A distributed electric propulsion tilt-rotor loitering vehicle power system

Abstract

The present application relates to low altitude aircraft power system technical field, especially pointing to a kind of distributed electric propulsion's tilt rotor low altitude vehicle power system, including propulsion unit, tilt actuator, power battery pack, flight control computer and battery management system.The present application can calculate and compensate the aerodynamic coupling disturbance of transition state of tilt in real time through the transition mode linkage of aerodynamic-power coupling disturbance suppression module and power collaborative control module, so that power distribution unit can predict aerodynamic disturbance in advance, effectively suppress the aerodynamic nonlinear disturbance caused by mutual interference between wing and fuselage downwash airflow, propeller and the like, reduce the flight attitude fluctuation under transition mode;Meanwhile, power collaborative control module realizes the fine distribution of power based on model predictive control framework, considers actuator dynamic constraint and energy efficiency, so that vehicle is more stable in the mode conversion process of vertical take-off and landing, transition flight, high-speed cruise, and can adapt to complex low altitude aerodynamic environment.

Description

Technical Field

[0001] This invention relates to the field of low-altitude aircraft propulsion system technology, and in particular to a distributed electric propulsion tiltrotor low-altitude vehicle propulsion system. Background Technology

[0002] With the rapid development of low-altitude intelligent transportation vehicles, the requirements for power systems are increasing. Currently, the power systems of low-altitude intelligent transportation vehicles are mainly divided into two categories: pure electric and hybrid power systems, aiming to achieve cleaner and lower-carbon energy. In terms of configuration, eVTOL (electric vertical takeoff and landing aircraft) can be divided into three types: multi-rotor, compound wing, and tiltrotor. Among them, the tiltrotor configuration is widely considered the best solution due to its advantages in light weight, high thrust, speed, and range. As the preferred configuration for low-altitude intelligent transportation vehicles, the tiltrotor aircraft combines the vertical takeoff and landing capabilities of a rotorcraft with the high-speed cruise capabilities of a fixed-wing aircraft by changing the thrust direction of the power unit.

[0003] Existing tiltrotor propulsion systems mostly use mechanical linkages or hydraulic mechanisms to drive the tilt of the overall nacelle. Since the thrust output and tilt angle of each power unit are mostly mechanically linked or simply master-slave, there is no refined optimization cost function or model predictive control algorithm based on real-time flight status. The attitude fluctuations are large in the transition mode, and power saturation and thrust loss are prone to occur. It is impossible to achieve seamless connection between vertical take-off and landing and high-speed cruise. Therefore, it is necessary to design a distributed electric propulsion tiltrotor low-altitude vehicle propulsion system. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a distributed electric propulsion tiltrotor low-altitude vehicle power system.

[0005] The technical solution adopted in this invention is: a distributed electric propulsion tiltrotor low-altitude vehicle power system, including a propulsion unit, a tilt actuator, a power battery pack, a flight control computer, and a battery management system;

[0006] The flight control computer serves as the system control core, communicating and interacting with the propulsion unit, tilt actuator, and battery management system. The power battery pack provides electrical power to the propulsion unit, tilt actuator, and flight control computer. The battery management system monitors the status and manages the safety of the power battery pack. The propulsion unit adjusts its thrust direction under the drive of the tilt actuator. The flight control computer has an embedded intelligent power-coordinated control system, which achieves precise control of the propulsion unit and tilt actuator through hierarchical linkage and data interaction of various functional modules.

[0007] The propulsion unit is configured in multiple distributed configurations, including a high-power-density permanent magnet synchronous motor, a propeller, and a motor controller. Some propulsion units are mounted on the tilt mechanism to change the thrust direction with the tilt mechanism, thereby achieving seamless switching between vertical lift and horizontal cruise thrust. The propulsion unit adopts a parallel redundant design, and its motor operating parameters are uploaded to the flight control computer in real time to provide data support for parameter identification of the digital twin performance boundary prediction module.

[0008] The tilt actuator is connected to the tilt mechanism and adopts an electro-hydraulic actuator structure. It receives tilt angle commands from the flight control computer and drives the propulsion unit to make continuous closed-loop adjustments of the angle from 0° to 95° between the vertical take-off and landing attitude and the horizontal cruise attitude. Its tilt angle data is fed back to the flight control computer in real time, providing core judgment basis for the intelligent power cooperative control system.

[0009] The power battery pack is a modular structure composed of high-rate soft-pack lithium-ion batteries connected in series and parallel. It supplies power to the motor controllers, tilt actuators and flight control computers of each propulsion unit through a high-voltage DC bus. It is equipped with a modular thermal management structure to achieve temperature balance of each battery module. Its battery status data is collected by the battery management system and uploaded to the flight control computer.

[0010] The battery management system is connected to the flight control computer via a CAN bus. It includes a main management module and several slave management modules. It collects and reports the state of charge (SOC), cell voltage, temperature and SOC balance index of the power battery pack to the flight control computer in real time. At the same time, it performs battery overcharge, over-discharge and over-temperature protection and module thermal balance control. The balance index reported by the battery is directly used as the basis for energy balance optimization of the power cooperative control module of the flight control computer.

