An aircraft stall angle of attack active boundary protection control system and method

Abstract

This invention relates to the field of flight control technology and discloses an active boundary protection control system and method for aircraft stall angle of attack, comprising: a multi-source redundant sensing module, a boundary protection decision module, an adaptive execution control module, and a human-machine interaction and collaboration module connected in sequence; the multi-source redundant sensing module is used to collect multi-source parameters related to aircraft flight and output an effective flight state dataset; the boundary protection decision module is used to complete flight state classification and dynamic allocation of human-machine permissions based on the effective flight state dataset and output stall boundary protection control commands of corresponding levels; the adaptive execution control module is used to complete adaptive control of the aircraft control surfaces and power mechanisms, as well as dynamic recovery control after stall angle of attack is exceeded, based on the stall boundary protection control commands and combined with the flight scenario; the human-machine interaction and collaboration module is used to realize two-way human-machine interaction between the pilot and the system, and to complete the input of control commands, system status feedback, and switching of operational control rights.

Description

Technical Field

[0001] This invention relates to the field of flight control technology, specifically to an active boundary protection control system and method for aircraft stall angle of attack. Background Technology

[0002] Stall is one of the core causes of loss of control of aircraft flight attitude and aviation safety accidents. Stall angle of attack boundary protection technology, as a core component of modern fly-by-wire flight control systems, directly determines the handling safety and flight stability of an aircraft throughout its entire flight envelope. With the continuous development of aviation flight control technology, major domestic and foreign aircraft manufacturers have implemented flight envelope protection, maneuverability enhancement, and active stall protection systems in various models of civil airliners and general aviation aircraft. This has formed two main technical solutions: pilot passive warning protection and fixed threshold semi-active intervention. When the aircraft's angle of attack approaches the preset stall threshold, stall intervention can be achieved through automatic stick push, control surface deflection limitation, and engine thrust adjustment. Under normal flight conditions, it has a certain stall protection capability and reduces the risk of stall accidents caused by human error to a certain extent.

[0003] However, existing stall angle-of-attack boundary protection technologies have not constructed a closed-loop flight control protection architecture with multi-source redundant perception, boundary protection decision-making, adaptive execution control, and human-machine interaction collaboration, all of which are layered and decoupled. This results in the existing systems being unable to resolve the core contradiction between the reliability of stall active protection, the ability to adapt to all scenarios, and the absolute guarantee of the pilot's operational control. They are unable to perform adaptive active protection and stable attitude recovery in complex flight scenarios such as turbulence, crosswinds, and single-engine failure after the angle of attack exceeds the stall boundary. They are also unable to guarantee the pilot's control priority and control under all flight conditions, which can easily lead to aircraft safety risks such as human-machine control conflicts, system mis-triggers, and protection failures. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes an active boundary protection control system and method for aircraft stall angle of attack, which solves the aforementioned technical problems.

[0005] Firstly, an active boundary protection control system for aircraft stall angle of attack is provided, based on a hierarchical decoupled closed-loop flight control protection architecture, including: a multi-source redundant sensing module, a boundary protection decision module, an adaptive execution control module, and a human-machine interaction and collaboration module that are connected in sequence via communication. The multi-source redundant sensing module is used to collect multi-source parameters related to aircraft flight and output a valid flight status dataset. The boundary protection decision module is used to complete the flight status classification and dynamic allocation of human and machine permissions based on the effective flight status dataset, and output stall boundary protection control commands of the corresponding level. The adaptive execution control module is used to perform adaptive control of the aircraft control surfaces and power mechanism based on stall boundary protection control commands and in combination with flight scenarios, as well as dynamic recovery control after stall angle of attack is exceeded. The human-machine interaction and collaboration module is used to realize two-way human-machine interaction between the pilot and the system, and to complete the input of control commands, system status feedback and switching of operation control.

[0006] Furthermore, the multi-source redundant sensing module includes a three-level redundant angle of attack acquisition unit, a multi-source auxiliary parameter acquisition unit, a sensor health status determination unit, and an extended Kalman filter fusion unit that are interconnected. The three-level redundant angle of attack acquisition unit and the multi-source auxiliary parameter acquisition unit are used to complete the multi-channel redundant acquisition of aircraft angle of attack, flight status and atmospheric environment parameters, and output the raw acquisition dataset. The sensor health status determination unit is used to complete the real-time identification and classification of the health / fault status of each sensor and output sensor health status data. The extended Kalman filter fusion unit is used to preprocess the original acquired dataset, remove outliers, and perform dynamic weight fusion to output the optimal effective flight state dataset.

[0007] Furthermore, the boundary protection decision module includes a dynamic stall threshold calculation unit, a flight status classification determination unit, and a dynamic access control unit that are interconnected. The dynamic stall threshold calculation unit is used to calculate the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions based on the effective flight state dataset. The flight status classification and determination unit is used to complete the real-time determination of three levels of flight status: normal flight, early warning intervention, and emergency control, based on the correspondence between real-time angle of attack and dynamic threshold. The dynamic permission management unit is used to dynamically allocate human-machine operation permissions based on the flight status classification results and output stall boundary protection control commands of the corresponding level.

[0008] Furthermore, the adaptive execution control module includes a flight scene recognition unit, a scene-specific adaptive control law calculation unit, and a dynamic recovery control unit that are interconnected. The flight scene recognition unit is communicatively connected to the boundary protection decision module and is used to complete the real-time recognition of turbulence / crosswind, single engine failure / asymmetric thrust, and propeller stall flight scenes based on the effective flight state dataset. The scenario-specific adaptive control law solving unit is used to solve and output adapted control surface deflection control commands and power mechanism thrust control commands based on the identified flight scenario and stall boundary protection control commands. The dynamic recovery control unit is used to dynamically adjust the control intensity based on the real-time flight status of the aircraft during the stall recovery phase, and output smooth stall recovery control commands.

