compound flight state is an unstable state, state matching is performed according to a second trimming mode corresponding to the unstable state; The state matching in the case where the compound flight state is a stable state, for different compound flight states, based on a flight parameter state, through parameter setting and a first trimming mode corresponding to the stable state, or in the case where the compound flight state is an unstable state, according to a second trimming mode corresponding to the unstable state, includes: In the case where the compound flight state is a stable state, the position parameter, the attitude parameter and the speed parameter of the simulation model corresponding to the flight target are set, and the position parameter, the attitude parameter and the speed parameter are taken as initial conditions for state matching; Based on the flight parameter state and the initial conditions, a first trimming mode matched with the compound flight state is selected from a plurality of preset trimming modes; The selected first trimming mode is used to perform trimming operation on the simulation model under the initial conditions, so as to complete state matching; In the case where the compound flight state is an unstable state, based on the flight parameter state, a long short-term memory network is used to predict the dynamic change trend of the unstable shaft system, and based on the dynamic change trend, the final stress state under the corresponding shaft system is determined; Based on the final stress state, a second trimming mode is selected, which does not perform trimming operation on the unstable shaft system; The selected second trimming mode is used to perform trimming operation on the simulation model, so as to complete state matching.

Images