[0011] The flight control computer adopts a heterogeneous dual-core architecture of ARM Cortex-M7 and FPGA, and is connected to the motor controllers and tilt actuators of each propulsion unit via Ethernet signals. It issues control commands, collects high-speed operation data of each unit and mechanism, executes algorithm calculations, and realizes the timing linkage of each module of the intelligent power cooperative control system.

[0012] The flight control computer is embedded with an intelligent power-coordinated control system, which includes a power-coordinated control module, an aerodynamic-power coupling disturbance suppression module, a high-frequency vibration active suppression module, a fault diagnosis and reconfiguration unit, a digital twin performance boundary prediction module, and a flight experience book self-evolution module.

[0013] As a further description of the above technical solution:

[0014] The power coordination control module is the core control module of the intelligent power coordination control system, which coordinates real-time flight mode determination, accurate calculation of desired force and torque, optimized allocation of multiple power constraints, and low-level tracking and execution of control commands. The aerodynamic-power coupling disturbance suppression module is a dedicated auxiliary module for transition modes, connected between the mode identification unit and the power allocation unit of the power coordination control module. It is activated only during transition modes and provides aerodynamic disturbance compensation to the power coordination control module. The high-frequency vibration active suppression module, as a supplement to the low-level tracking control unit, outputs an anti-phase additional current command to the propulsion unit's motor controller to achieve active cancellation of high-frequency vibration. The fault diagnosis and reconfiguration... The unit detects the fault type of the propulsion unit in real time. When a fault occurs, it sends a fault signal to the power coordination control module and completes the rapid fault-tolerant reconstruction of the remaining power units, adjusting the power distribution logic of the power coordination control module. The digital twin performance boundary prediction module works in parallel with the power coordination control module, identifying the motor parameters of the propulsion unit online and dynamically predicting its actual performance boundary, updating the dynamic constraints to the power coordination control module in real time. The flight experience self-evolution module is deployed on the ground station and cloud server, communicating and interacting with the flight control computer. It updates the control parameters obtained through offline optimization to the power coordination control module of the flight control computer, realizing the continuous self-evolution of the system control performance.

[0015] As a further description of the above technical solution:

[0016] The power coordination control module includes a modal recognition unit, a desired force and torque calculation unit, a power distribution unit, and a low-level tracking control unit;

[0017] The mode recognition unit is used to receive feedback data from the inertial measurement unit (IMU), airspeed indicator, and tilt actuator. Based on the airspeed, attitude angle, and tilt angle signals of the aircraft, it determines the current flight mode in real time. The flight modes include vertical takeoff and landing mode, transition mode, and high-speed cruise mode.

[0018] The expected force and torque calculation unit is used to calculate the total thrust vector and expected three-axis torque values ​​required in the current mode body coordinate system according to the position and attitude control commands issued by the upper navigation computer of the flight control computer;

[0019] The power distribution unit is used to control the total thrust vector and the expected value of the three-axis torque, and to optimize the thrust and tilt angle of each propulsion unit. It constructs a cost function that considers actuator dynamic constraints and energy efficiency to solve for the thrust and tilt angle commands of each propulsion unit. The control cycle is 10ms to 50ms. The cost function is used to balance thrust tracking accuracy, control smoothness, and energy consumption. The core optimization logic is to minimize the weighted sum of thrust tracking error, control quantity changes, and energy consumption. The core formula of the cost function is as follows: The meanings of each parameter are as follows: The cost function value is used to measure the degree of optimization in power allocation; , , These are weighting coefficients for thrust tracking accuracy, control smoothness, and energy consumption, respectively. The actual output total thrust and torque vector of each propulsion unit. This represents the desired total thrust and torque vector; For the current control command, The control command from the previous moment; For the first Actual power consumption of each propulsion unit To advance the total number of units;

[0020] The underlying tracking control unit is used to convert the tension command into the speed and torque command of the motor controller through the motor torque and tension model, and to convert the tilt angle command into the position command of the tilt actuator, so as to drive each propulsion unit to execute precisely.

[0021] As a further description of the above technical solution:

[0022] The aerodynamic-dynamic coupling disturbance suppression module is connected between the mode identification unit and the power distribution unit. It is activated only in the transition mode. It is used to calculate the local airflow velocity of the propulsion unit based on the airspeed vector, the airframe rotation angular velocity, the propulsion unit installation position and tilt angle. It also queries the aerodynamic disturbance force and torque in combination with the three-dimensional interpolation table obtained from the CFD simulation experiment. The module generates the expected force and torque after compensation through feedforward compensation logic and superimposes it into the output of the expected force and torque calculation unit to achieve feedforward compensation of aerodynamic disturbance and effectively suppress attitude fluctuations in the transition mode.