[0009] Furthermore, the human-machine interaction collaboration module includes an operation command recognition unit, a multimodal status feedback unit, an emergency operation triggering unit, and a control power flexible handover unit that are interconnected. The control command recognition unit is used to collect and recognize the pilot's control input commands and transmit them to the boundary protection decision module; The multimodal state feedback unit is used to output multi-dimensional system state prompts and control suggestions to the pilot based on the system operation state and flight state classification results. The emergency control triggering unit is communicatively connected to the boundary protection decision module and is used to identify the pilot's emergency control operations, triggering system intervention to suspend control and immediately transfer control to the pilot. The flexible handover unit is used to complete the flexible transfer of control from the system to the pilot based on the aircraft's restored stable flight status and the pilot's confirmed operation.

[0010] Secondly, a method for active boundary protection control of aircraft stall angle of attack is provided, based on an active boundary protection control system for aircraft stall angle of attack described in any of the preceding claims, comprising: In response to the real-time acquisition trigger of the aircraft's flight status, redundant acquisition and fusion processing of the aircraft's multi-source flight parameters are performed to extract an effective flight status dataset. A dynamic stall boundary calculation model is constructed. The dynamic stall boundary calculation model is used to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions. By acquiring real-time aerodynamic parameters and attitude change data of the aircraft, and combining them with the extracted effective flight status dataset, the current flight scenario type and flight safety status classification can be obtained. Based on the dynamic stall angle of attack threshold, flight scenario type and flight safety status decomposition, an appropriate boundary protection control command is generated to adaptively control the aircraft control surfaces and power mechanism, and to complete the dynamic recovery control after the stall angle of attack is exceeded. The system provides feedback on its operational status and operational suggestions to the pilot through a human-machine interaction link. In response to the pilot's control commands, the system flexibly switches control between the pilot and the pilot and executes corresponding flight maneuvers.

[0011] Furthermore, in response to the real-time acquisition trigger of aircraft flight status, redundant acquisition and fusion processing of multi-source flight parameters of the aircraft are performed to extract an effective flight status dataset, including: The aircraft angle of attack parameters are synchronously acquired and cross-validated by a three-level redundant angle of attack acquisition unit to obtain three channels of redundant raw angle of attack data. The sensor health status determination unit synchronously collects the aircraft's flight status and atmospheric environmental parameters to obtain auxiliary raw data such as airspeed, altitude, temperature, air pressure, load and flap position. By using an extended Kalman filter fusion module to preprocess, remove outliers, and dynamically weight the three redundant angle-of-attack raw data and auxiliary raw data, an effective flight state dataset is obtained.

[0012] Furthermore, a dynamic stall boundary calculation model is constructed. This model is used to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions, including: A database of stall boundary references is constructed by predefining the basic stall angle of attack, standard atmospheric parameters, maximum rated load and aerodynamic configuration reference parameters of the aircraft model. The effective flight status dataset is standardized to extract atmospheric environmental parameters, flight load parameters, and flap position parameters under the current operating conditions. The atmospheric density under the current operating conditions is calculated based on atmospheric environmental parameters, and the environmental correction factor is obtained by combining it with standard atmospheric parameters. Based on flight load parameters and flap position parameters, combined with maximum rated load and aerodynamic configuration reference parameters, the flight state correction coefficient is calculated. By combining the aircraft's basic stall angle of attack, environmental correction coefficient, and flight state correction coefficient, the dynamic stall angle of attack threshold under the current operating condition is calculated. The corresponding buffer threshold is then obtained by synchronously matching the preset buffer threshold tuning rules.

[0013] Furthermore, based on the dynamic stall angle of attack threshold, flight scenario type, and flight safety state decomposition, appropriate boundary protection control commands are generated to adaptively control the aircraft control surfaces and power mechanism, completing dynamic recovery control after stall angle of attack exceedance, including: Based on the flight safety status classification, match the corresponding human-machine access control rules and basic control strategies; Based on the human-machine permission allocation rules and basic control strategies, for the identified flight scenario type, the appropriate scenario-specific control law is invoked to calculate and generate corresponding aileron, rudder, and elevator deflection control commands and throttle thrust control commands. During the stall recovery phase, based on real-time airspeed and load conditions, the aileron, rudder, and elevator deflection control commands and throttle thrust control commands are optimized, and the thrust increase and elevator push rod force are dynamically adjusted to generate smooth recovery control commands. The final control commands are sent to the actuators, which drive the aircraft control surfaces and power mechanisms to complete the corresponding actions, thereby achieving adaptive protection of the stall boundary in all flight states and dynamic recovery control after the stall angle of attack is exceeded.

[0014] The invention employing the above technical solution has the following advantages: 1. This invention solves the problem of human-machine permission conflict through three-level permission allocation and flexible handover logic, fully guaranteeing the pilot's control. At the same time, the system can quickly intervene in extreme scenario protection to reduce the risk of human error. The emergency control mechanism ensures the pilot's control in special circumstances, thereby minimizing the occurrence of similar accidents.

[0015] 2. This invention effectively avoids the risks of nitrogen source failure and extreme failure scenarios through a three-level sensor redundancy and multi-source data fusion architecture; the graded fault tolerance strategy ensures that the system can still operate in degraded mode when the sensor fails, thereby improving flight reliability.

[0016] 3. The adaptive control law of this invention covers complex scenarios such as turbulence, crosswind, single-engine failure, and spiral stall. The dynamic threshold adapts to changes in environment and state, solving the problems of incomplete scenario coverage and fixed thresholds in existing technologies. Attached Figure Description

[0017] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly described below. In all the drawings, the elements or parts are not necessarily drawn to scale.

[0018] Figure 1 This is a flowchart illustrating the working logic of the protection control law in the active boundary protection control system and method for aircraft stall angle of attack of the present invention. Figure 2 This is a diagram of the dynamic stall threshold calculation model in the active boundary protection control system and method for aircraft stall angle of attack of the present invention; Figure 3 This is a schematic diagram of the scenario-based adaptive control logic in the active boundary protection control system and method for aircraft stall angle of attack of the present invention. Figure 4 This is a dynamic recovery control logic diagram in an active boundary protection control system and method for aircraft stall angle of attack according to the present invention. Figure 5 This is a structural diagram of an active boundary protection control system for aircraft stall angle of attack according to the present invention; Figure 6 This is a schematic diagram illustrating the algorithm implementation principle of an active boundary protection control system and method for aircraft stall angle of attack in this invention. Figure 7 This is a flowchart of an active boundary protection control method for aircraft stall angle of attack according to the present invention. Detailed Implementation

[0019] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0020] like Figures 1 to 7 As shown, the present invention provides an active boundary protection control system for aircraft stall angle of attack, based on a hierarchical decoupled closed-loop flight control protection architecture, comprising: a multi-source redundant sensing module, a boundary protection decision module, an adaptive execution control module, and a human-machine interaction and collaboration module that are connected in sequence via communication. The multi-source redundant sensing module is used to collect multi-source parameters related to aircraft flight and output a valid flight status dataset; The boundary protection decision module is used to classify flight status and dynamically allocate human and machine permissions based on the effective flight status dataset, and output stall boundary protection control commands of the corresponding level. The adaptive execution control module is used to perform adaptive control of the aircraft control surfaces and power mechanism based on stall boundary protection control commands and in combination with flight scenarios, as well as dynamic recovery control after stall angle of attack is exceeded. The human-machine interaction and collaboration module is used to enable two-way human-machine interaction between the pilot and the system, and to complete the input of control commands, system status feedback and switching of operation control.

[0021] In this embodiment, the multi-source redundant sensing module includes a three-level redundant angle of attack acquisition unit, a multi-source auxiliary parameter acquisition unit, a sensor health status determination unit, and an extended Kalman filter fusion unit that are interconnected. The three-level redundant angle of attack acquisition unit and the multi-source auxiliary parameter acquisition unit are used to complete the multi-channel redundant acquisition of aircraft angle of attack, flight status and atmospheric environment parameters, and output the raw acquisition dataset. The sensor health status determination unit is used to complete the real-time identification and classification of the health / fault status of each sensor and output sensor health status data. The extended Kalman filter fusion unit is used to preprocess the original acquired dataset, remove outliers, and perform dynamic weight fusion to output the optimal effective flight state dataset.

[0022] Specifically, a three-level redundant angle-of-attack acquisition unit: This unit adopts a three-level sensor redundancy configuration of "primary / backup / redundancy", and the specific implementation is as follows: Main sensor: Equipped with one set of high-precision angle of attack sensor with a measurement error ≤ ±0.1°, serving as the core data input source for angle of attack; Backup sensor: Configure one set of high-precision angle of attack sensor of the same specifications as the main sensor, which collects data synchronously with the main sensor and performs real-time cross-comparison of data; Redundant sensors: Configure one set of low-cost angle of attack sensor + indirect calculation module. The indirect calculation module can derive the angle of attack reference value based on airspeed, lift coefficient and load parameters, serving as a backup data source for extreme failure scenarios.

[0023] Three angle-of-attack sensors synchronously acquire real-time angle-of-attack data of the aircraft, with a sampling frequency of no less than 100Hz. All data are connected to the extended Kalman filter fusion unit via CAN bus / ARINC429 protocol.

[0024] Multi-source auxiliary parameter acquisition unit: This unit is equipped with multiple redundant auxiliary sensors, specifically including: 2 sets of airspeed sensors, 2 sets of altitude sensors, 2 sets of atmospheric data sensors (for measuring atmospheric temperature and atmospheric pressure), 1 set of load sensor, and 1 flap position sensor. All auxiliary sensors are connected to the extended Kalman filter fusion unit via an aviation-grade bus to achieve redundant acquisition of multi-dimensional parameters such as flight airspeed, altitude, atmospheric temperature, atmospheric pressure, load, and flap position. The sampling frequency is synchronized with that of the angle-of-attack sensor.

[0025] Sensor health status determination unit: This unit is used to perform real-time health / fault assessments on the main, backup, and redundant angle-of-attack sensors and each auxiliary sensor. The health status obtained will be directly used as the core input parameter of the extended Kalman filter fusion unit to achieve closed-loop linkage of health status, noise adjustment, status prediction, status correction, and health status.

[0026] The specific implementation method for health determination of angle-of-attack sensors is as follows: No. The real-time measurement value of the road angle of attack sensor is The fused reference angle of attack after state correction by the extended Kalman filter at the previous moment was: The standard deviation of multiple consecutive frames of data is Data update delay is The preset model compatibility deviation threshold is: (Engineering Values) The nominal standard deviation of the sensor is The maximum allowed update delay is (Engineering Values) ).

[0027] The sensor health determination criteria are: Data deviation is normal: ; Data stability is normal: ; Communication and response are normal: And there were no more than 3 consecutive frames of packet loss.

[0028] A sensor is considered healthy if and only if all of the above conditions are met simultaneously; otherwise, it is considered faulty. Sensor Health Status The quantization expression is: in, Indicates sensor health. This indicates a sensor malfunction.

[0029] For auxiliary sensors, health determination uses the same logic for deviation verification, stability verification, and communication status verification. When a single auxiliary sensor fails, it automatically switches to data from a backup sensor of the same type, without affecting the operation of the core protection logic.

[0030] Extended Kalman filter fusion unit: This unit employs an improved Extended Kalman Filter (EKF) algorithm, using sensor health status and data jitter as core inputs. It achieves a fully closed-loop operation encompassing adaptive noise covariance adjustment, state / covariance prediction, Kalman gain calculation, state correction, and weight fusion / health status feedback. This differs from the conventional EKF algorithm with fixed noise covariance. The specific implementation steps are as follows: Data preprocessing: Outlier removal and missing value completion are performed on the raw data collected by each sensor: abnormal data caused by sensor jitter, transient interference or communication anomalies are removed based on the 3σ criterion; missing data is completed by linear interpolation to ensure the continuity and validity of the input data.