[0023] As a further description of the above technical solution:

[0024] The high-frequency vibration active suppression module employs an adaptive notch filter combined with the LMS least mean square algorithm. Independent of the main control loop and as part of the underlying tracking control unit, it has a sampling frequency of no less than 10kHz. It generates a reference signal of the same frequency based on the motor speed. Using the vibration signals measured by accelerometers at key vibration points of the wing and fuselage as errors, it generates additional current commands through the LMS weight online adjustment logic and superimposes them onto the original current loop setpoint of the motor controller. This causes the motor to generate an electromagnetic force opposite to the phase of the vibration, actively canceling the high-frequency vibration of the propeller and improving the vehicle's ride comfort and structural lifespan.

[0025] As a further description of the above technical solution:

[0026] The fault diagnosis and reconfiguration unit configures an independent nonlinear extended state observer (NLESO) for each propulsion unit to estimate the deviation between the actual output and the commanded output of each propulsion unit in real time. When the deviation exceeds a set threshold for five consecutive control cycles, the corresponding propulsion unit is determined to have failed and fault-tolerant reconfiguration is triggered. The fault-tolerant reconfiguration is used to set the upper limit of available thrust of the failed propulsion unit to zero, adjust the dimension and parameters of the control efficiency matrix in real time, and adaptively optimize the weight of the cost function of the power distribution unit, increase the weight of attitude-related degrees of freedom, prioritize the stability of the aircraft attitude, and use the remaining thrust for position tracking.

[0027] As a further description of the above technical solution:

[0028] The digital twin performance boundary prediction module works in parallel with the power distribution unit, operating on the high-performance core of the flight control computer, the ground station, and the cloud server. It is used to identify key motor parameters online using the recursive least squares method (FFRLS) with a forgetting factor, characterize the degree of demagnetization of the motor magnets using a magnet health factor, dynamically predict the maximum available thrust and thrust response rate of the propulsion unit, and update the dynamic performance boundary in real time to the MPC constraints of the power distribution unit, replacing static constraints to ensure that control commands are always within the true physical limits of the propulsion unit. The core formula for the magnet health factor is: ,in, For the health factors of magnetic steel; The actual flux linkage of permanent magnets identified online using the FFRLS algorithm; The nominal flux linkage is for permanent magnets.

[0029] As a further description of the above technical solution:

[0030] The self-evolving flight experience book module is deployed on ground stations and cloud servers, interacting with the flight control computer via 4G, 5G, and satellite communication links. It is used to record flight data throughout the entire lifecycle of the power system. Using a Bayesian optimization algorithm, it constructs a multi-index weighted optimization objective with multiple optimization indicators such as attitude error integral, energy consumption, vibration RMS value, and false alarm rate. It optimizes the key parameters of the power cooperative control module offline, generates a personalized flight experience book for the vehicle, and updates it to the flight control computer before the next flight, realizing the continuous self-evolution of the system control performance.

[0031] As a further description of the above technical solution:

[0032] The battery management system is used to collect the SOC value of each battery module in real time, and calculate the state-of-charge balance of the power battery pack through the balance index. The core calculation formula of the balance index is: ,in, For the state of charge balance of the battery module; This represents the maximum SOC value for all battery modules. This represents the minimum SOC value for all battery modules. The SOC is the average value of all battery modules. The power distribution unit of the flight control computer is used to dynamically adjust the energy optimization weight according to the balance index, and introduces an energy balance penalty function into the cost function to incorporate battery energy balance into the power distribution optimization objective. A smaller energy consumption weight is set for the propulsion unit corresponding to the battery module with a higher SOC to encourage it to output more power; a larger energy consumption weight is set for the propulsion unit corresponding to the battery module with a lower SOC to limit its power output, thereby achieving balanced discharge of battery modules.

[0033] The present invention has the following beneficial effects:

[0034] 1. This invention, through the transition mode linkage of the aerodynamic-dynamic coupling disturbance suppression module and the power cooperative control module, can calculate and compensate for aerodynamic coupling disturbances in the tilt transition state in real time. This allows the power distribution unit to predict aerodynamic interference in advance, effectively suppressing aerodynamic nonlinear disturbances caused by the downwash airflow between the wing and fuselage, mutual interference between propellers, etc., and reducing flight attitude fluctuations in the transition mode. At the same time, the power cooperative control module realizes fine-grained power distribution based on the model predictive control framework, taking into account the dynamic constraints of actuators and energy efficiency, making the vehicle more stable during mode transitions in vertical take-off and landing, transition flight, and high-speed cruise. It can adapt to complex low-altitude aerodynamic environments and greatly improve the environmental adaptability of the vehicle.

[0035] 2. The high-frequency vibration active suppression module of the present invention adopts the LMS adaptive algorithm to actively cancel high-frequency vibrations of specific frequencies such as 1P and 2P generated by propeller rotation. The sampling frequency of this module is not less than 10kHz, which can respond to high-frequency vibrations in real time, effectively reducing the vibration amplitude of the wing and the vibration noise in the cabin, and improving the ride comfort. At the same time, it eliminates the fatigue damage of high-frequency vibration to the vehicle structure, reduces the risk of high-frequency fatigue failure of the structure, and significantly extends the structural service life of key components such as the tilting mechanism, wing, and propeller.