[0031] Core parameter definitions: Define the system state vector ,in For angle of attack, For airspeed, For height, For sideslip angle, For roll angle, Define the yaw angle; define the sensor health set. ,in, , , These are the health status values ​​of the primary, backup, and redundant angle-of-attack sensors, respectively; the pre-processed values ​​are defined as follows: Road sensor data The standard deviation of the criterion is .

[0032] Adaptive noise covariance calculation process Noise covariance The adaptive calculation formula is: in, The initial fixed process noise covariance matrix; This represents the total number of angle-of-attack sensors, fixed at 3. The more faulty sensors there are, the more health sensors there are. The smaller, The larger the value, the better it is to accommodate model uncertainty. Maximum not exceeding Observation noise covariance matrix In the middle, the first Observation noise covariance corresponding to road sensor The adaptive calculation formula is: in, For the first The initial fixed observation noise covariance of the road sensor; For the first The nominal standard deviation of the road sensor; the greater the data jitter, The larger, The larger the size, the more adaptively the sensors adjust. Composition of observation noise covariance matrix It is directly used as the core input for solving the Kalman gain.

[0033] Optimized EKF iterative calculation: Prediction step: State prediction: Covariance prediction: in, The system's nonlinear state equations; This is the system control input at time k-1; The Jacobian matrix corresponding to the state equation is calculated using piecewise linearization. The above adaptively adjusted process noise covariance matrix is ​​not a fixed value.

[0034] Update steps: Kalman gain calculation: in, The Jacobian matrix corresponding to the observation equation; This is the observation noise covariance matrix after adaptive adjustment for each of the above sensors; it is not a fixed matrix.

[0035] Status correction: in, for Sensor fusion observations at any given time; The system's nonlinear observation equation is used; after correction, the optimal estimate of the angle of attack is obtained, which is output to subsequent modules and fed back to the sensor health status determination unit as a fusion reference value for the next time step health determination. Covariance update: in, Given the identity matrix, this covariance update formula is an improved format that enhances filtering stability and, through... Ensure seamless integration of update steps and noise adaptation across the entire process.

[0036] Dynamic weighted fusion and data voting: Angle of attack data are fused using a dynamic weighted allocation rule: Let the base weights of the primary, backup, and redundant angle of attack sensors be respectively... , , The weights of faulty sensors are set to 0, and the weights of healthy sensors are renormalized according to their original weight ratios to obtain dynamic weights. : in, A collection of health sensors.

[0037] Final angle of attack The calculation formula is: This fused angle of attack serves as both an effective angle of attack output for the flight control system and a fused reference value for determining the health status of the sensors at the next moment, while also providing data for adaptive adjustment of noise covariance.

[0038] Simultaneously, a three-out-of-two data voting mechanism is set up: if the difference between any two data streams from the main / backup / redundant sensors is >0.3°, data voting is performed in conjunction with the fusion reference value, fault data with a deviation >0.5° from the fusion reference value is eliminated, and the remaining valid data is selected as the algorithm input.

[0039] Auxiliary parameters such as airspeed and altitude are obtained by averaging data collected from similar health sensors to form a stable set of auxiliary state parameters. This set, together with the fused angle of attack, forms the optimal flight state dataset, which is then output to the boundary protection decision module.

[0040] Tiered fault tolerance strategy: This module is equipped with a tiered fault tolerance strategy, which performs differentiated processing for different sensor failure scenarios. The specific implementation is as follows: Single sensor failure: When the difference between the main and backup angle of attack sensor data is ≤0.3°, the average value is taken; when the difference is >0.3°, abnormal data is eliminated by combining the reference value of the redundant sensor. If the difference between the main sensor and the backup + redundant data is >0.5°, the main sensor is determined to be faulty, and the backup sensor will automatically take over. When the auxiliary sensor fails, the data will automatically switch to the backup sensor of the same type without affecting the core protection logic.

[0041] Dual sensor failure (simultaneous failure of primary and backup angle of attack sensors): Activate redundant sensor + Kalman filter prediction mechanism, deduce the angle of attack change trend based on historical 0.5s effective data, and generate temporary angle of attack data; at the same time, send a multi-modal alarm "sensor failure, system degraded operation", the control law automatically switches to conservative mode, the buffer threshold is expanded to 2°, and the intervention intensity is reduced.

[0042] Multi-sensor failure (primary + backup + redundant angle of attack sensors all fail): The indirect parameter derivation model is activated, and the estimated angle of attack is calculated in reverse based on airspeed, thrust, and elevator deflection angle. The system switches to emergency protection mode, retaining only the core recovery logic (moderate stick push + stable throttle), and continuously prompts the pilot "sensor failure, return to base recommended".

[0043] In this embodiment, the boundary protection decision module includes a dynamic stall threshold calculation unit, a flight status classification determination unit, and a dynamic access control unit that are interconnected. The dynamic stall threshold calculation unit is used to calculate the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions based on the effective flight state dataset; The flight status classification and determination unit is used to make real-time determinations of three flight statuses: normal flight, early warning intervention, and emergency control, based on the correspondence between real-time angle of attack and dynamic threshold. The dynamic access control unit is used to dynamically allocate human-machine control permissions based on the flight status classification results and output stall boundary protection control commands of the corresponding level.

[0044] Specifically, the dynamic stall threshold calculation unit: This unit constructs a dynamic stall boundary calculation model, integrating atmospheric environment and flight state parameters to correct the stall boundary in real time. The specific implementation steps are as follows: Input parameters: Input parameters include: the fused angle of attack output by the multi-source redundant sensing module. ,airspeed ,high Atmospheric temperature Atmospheric pressure Load capacity flap position .

[0045] Correction factor calculation: Environmental correction factor Based on height ,temperature air pressure Calculate the current atmospheric density Combined with standard atmospheric density (Value taken as 1.225 kg / m) 3 Calculate the environmental correction factor to correct for the influence of the atmospheric environment on the stall boundary. The calculation formula is as follows: State correction coefficient The formula for adapting to changes in load and flap status is as follows: in, This refers to the maximum rated load of the model. This is the normalized value of the flap deflection angle, with a range of values ​​of [value missing]. .

[0046] Dynamic threshold output: Dynamic stall angle of attack threshold The calculation formula is: in, The basic stall angle of attack for the aircraft is pre-calibrated based on the aircraft's aerodynamic characteristics.

[0047] Synchronously set buffer threshold , The range of values ​​is It can be adjusted online according to the characteristics of the model; the final output warning and protection trigger threshold range is: .

[0048] Simultaneously, the buffer threshold adaptation rules are optimized for high and low altitude operating conditions: In high-altitude, low-density atmospheric environments, the buffer threshold is adjusted to The condition for recovery is the angle of attack. And duration ; In low-altitude atmospheric environments, the buffer threshold is adjusted to 1.15, the condition for recovery is angle of attack. And duration .

[0049] Flight status classification determination unit: This unit is based on real-time fused angle of attack. Based on the correspondence with the dynamic stall threshold, the flight state is divided into three levels, and the specific judgment rules are as follows: Level 1 Normal Flight Status: The determination criteria are as follows The aircraft is in a safe flight zone and there is no risk of stalling; Level 2 early warning intervention status: The judgment criteria are as follows: The aircraft's angle of attack is approaching the stall boundary, posing a potential stall risk; Level 3 Emergency Control Status: The determination condition is as follows The aircraft's angle of attack exceeded the stall limit and it has entered the stall risk zone.

[0050] Dynamic permission management unit: Based on the flight status classification results, this unit designs a three-level dynamic allocation rule for permissions: "Normal / Early Warning / Emergency," along with a flexible control handover logic. The specific implementation is as follows: Three-level permission allocation rules: Level 1 Normal Flight Status Permission Rules: The pilot has full control. The system only monitors flight parameters in the background in real time, without intervening in any operation or outputting any control commands. Level 2 warning intervention status permission rules: The system issues multimodal warning prompts through the human-machine interaction collaboration module and provides control suggestions (such as "pay attention to angle of attack, push the stick appropriately"). The pilot retains the highest control priority, and the system does not forcibly intervene or output any forced control commands. Level 3 Emergency Control Status Authority Rules: The system prioritizes intervention to implement protective control, outputting stall boundary protection control commands to the adaptive execution control module; simultaneously, an emergency over-control mechanism is set up, and when the pilot applies a control force exceeding a preset threshold (engineering value ≥50N), system intervention is immediately suspended, and control is completely transferred to the pilot; after control, the system continuously monitors the flight status, and if the flight attitude approaches the stall threshold again, the system can re-intervene for protection only after the pilot presses the confirmation button on the control stick.

[0051] Logic for flexible return of control: Once the aircraft's angle of attack returns to a safe range and the basic conditions for handing over control are met, a flexible handover procedure will be executed, as follows: Basic conditions for return: The following must be met simultaneously: ① The angle of attack returns to the safe range, meets the recovery judgment condition at the corresponding altitude, and the duration is ≥0.5s; ②Airspeed fluctuation ≤5%; ③ Altitude fluctuation ≤10m, flight status is completely stable.

[0052] The handover process is as follows: The system issues a control handover prompt via tactile vibration (low-frequency continuous) on the joystick of the human-machine interface module. The pilot can confirm the handover in two ways: ① Press the confirmation button on the joystick; ②Continuous normal operation ≥1s.

[0053] Once the pilot confirms, the system immediately relinquishes full control, ceases all background intervention, and switches to Level 1 normal flight status permission rules. If the pilot does not respond, the system remains in auxiliary monitoring status, repeating the prompt every 3 seconds until the pilot confirms or the flight status changes (angle of attack approaches the stall threshold again).

[0054] In this embodiment, the adaptive execution control module includes a flight scene recognition unit, a scene-specific adaptive control law calculation unit, and a dynamic recovery control unit that are interconnected. The flight scene recognition unit communicates with the boundary protection decision module to complete the real-time recognition of turbulence / crosswind, single engine failure / asymmetric thrust, and propeller stall flight scenes based on the effective flight state dataset; The scenario-specific adaptive control law solving unit is used to solve and output adapted control surface deflection control commands and power mechanism thrust control commands based on the identified flight scenario and stall boundary protection control commands. The dynamic recovery control unit is used to dynamically adjust the control intensity based on the real-time flight status of the aircraft during the stall recovery phase, and output smooth stall recovery control commands.

[0055] Specifically, the flight scene recognition unit: This unit extracts flight state features in real time based on the effective flight state dataset output by the multi-source redundant sensing module, and completes the identification of typical complex flight scenarios. The specific identification rules are as follows: Turbulent / crosswind scene recognition: Real-time calculation of sideslip angle Roll angle Yaw angle The rate of change, when , , If any two of the three conditions are met, the scene is determined to be a turbulent / crosswind scenario; Single-engine failure / asymmetric thrust scenario identification: Real-time monitoring of thrust output of each engine; when the thrust difference between engines... ( When the aircraft model is adapted to a threshold (usually 15% of the rated thrust), it is considered a single-engine failure / asymmetric thrust scenario. Propeller stall scene recognition: For propeller-powered aircraft, when the angle of attack exceeds the stall threshold and the propeller thrust shows abnormal decline, it is determined to be a propeller stall scene.