[0036] 3. The fault diagnosis and reconstruction unit of this invention achieves real-time detection of propulsion unit faults based on the NLESO algorithm. Combined with real-time performance data from the digital twin performance boundary prediction module, it can accurately determine the fault type and the upper limit of thrust. In the fault condition of complete failure of a single wingtip propulsion unit, the remaining power units can be reconstructed in a fault-tolerant manner within a long time, prioritizing the stability of the vehicle's attitude and ensuring that the position tracking accuracy does not deviate significantly. At the same time, the digital twin performance boundary prediction module identifies motor parameters online and dynamically predicts performance boundaries. It can detect performance degradation problems such as demagnetization of motor magnets and wear of bearings in real time, avoiding control saturation or instability caused by the discrepancy between the actual thrust boundary and the theoretical boundary. It improves the flight safety of the vehicle from three dimensions: fault prevention, fault detection, and fault reconstruction. In addition, the parallel redundancy design of the distributed propulsion unit achieves fault-safe redundancy without sacrificing efficiency and weight, further strengthening safety assurance. Attached Figure Description

[0037] Figure 1 This is a framework diagram of the system of the present invention. Detailed Implementation

[0038] Reference Figure 1 The present invention provides a distributed electric propulsion tiltrotor low-altitude vehicle power system, including a propulsion unit, a tilt actuator, a power battery pack, a flight control computer, and a battery management system;

[0039] The propulsion unit is the core of the power output, consisting of a high-power-density permanent magnet synchronous motor, a propeller, and a motor controller. It employs a multi-unit parallel redundancy design, with some propulsion units mounted on the tilting mechanism and the remainder as fixed propulsion units. The permanent magnet synchronous motor features a high-power-density and high-torque-density design, integrating a field-weakening speed-spreading algorithm to meet the high-dynamic flight requirements of low-altitude vehicles. The propeller is made of carbon fiber, offering lightweight design, high aerodynamic efficiency, and reduced rotational drag and vibration. The motor controller integrates FOC vector control, motor status monitoring, overcurrent protection, and overtemperature protection functions, communicating with the flight control computer via CAN. Each controller independently drives one propulsion unit, providing a physical basis for fault tolerance. The core function of this unit is to convert electrical energy into mechanical energy, receiving flight control algorithm commands through the motor controller to precisely output the lift and thrust required for vehicle flight.

[0040] The tilt actuator employs an electro-hydraulic actuator structure, connected to the tilt mechanism. Through closed-loop position control, it achieves precise adjustment of the propulsion unit's tilt angle, ranging from 0° to 95° with a control accuracy of no less than ±0.5°. 0° corresponds to the propeller axis pointing vertically upward (vertical takeoff and landing mode), and 95° corresponds to the propeller axis pointing horizontally forward (high-speed cruise mode). The angle can be adjusted continuously and smoothly during transition modes. The electro-hydraulic actuator features high response, high load capacity, and anti-interference characteristics, adapting to complex aerodynamic environments at low altitudes. Its core function is to receive tilt angle commands from the flight control algorithm, enabling precise switching of the propulsion unit's thrust direction and providing mechanical support for all-mode flight.

[0041] The power battery pack is the core of the system's energy supply. It is a modular structure composed of high-rate soft-pack lithium-ion batteries connected in series and parallel, with a rated voltage of 800V. Power is supplied to each motor controller via a high-voltage DC bus, meeting the high-rate discharge requirements of the distributed propulsion unit. The battery pack adopts a modular design and modular thermal management architecture. Each battery module is equipped with independent water-cooling and air-cooling thermal management units to achieve temperature balance across modules and prevent localized overheating. The modules use a series-parallel connection to improve redundancy and reliability. Its core function is to provide clean, stable, and high-rate electrical energy to the entire power system, enabling modular energy storage and supply, and providing energy assurance for the execution of instructions from the algorithm module.

[0042] The battery management system adopts a master-slave architecture, including a master management module and several slave management modules. Each slave module corresponds one-to-one with a battery module, enabling cell-level state monitoring. The system collects real-time data on the state of charge (SOC, detection accuracy ≥ ±2%), individual cell voltage, and temperature of each module. It calculates the degree of SOC balance among the battery modules using a balance index; the core calculation formula is as follows: ,in: The state of charge balance of the battery module (values ​​range from 0 to 1, with a higher value of 1 indicating more balanced discharge). This represents the maximum SOC value for all battery modules. This represents the minimum SOC value for all battery modules. It calculates the average SOC of all battery modules; it also performs overcharge, over-discharge, and over-temperature protection, as well as module thermal equalization control. After aggregating the data, the main management module interacts with the flight control computer via the CAN bus, providing precise data support for energy balance optimization of power distribution (introducing an energy balance penalty function). Its core function is to ensure battery safety, achieve refined battery management, and provide a data foundation for the coordinated optimization of power and energy.