[0056] Scenario-specific adaptive control law solution unit: This unit designs differentiated adaptive control logic for different flight scenarios, calculates the corresponding control surface and power control commands, and implements them as follows: The control law for turbulent / crosswind scenarios outputs adaptive aerodynamic compensation commands to counteract attitude instability caused by aerodynamic disturbances. The specific control commands are as follows: Aileron deflection command: in, , For adaptive coefficients, The value range is 0.8-1.2. The value ranges from 0.3 to 0.5, and is dynamically adjusted based on real-time airspeed. Rudder deflection command: in, , For adaptive coefficients, The value range is 0.6-0.9. The value ranges from 0.2 to 0.4, and is dynamically adjusted based on real-time airspeed. Rudder deflection command: in, , For adaptive coefficients, Range of values , Range of values Based on real-time airspeed dynamic adjustment; Elevator trim commands: To counteract angle of attack fluctuations; Automatic throttle control command: based on the deviation between target airspeed and actual airspeed. Adjust the thrust. ,in Range of values .

[0057] Control law for single-engine failure / asymmetric thrust scenarios: For single-engine failure / asymmetric thrust scenarios, the compensation thrust and aerodynamic trim commands are calculated in real time to counteract the roll and yaw trends caused by asymmetric thrust. The specific control logic is as follows: Thrust compensation command: Calculate the compensated thrust. Increase the thrust of the undamaged engines, up to a maximum of the rated thrust, to supplement the thrust required for flight; Aerodynamic trim commands: coordinate aileron and rudder deflection, output yaw and roll suppression commands, counteract attitude disturbances caused by asymmetric thrust, and ensure the lateral and directional stability of the aircraft.

[0058] Control law for propeller aircraft stall scenarios: For propeller-driven aircraft stall scenarios, control logic is designed to adapt to the propeller's dynamic characteristics, as follows: Propeller pitch control: Automatically reduces propeller pitch after the angle of attack exceeds the stall threshold. Increase propeller thrust; Engine speed control: Dynamically adjusts engine speed; the control formula is as follows: ,in, The engine's rated speed. The value range is 50-100 r / (min·°); ·Simultaneously, the human-machine interaction and collaboration module sends a voice prompt to the pilot to "push the stick + adjust the pitch", which is suitable for the stall protection requirements of single-engine / twin-engine propeller aircraft.

[0059] Dynamic recovery control unit: This unit is used during the stall recovery phase. It dynamically adjusts the control force based on the aircraft's real-time flight status to avoid abrupt attitude changes caused by rigid control logic. The specific implementation is as follows: Calculation of safe minimum airspeed: First, calculate the safe minimum airspeed based on real-time flight status. The calculation formula is: in, For real-time aircraft quality, It is the acceleration due to gravity. The lift coefficient corresponding to the flight state ( , (Mach number) Given the current atmospheric density, This is the aerodynamic reference area for the aircraft.

[0060] Dynamic control command calculation: Thrust dynamic adjustment: The core rule is that the lower the airspeed, the greater the throttle increase. The formula for calculating the throttle increase ΔT is: in, This is the proportionality coefficient. This formula, representing the aircraft's real-time airspeed, ensures sufficient lift during the stall recovery phase.

[0061] Dynamic deflection of control surfaces: The core rule is that the greater the load, the smoother the elevator push-stick force, and the more precise the elevator push-stick command. The calculation formula is: in, For model compatibility coefficient, This formula is used to measure the real-time load of the aircraft and avoid sudden attitude changes caused by excessive force on the control stick under heavy load conditions.

[0062] Finally, a smooth stall recovery control command is output and sent to actuators such as the elevator and throttle to complete a smooth recovery after the stall angle of attack is exceeded.

[0063] In this embodiment, the human-computer interaction collaboration module includes a control command recognition unit, a multimodal status feedback unit, an emergency control triggering unit, and a control power flexible handover unit that are interconnected. The control command recognition unit is used to collect and recognize the pilot's control input commands and transmit them to the boundary protection decision module; The multimodal state feedback unit is used to output multi-dimensional system state prompts and control suggestions to the pilot based on the system operation status and flight status classification results; The emergency control triggering unit communicates with the boundary protection decision module to identify the pilot's emergency control maneuvers, triggering system intervention to suspend control and immediately transfer control to the pilot. The flexible handover unit is used to complete the flexible transfer of control from the system to the pilot based on the aircraft's restored stable flight status and the pilot's confirmed operation.

[0064] Specifically, the manipulation command recognition unit: This unit collects the pilot's input via a haptic feedback joystick, including joystick deflection, control force, and button trigger signals. It identifies the pilot's control intentions and commands in real time and transmits the identified commands to the boundary protection decision module and the adaptive execution control module as input for permission allocation and control command calculation.

[0065] Multimodal state feedback unit: Based on the flight status classification results and system operation status, this unit outputs multi-dimensional, full-process status prompts to the pilot, as specifically implemented as follows: Level 1 Normal Flight Status: No additional prompts, flight parameters are displayed normally only in the system background; Level 2 warning and intervention status: Warnings are issued through a multimodal approach, including visual (cabin display alarm prompts), auditory (voice alarms), and tactile (high-frequency vibration of the joystick), and targeted control suggestions are pushed out simultaneously; Level 3 Emergency Control Status: Continuously outputs strong visual and auditory alarms, and simultaneously prompts the system intervention status and the pilot's over-control mode, allowing the pilot to clearly grasp the system's operating status throughout the entire process; Sensor Failure / System Degradation Status: Synchronously outputs multimodal prompts of fault type, system degradation mode, and operation suggestions to avoid pilot misjudgment.

[0066] Emergency control trigger unit: This unit collects the control force applied by the pilot to the joystick in real time. When the control force is greater than or equal to the preset threshold, the emergency over-control mechanism is immediately triggered, and an over-control command is sent to the boundary protection decision module to suspend all automatic interventions of the system and immediately transfer control to the pilot, ensuring the pilot's absolute control priority in special scenarios.

[0067] Flexible transfer of control unit: This unit works in conjunction with the dynamic access control management unit of the boundary protection decision module. When the aircraft meets the basic conditions for handing over control, a flexible handover process is executed: a handover prompt is sent to the pilot via low-frequency continuous tactile vibration of the control stick, the pilot's confirmation operation is recognized (pressing the confirmation button or continuous normal operation for ≥1s), and a handover command is immediately sent to the boundary protection decision module after confirmation, completing the complete transfer of control and stopping all background interventions. If the pilot does not respond, the prompt is repeated every 3s until the pilot confirms or the flight status changes.