[0043] The flight control computer serves as the core of the system's control and algorithm computation. It employs a heterogeneous dual-core architecture of ARM Cortex-M7 and FPGA, running a real-time operating system, and boasts high computing power, high real-time performance, and high reliability. The ARM Cortex-M7 is responsible for real-time control command issuance, multi-source data acquisition, and task scheduling and priority management for each module. The FPGA handles high-speed algorithm computation (MPC, LMS, NLESO, FFRLS, etc.), data preprocessing, timing synchronization of each module, and high-speed output of motor control signals, effectively improving algorithm efficiency and control real-time performance. The flight control computer connects to the distributed propulsion unit and tilt actuator via Ethernet, to the battery management system via CAN bus, and to the flight experience self-evolution module on the ground station and cloud server via 4G, 5G, and satellite communication. It also provides the operating platform for the six major algorithm modules of the intelligent power-coordinated control system. Its core function is to coordinate the operation of the entire power system, complete the computation of various core algorithms, issue and provide feedback on control commands, and achieve deep collaboration among various hardware units and algorithm modules.

[0044] The flight control computer is embedded with an intelligent power-coordinated control system, which includes a power-coordinated control module, an aerodynamic-power coupling disturbance suppression module, a high-frequency vibration active suppression module, a fault diagnosis and reconfiguration unit, a digital twin performance boundary prediction module, and a flight experience book self-evolution module.

[0045] The power coordination control module is the core control module of the system, serving as the decision-making and command center for the entire system. It comprises a modal recognition unit, a desired force and torque calculation unit, a power distribution unit, and a low-level tracking control unit. These four sub-units form a closed-loop control chain of perception, calculation, distribution, and execution, coordinating the start-stop logic and data interaction of other algorithm modules. The core algorithm is cost function multi-constraint optimization and model predictive control (MPC). Its specific design and functions are as follows:

[0046] Modal recognition unit: As the sensing core for modal determination of the entire system, it collects real-time data from the inertial measurement unit (IMU) on three axes of acceleration and angular velocity, the airspeed indicator's true airspeed, and the tilt angle feedback data from the tilt actuator through the high-speed data interface of the flight control computer. The sampling frequency is synchronized with the power distribution unit control cycle (10ms-50ms). Its core logic is modal threshold determination based on multi-dimensional data fusion. The specific determination rules are as follows: when airspeed < 10m / s and average tilt angle < 10°, it is determined to be vertical takeoff and landing mode; when 10m / s ≤ airspeed ≤ 30m / s or 10° ≤ tilt angle ≤ 85°, it is determined to be transition mode; when airspeed > 30m / s and tilt angle > 85°, it is determined to be high-speed cruise mode. To avoid abrupt changes in commands during mode switching, the unit incorporates a modal smoothing transition mechanism. The modal determination results are processed by a first-order low-pass filter to ensure the continuity of control commands during mode switching. Simultaneously, the modal determination signal is synchronized in real time to the aerodynamic-dynamic coupling disturbance suppression module (as its start / stop trigger signal), the desired force and torque calculation unit (as the basis for algorithm adaptation), and the power distribution unit (as the basis for optimization weight adjustment), thereby achieving full system modal adaptation.

[0047] The desired force and torque calculation unit, as the core of target transformation, receives trajectory tracking commands (position or attitude commands) from the upper-level navigation computer of the flight control system. Based on the judgment results of the modal recognition unit, it adaptively selects the calculation algorithm: In vertical takeoff and landing mode, it employs an Active Disturbance Rejection Control (ADRC) algorithm to suppress airflow disturbances during vertical takeoff and landing through an extended state observer, improving lift control accuracy; in transition mode, it uses PID and feedforward compensation algorithms, with the feedforward term being the compensation amount from the aerodynamic-dynamic coupling disturbance suppression module, achieving rapid tracking of target force and torque; in high-speed cruise mode, it uses a simplified PID algorithm to reduce computational complexity and ensure real-time control. During the calculation process, the upper-level navigation commands are first converted into desired motion states (acceleration, angular velocity) in the body coordinate system. Then, the required total thrust vector and three-axis torque expectations are obtained through inverse calculation using a dynamic model, forming the desired force and torque vector. The calculation error is controlled within ±2%, ensuring the accuracy of subsequent power distribution.

[0048] The power distribution unit uses the total thrust vector and the expected values ​​of the three-axis torques as control objectives, and constructs a multi-constraint optimization cost function (the core formula is...). The meanings of each parameter are as follows: The cost function value is used to measure the degree of optimization in power allocation; , , These are weighting coefficients for thrust tracking accuracy, control smoothness, and energy consumption, respectively. The actual output total thrust and torque vector of each propulsion unit. The desired total thrust and torque vector; For the current control command, The control command from the previous moment; This represents the actual power consumption of the i-th propulsion unit. The total number of propulsion units is given; this cost function takes into account both actuator dynamic constraints and energy efficiency, and obtains the tension and tilt angle commands for each propulsion unit, with a control cycle of 10ms to 50ms.

[0049] The underlying tracking control unit converts algorithm instructions into executable instructions for the motor controller and tilting actuator, thus achieving the connection between the algorithm and the hardware.

[0050] The aerodynamic-dynamic coupling disturbance suppression module is a dedicated auxiliary control module for transition modes, connected between the mode identification unit and the power distribution unit. It is activated only in the transition mode. The core algorithm is the calculation of local airflow combined velocity and the INDI incremental nonlinear dynamic inverse feedforward compensation algorithm: the local airflow combined velocity, angle of attack and sideslip angle of each propulsion unit are calculated by using airspeed vector, airframe rotation angular velocity, propulsion unit position and tilt angle. Combined with the interpolation table of CFD simulation experiment, the aerodynamic disturbance force and torque are queried. The corrected expected force and torque are generated by feedforward compensation logic, so that the power distribution unit can predict the aerodynamic disturbance in advance, realize the active suppression of aerodynamic disturbance in the transition mode, and reduce attitude fluctuation.

[0051] The high-frequency vibration active suppression module is independent of the main control loop or as part of the underlying tracking control unit. It has a sampling frequency of no less than 10kHz, the highest task priority, and the core algorithm is the LMS least mean square adaptive algorithm, which is used to actively cancel the 1P and 2P high-frequency vibrations of the distributed propulsion unit. It generates sine and cosine reference signals with the same frequency as the vibration based on the motor speed. Based on the vibration error signal, it generates an inverse current command through the LMS weight online adjustment logic and superimposes it on the motor current loop setpoint, so that the motor generates an inverse electromagnetic force, actively cancels the high-frequency vibration, and improves ride comfort and structural life.

[0052] The fault diagnosis and reconfiguration unit is equipped with an independent observer for each propulsion unit. The core algorithm is the NLESO nonlinear extended state observer algorithm, which realizes real-time fault detection and rapid fault-tolerant reconfiguration: the deviation between the actual output and commanded output of each propulsion unit is estimated by the NLESO algorithm. After the fault is determined, the upper limit of the thrust of the faulty unit is determined, the control efficiency matrix is ​​adjusted and the cost function weights are optimized. Attitude stability is prioritized, and the remaining healthy units are used to complete the power reconfiguration, thereby improving flight safety.

[0053] The digital twin performance boundary prediction module works in parallel with the power distribution unit. The core algorithm is FFRLS recursive least squares with a forgetting factor and a performance boundary prediction algorithm, which realizes online identification of motor parameters and perception of performance degradation. The module supports local deployment of the flight control computer and dual-mode deployment of ground station and cloud. The specific implementation logic is as follows:

[0054] Flight control computer local deployment: The module runs on the high-performance FPGA core of the flight control computer. The flight control computer adopts a heterogeneous dual-core architecture of ARM Cortex-M7 and FPGA. The FPGA has powerful parallel computing capabilities, which can quickly process real-time data from multiple sources such as motor voltage, current, and speed, and complete FFRLS parameter identification, magnet health factor calculation and maximum available thrust prediction. The computing latency is matched with the control cycle of the power distribution unit (10ms-50ms) to ensure that the dynamic performance boundary is updated to the MPC constraints in real time.

[0055] Ground station and cloud deployment: The module is deployed on the ground station server and cloud computing platform. It establishes a real-time data transmission channel with the flight control computer through 4G, 5G and satellite communication links. The flight control computer uploads motor operation data and flight status data to the ground station and cloud in real time. The module uses the high-performance computing resources (multi-core CPU, GPU) of the ground station and cloud to complete complex calculations and transmits the predicted dynamic performance boundaries (maximum available thrust, thrust response rate) back to the power distribution unit of the flight control computer in real time. It is suitable for scenarios with additional redundancy requirements for local computing power of the flight control.

[0056] Two deployment modes can be switched according to the actual application scenario, while the core algorithm and functions remain consistent. The module identifies key motor parameters online using the FFRLS algorithm based on the voltage equation of the permanent magnet synchronous motor, and uses the magnet health factor (the core formula is...) To assess motor performance degradation, the meanings of each parameter are as follows: The magnet health factor (values ​​range from 0 to 1, with values ​​closer to 1 indicating better magnet performance). The actual flux linkage of permanent magnets identified online using the FFRLS algorithm; The nominal flux linkage of the permanent magnet (the standard value for motor design) is specified; the maximum available thrust and thrust response rate of the propulsion unit are dynamically predicted and updated in real time to the constraints of the MPC control to ensure that the commands are within the actual physical limits.

[0057] The flight experience book self-evolution module is deployed on ground stations and cloud servers, working offline and non-real-time with the flight control computer. The core algorithm is the Bayesian optimization algorithm, which realizes personalized self-evolution of control parameters: it records flight data throughout the entire life cycle of the power system, uses the MPC weight matrix, fault diagnosis threshold, etc. as parameters to be optimized, constructs a multi-index weighted optimization objective, obtains the optimal parameter set through offline iteration of Bayesian optimization, compiles it into a personalized flight experience book and updates it to the flight control computer, realizing the system's self-evolution capability of getting better with each flight.