[0068] In other implementations, an active boundary protection control method for aircraft stall angle of attack is provided, based on any of the preceding active boundary protection control systems for aircraft stall angle of attack, including: Step S01: In response to the real-time acquisition trigger of the aircraft's flight status, redundant acquisition and fusion processing of the aircraft's multi-source flight parameters are performed to extract an effective flight status dataset. Step S02: Construct a dynamic stall boundary calculation model. Use the dynamic stall boundary calculation model to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions. Step S03: Obtain real-time aerodynamic parameters and attitude change data of the aircraft, and combine them with the extracted effective flight status dataset to obtain the current flight scenario type and flight safety status classification. Step S04: Based on the dynamic stall angle of attack threshold, flight scenario type and flight safety status decomposition, generate appropriate boundary protection control commands, perform adaptive control on the aircraft control surfaces and power mechanism, and complete the dynamic recovery control after the stall angle of attack is exceeded. Step S05: Feedback the system's operating status and operational suggestions to the pilot through the human-machine interaction link, and in response to the pilot's control trigger commands, complete the flexible switching of control between the system and the pilot and the execution of corresponding flight maneuvers.

[0069] In this embodiment, in response to the real-time acquisition trigger of the aircraft's flight status, redundant acquisition and fusion processing of the aircraft's multi-source flight parameters are performed to extract an effective flight status dataset, including: The aircraft angle of attack parameters are synchronously acquired and cross-validated by a three-level redundant angle of attack acquisition unit to obtain three channels of redundant raw angle of attack data. The sensor health status determination unit synchronously collects the aircraft's flight status and atmospheric environmental parameters to obtain auxiliary raw data such as airspeed, altitude, temperature, air pressure, load and flap position. By using an extended Kalman filter fusion module to preprocess, remove outliers, and dynamically weight the three redundant angle-of-attack raw data and auxiliary raw data, an effective flight state dataset is obtained.

[0070] In this embodiment, a dynamic stall boundary calculation model is constructed. This model is used to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions, including: A database of stall boundary references is constructed by predefining the basic stall angle of attack, standard atmospheric parameters, maximum rated load and aerodynamic configuration reference parameters of the aircraft model. The effective flight status dataset is standardized to extract atmospheric environmental parameters, flight load parameters, and flap position parameters under the current operating conditions. The atmospheric density under the current operating conditions is calculated based on atmospheric environmental parameters, and the environmental correction factor is obtained by combining it with standard atmospheric parameters. Based on flight load parameters and flap position parameters, combined with maximum rated load and aerodynamic configuration reference parameters, the flight state correction coefficient is calculated. By combining the aircraft's basic stall angle of attack, environmental correction coefficient, and flight state correction coefficient, the dynamic stall angle of attack threshold under the current operating condition is calculated. The corresponding buffer threshold is then obtained by synchronously matching the preset buffer threshold tuning rules.

[0071] In this embodiment, based on the dynamic stall angle of attack threshold, flight scenario type, and flight safety state decomposition, an appropriate boundary protection control command is generated to adaptively control the aircraft control surfaces and power mechanism, completing the dynamic recovery control after the stall angle of attack is exceeded, including: Based on the flight safety status classification, match the corresponding human-machine access control rules and basic control strategies; Based on the human-machine permission allocation rules and basic control strategies, for the identified flight scenario type, the appropriate scenario-specific control law is invoked to calculate and generate the corresponding aileron, rudder, and elevator deflection control commands and throttle thrust control commands. During the stall recovery phase, based on real-time airspeed and load conditions, the aileron, rudder, and elevator deflection control commands and throttle thrust control commands are optimized, and the thrust increase and elevator push rod force are dynamically adjusted to generate smooth recovery control commands. The final control commands are sent to the actuators, which drive the aircraft control surfaces and power mechanisms to complete the corresponding actions, thereby achieving adaptive protection of the stall boundary in all flight states and dynamic recovery control after the stall angle of attack is exceeded.