[0058] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A distributed electric propulsion tiltrotor low-altitude vehicle power system, characterized in that, This includes the propulsion unit, tilting actuator, power battery pack, flight control computer, and battery management system; The flight control computer serves as the system control core, communicating and interacting with the propulsion unit, tilt actuator, and battery management system. The power battery pack provides electrical power to the propulsion unit, tilt actuator, and flight control computer. The battery management system monitors the status and manages the safety of the power battery pack. The propulsion unit adjusts its thrust direction under the drive of the tilt actuator. The flight control computer has an embedded intelligent power-coordinated control system, which achieves precise control of the propulsion unit and tilt actuator through hierarchical linkage and data interaction of various functional modules. The propulsion unit is provided in multiple and distributed manner, including a high power density permanent magnet synchronous motor, a propeller and a motor controller. Some propulsion units are installed on the tilting mechanism to change the thrust direction with the tilting mechanism, so as to achieve seamless switching between vertical lift and horizontal cruise thrust. The propulsion unit adopts a parallel redundant design, and its motor operating parameters are uploaded to the flight control computer in real time, providing data support for parameter identification of the digital twin performance boundary prediction module. The tilt actuator is connected to the tilt mechanism and adopts an electro-hydraulic actuator structure. It receives tilt angle commands from the flight control computer and drives the propulsion unit to make continuous closed-loop adjustments of the angle from 0° to 95° between the vertical take-off and landing attitude and the horizontal cruise attitude. Its tilt angle data is fed back to the flight control computer in real time, providing core judgment basis for the intelligent power cooperative control system. The power battery pack is a modular structure composed of high-rate soft-pack lithium-ion batteries connected in series and parallel. It supplies power to the motor controllers, tilt actuators and flight control computers of each propulsion unit through a high-voltage DC bus. It is equipped with a modular thermal management structure to achieve temperature balance of each battery module. Its battery status data is collected by the battery management system and uploaded to the flight control computer. The battery management system is connected to the flight control computer via a CAN bus. It includes a main management module and several slave management modules. It collects and reports the state of charge (SOC), cell voltage, temperature and SOC balance index of the power battery pack to the flight control computer in real time. At the same time, it performs battery overcharge, over-discharge and over-temperature protection and module thermal balance control. The balance index reported by the battery is directly used as the basis for energy balance optimization of the power cooperative control module of the flight control computer. The flight control computer adopts a heterogeneous dual-core architecture of ARM Cortex-M7 and FPGA, and is connected to the motor controllers and tilt actuators of each propulsion unit via Ethernet signals. It issues control commands, collects high-speed operation data of each unit and mechanism, executes algorithm calculations, and realizes the timing linkage of each module of the intelligent power cooperative control system. The flight control computer is embedded with an intelligent power-coordinated control system, which includes a power-coordinated control module, an aerodynamic-power coupling disturbance suppression module, a high-frequency vibration active suppression module, a fault diagnosis and reconfiguration unit, a digital twin performance boundary prediction module, and a flight experience book self-evolution module.

2. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The power coordination control module is the core control module of the intelligent power coordination control system, which coordinates real-time flight mode determination, accurate calculation of desired force and torque, optimized allocation of multiple power constraints, and low-level tracking and execution of control commands. The aerodynamic-power coupling disturbance suppression module is a dedicated auxiliary module for transition modes, connected between the mode identification unit and the power allocation unit of the power coordination control module. It is activated only during transition modes and provides aerodynamic disturbance compensation to the power coordination control module. The high-frequency vibration active suppression module, as a supplement to the low-level tracking control unit, outputs an anti-phase additional current command to the propulsion unit's motor controller to achieve active cancellation of high-frequency vibration. The fault diagnosis and reconfiguration... The unit detects the fault type of the propulsion unit in real time. When a fault occurs, it sends a fault signal to the power coordination control module and completes the rapid fault-tolerant reconstruction of the remaining power units, adjusting the power distribution logic of the power coordination control module. The digital twin performance boundary prediction module works in parallel with the power coordination control module, identifying the motor parameters of the propulsion unit online and dynamically predicting its actual performance boundary, updating the dynamic constraints to the power coordination control module in real time. The flight experience self-evolution module is deployed on the ground station and cloud server, communicating and interacting with the flight control computer. It updates the control parameters obtained through offline optimization to the power coordination control module of the flight control computer, realizing the continuous self-evolution of the system control performance.

3. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The power coordination control module includes a modal recognition unit, a desired force and torque calculation unit, a power distribution unit, and a low-level tracking control unit; The mode recognition unit is used to receive feedback data from the inertial measurement unit (IMU), airspeed indicator, and tilt actuator. Based on the airspeed, attitude angle, and tilt angle signals of the aircraft, it determines the current flight mode in real time. The flight modes include vertical takeoff and landing mode, transition mode, and high-speed cruise mode. The expected force and torque calculation unit is used to calculate the total thrust vector and expected three-axis torque values ​​required in the current mode body coordinate system according to the position and attitude control commands issued by the upper navigation computer of the flight control computer; The power distribution unit is used to control the total thrust vector and the expected value of the three-axis torque, and to optimize the thrust and tilt angle of each propulsion unit. It constructs a cost function that considers actuator dynamic constraints and energy efficiency to solve for the thrust and tilt angle commands of each propulsion unit. The control cycle is 10ms to 50ms. The cost function is used to balance thrust tracking accuracy, control smoothness, and energy consumption. The core optimization logic is to minimize the weighted sum of thrust tracking error, control quantity changes, and energy consumption. The core formula of the cost function is as follows: The meanings of each parameter are as follows: The cost function value is used to measure the degree of optimization in power allocation; , , These are weighting coefficients for thrust tracking accuracy, control smoothness, and energy consumption, respectively. The actual output total thrust and torque vector of each propulsion unit. This represents the desired total thrust and torque vector; For the current control command, The control command from the previous moment; For the first Actual power consumption of each propulsion unit To advance the total number of units; The underlying tracking control unit is used to convert the tension command into the speed and torque command of the motor controller through the motor torque and tension model, and to convert the tilt angle command into the position command of the tilt actuator, so as to drive each propulsion unit to execute precisely.

4. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The aerodynamic-dynamic coupling disturbance suppression module is connected between the mode identification unit and the power distribution unit. It is activated only in the transition mode. It is used to calculate the local airflow velocity of the propulsion unit based on the airspeed vector, the airframe rotation angular velocity, the propulsion unit installation position and tilt angle. It also queries the aerodynamic disturbance force and torque in combination with the three-dimensional interpolation table obtained from the CFD simulation experiment. The module generates the expected force and torque after compensation through feedforward compensation logic and superimposes it into the output of the expected force and torque calculation unit to achieve feedforward compensation of aerodynamic disturbance and effectively suppress attitude fluctuations in the transition mode.

5. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The high-frequency vibration active suppression module employs an adaptive notch filter combined with the LMS least mean square algorithm. Independent of the main control loop and as part of the underlying tracking control unit, it has a sampling frequency of no less than 10kHz. It generates a reference signal of the same frequency based on the motor speed. Using the vibration signals measured by accelerometers at key vibration points of the wing and fuselage as errors, it generates additional current commands through the LMS weight online adjustment logic and superimposes them onto the original current loop setpoint of the motor controller. This causes the motor to generate an electromagnetic force opposite to the phase of the vibration, actively canceling the high-frequency vibration of the propeller and improving the vehicle's ride comfort and structural lifespan.

6. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The fault diagnosis and reconfiguration unit configures an independent nonlinear extended state observer (NLESO) for each propulsion unit to estimate the deviation between the actual output and the commanded output of each propulsion unit in real time. When the deviation exceeds a set threshold for five consecutive control cycles, the corresponding propulsion unit is determined to have failed and fault-tolerant reconfiguration is triggered. The fault-tolerant reconfiguration is used to set the upper limit of available thrust of the failed propulsion unit to zero, adjust the dimension and parameters of the control efficiency matrix in real time, and adaptively optimize the weight of the cost function of the power distribution unit, increase the weight of attitude-related degrees of freedom, prioritize the stability of the aircraft attitude, and use the remaining thrust for position tracking.

7. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The digital twin performance boundary prediction module works in parallel with the power distribution unit, operating on the high-performance core of the flight control computer, the ground station, and the cloud server. It is used to identify key motor parameters online using the recursive least squares method (FFRLS) with a forgetting factor, characterize the degree of demagnetization of the motor magnets using a magnet health factor, dynamically predict the maximum available thrust and thrust response rate of the propulsion unit, and update the dynamic performance boundary in real time to the MPC constraints of the power distribution unit, replacing static constraints to ensure that control commands are always within the true physical limits of the propulsion unit. The core formula for the magnet health factor is: ,in, For the health factors of magnetic steel; The actual flux linkage of permanent magnets identified online using the FFRLS algorithm; The nominal flux linkage is for permanent magnets.

8. The distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The self-evolving flight experience book module is deployed on ground stations and cloud servers, interacting with the flight control computer via 4G, 5G, and satellite communication links. It is used to record flight data throughout the entire lifecycle of the power system. Using a Bayesian optimization algorithm, it constructs a multi-index weighted optimization objective with multiple optimization indicators such as attitude error integral, energy consumption, vibration RMS value, and false alarm rate. It optimizes the key parameters of the power cooperative control module offline, generates a personalized flight experience book for the vehicle, and updates it to the flight control computer before the next flight, realizing the continuous self-evolution of the system control performance.

9. A distributed electric propulsion tiltrotor low-altitude vehicle power system according to claim 1, characterized in that, The battery management system is used to collect the SOC value of each battery module in real time, and calculate the balance of the state of charge of the power battery pack through the balance index. The core calculation formula of the balance index is: ,in, For the state of charge balance of the battery module; This represents the maximum SOC value for all battery modules. This represents the minimum SOC value for all battery modules. The SOC is the average value of all battery modules. The power distribution unit of the flight control computer is used to dynamically adjust the energy optimization weight according to the balance index, and introduces an energy balance penalty function into the cost function to incorporate battery energy balance into the power distribution optimization objective. A smaller energy consumption weight is set for the propulsion unit corresponding to the battery module with a higher SOC to encourage it to output more power; a larger energy consumption weight is set for the propulsion unit corresponding to the battery module with a lower SOC to limit its power output, thereby achieving balanced discharge of battery modules.

Images