[0072] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. An active boundary protection control method for aircraft stall angle of attack, applied to an active boundary protection control system for aircraft stall angle of attack, wherein the system is based on a hierarchical decoupled closed-loop flight control protection architecture, comprising: The module consists of a multi-source redundant sensing module, a boundary protection decision module, an adaptive execution control module, and a human-machine interaction collaboration module, which are connected in sequence via communication. The multi-source redundant sensing module is used to collect multi-source parameters related to aircraft flight and output a valid flight status dataset. The boundary protection decision module is used to complete the flight status classification and dynamic allocation of human and machine permissions based on the effective flight status dataset, and output stall boundary protection control commands of the corresponding level. The adaptive execution control module is used to perform adaptive control of the aircraft control surfaces and power mechanism based on stall boundary protection control commands and in combination with flight scenarios, as well as dynamic recovery control after stall angle of attack is exceeded. The human-machine interaction and collaboration module is used to realize two-way human-machine interaction between the pilot and the system, and to complete the input of control commands, system status feedback and switching of operation control. The multi-source redundant sensing module includes a three-level redundant angle of attack acquisition unit, a multi-source auxiliary parameter acquisition unit, a sensor health status determination unit, and an extended Kalman filter fusion unit that are interconnected with each other. The three-level redundant angle of attack acquisition unit and the multi-source auxiliary parameter acquisition unit are used to complete the multi-channel redundant acquisition of aircraft angle of attack, flight status and atmospheric environment parameters, and output the raw acquisition dataset. The sensor health status determination unit is used to complete the real-time identification and classification of the health / fault status of each sensor and output sensor health status data. The extended Kalman filter fusion unit is used to preprocess the original acquired dataset, remove outliers, and perform dynamic weight fusion to output the optimal effective flight state dataset. The boundary protection decision module includes a dynamic stall threshold calculation unit, a flight status classification judgment unit, and a dynamic permission management unit that are interconnected. The dynamic stall threshold calculation unit is used to calculate the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions based on the effective flight state dataset. The flight status classification and determination unit is used to complete the real-time determination of three levels of flight status: normal flight, early warning intervention, and emergency control, based on the correspondence between real-time angle of attack and dynamic threshold. The dynamic permission management unit is used to dynamically allocate human-machine operation permissions based on the flight status classification results and output stall boundary protection control commands of the corresponding level. The adaptive execution control module includes a flight scene recognition unit, a scene-specific adaptive control law calculation unit, and a dynamic recovery control unit that are interconnected. The flight scene recognition unit is communicatively connected to the boundary protection decision module and is used to complete the real-time recognition of turbulence / crosswind, single engine failure / asymmetric thrust, and propeller stall flight scenes based on the effective flight state dataset. The scenario-specific adaptive control law solving unit is used to solve and output adapted control surface deflection control commands and power mechanism thrust control commands based on the identified flight scenario and stall boundary protection control commands. The dynamic recovery control unit is used to dynamically adjust the control intensity based on the real-time flight status of the aircraft during the stall recovery phase, and output smooth stall recovery control commands. The human-machine interaction collaboration module includes an operation command recognition unit, a multimodal status feedback unit, an emergency operation triggering unit, and a control power flexible handover unit that are interconnected. The control command recognition unit is used to collect and recognize the pilot's control input commands and transmit them to the boundary protection decision module; The multimodal state feedback unit is used to output multi-dimensional system state prompts and control suggestions to the pilot based on the system operation state and flight state classification results. The emergency control triggering unit is communicatively connected to the boundary protection decision module and is used to identify the pilot's emergency control operations, triggering system intervention to suspend control and immediately transfer control to the pilot. The flexible handover unit is used to complete the flexible transfer of control from the system to the pilot based on the aircraft's restored stable flight state and the pilot's confirmed operation. Its characteristic is that it further includes: In response to the real-time acquisition trigger of the aircraft's flight status, redundant acquisition and fusion processing of the aircraft's multi-source flight parameters are performed to extract an effective flight status dataset. A dynamic stall boundary calculation model is constructed. The dynamic stall boundary calculation model is used to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions. By acquiring real-time aerodynamic parameters and attitude change data of the aircraft, and combining them with the extracted effective flight status dataset, the current flight scenario type and flight safety status classification can be obtained. Based on the dynamic stall angle of attack threshold, flight scenario type and flight safety status decomposition, an appropriate boundary protection control command is generated to adaptively control the aircraft control surfaces and power mechanism, and to complete the dynamic recovery control after the stall angle of attack is exceeded. The system provides feedback on its operating status and operational suggestions to the pilot through the human-machine interaction link, and responds to the pilot's control commands to complete the flexible switching of control between the system and the pilot and the execution of corresponding flight maneuvers. A dynamic stall boundary calculation model is constructed. This model is then used to perform environmental correction and state adaptation calculations on the effective flight state dataset to obtain the dynamic stall angle of attack threshold and buffer threshold under the current flight conditions, including: A stall boundary baseline database is constructed by predefining the basic stall angle of attack, standard atmospheric parameters, and maximum rated load of the aircraft model. The effective flight status dataset is standardized to extract atmospheric environmental parameters, flight load parameters, and flap position parameters under the current operating conditions. The atmospheric density under the current operating conditions is calculated based on atmospheric environmental parameters, and the environmental correction factor is obtained by combining it with standard atmospheric parameters. Based on the flight load parameters and flap position parameters, combined with the maximum rated load, the flight state correction coefficient is calculated. By combining the aircraft's basic stall angle of attack, environmental correction coefficient, and flight state correction coefficient, the dynamic stall angle of attack threshold under the current operating condition is calculated. The corresponding buffer threshold is then obtained by synchronously matching the preset buffer threshold tuning rules.

2. The active boundary protection control method for aircraft stall angle of attack according to claim 1, characterized in that, In response to the real-time acquisition trigger of aircraft flight status, redundant acquisition and fusion processing of multi-source flight parameters of the aircraft are performed to extract an effective flight status dataset, including: The aircraft angle of attack parameters are synchronously acquired and cross-validated by a three-level redundant angle of attack acquisition unit to obtain three channels of redundant raw angle of attack data. The multi-source auxiliary parameter acquisition unit synchronously collects the aircraft's flight status and atmospheric environmental parameters to obtain auxiliary raw data such as airspeed, altitude, temperature, air pressure, load, and flap position. By using an extended Kalman filter fusion module to preprocess, remove outliers, and dynamically weight the three redundant angle-of-attack raw data and auxiliary raw data, an effective flight state dataset is obtained.

3. The active boundary protection control method for aircraft stall angle of attack according to claim 1, characterized in that, Based on the dynamic stall angle of attack threshold, flight scenario type, and flight safety state decomposition, appropriate boundary protection control commands are generated to adaptively control the aircraft control surfaces and power mechanism, completing dynamic recovery control after stall angle of attack exceedance, including: Based on the flight safety status classification, match the corresponding human-machine access control rules and basic control strategies; Based on the human-machine permission allocation rules and basic control strategies, for the identified flight scenario type, the appropriate scenario-specific control law is invoked to calculate and generate corresponding aileron, rudder, and elevator deflection control commands and throttle thrust control commands. During the stall recovery phase, based on real-time airspeed and load conditions, the aileron, rudder, and elevator deflection control commands and throttle thrust control commands are optimized, and the thrust increase and elevator push rod force are dynamically adjusted to generate smooth recovery control commands. The final control commands are then sent to the actuators to drive the aircraft control surfaces and power mechanisms to complete the corresponding actions, achieving adaptive protection of the stall boundary in all-flight conditions and dynamic recovery control after stall angle of attack is exceeded.

Images