CN-121973978-A - Range extender control system and controller for unmanned aerial vehicle

CN121973978ACN 121973978 ACN121973978 ACN 121973978ACN-121973978-A

Abstract

The invention belongs to the technical field of aircraft control, in particular relates to a range extender control system for an unmanned aerial vehicle and a controller, and aims to solve the problems of limited endurance, poor flight stability and the like caused by discontinuous energy supply and delayed power response. The system comprises a range extender power source module, an electric energy conversion and distribution module, a flight state sensing module, a load power prediction module, a multi-target optimal scheduling module and a closed loop feedback execution module. The method comprises the steps of generating an optimal power generation instruction by combining real-time sensing of a flight state and load power prediction based on a recurrent neural network, realizing accurate rotation speed tracking through self-adaptive PID control, introducing a cost function combining fuel consumption, smooth power supply switching and battery health, dynamically weighting according to task types, improving endurance and system robustness, and having functions of fault emergency response, power battery hot plug and model online updating, and enhancing safety and maintainability.

Inventors

ZHANG XIAOMIN
ZHANG YUHUA

Assignees

江苏奥尼克电气股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. Range extender control system for unmanned aerial vehicle, characterized by comprising: the range extender power source module is used as an auxiliary power source for supplying power for the main direct current bus of the unmanned aerial vehicle; the electric energy conversion and distribution module is used for managing energy flow between the range extender power source module and the power battery pack; The flight state sensing module is used for acquiring flight state parameters of the unmanned aerial vehicle in real time to form a flight condition identification vector; The load power prediction module is used for predicting the total load power of the unmanned aerial vehicle in a future time period through a recurrent neural network model based on historical flight data and the flight condition identification vector; The multi-objective optimization scheduling module is used for receiving the prediction result of the total load power, the flight condition identification vector, the bus voltage deviation amount and the power battery state of charge value, and solving an optimal range extender objective power generation power sequence based on a comprehensive cost function comprising a fuel consumption rate penalty term, a power supply switching smoothness rewarding term and a battery health loss factor; And the closed loop feedback execution module is used for receiving the range extender target generated power sequence and converting the range extender target generated power sequence into a real-time target rotating speed signal which can be executed by the range extender electronic control unit so as to drive the range extender power source module.
2. The range extender control system for a drone of claim 1, wherein said flight status awareness module comprises: The inertial measurement unit, the barometric altimeter, the global positioning system receiver, the airspeed sensor and the flight control instruction interface are used for collecting acceleration, angular velocity, altitude, airspeed, attitude angle change rate and remote control instruction input characteristics; and the flight mode classification unit is used for analyzing the flight state parameters according to a preset threshold rule so as to output a current flight mode label.
3. The range extender control system for a unmanned aerial vehicle of claim 2, wherein the recurrent neural network model in the load power prediction module undergoes two-stage training prior to deployment, wherein the first stage uses real flight logs acquired offline for pre-training and the second stage uses newly accumulated flight data to fine-tune the model weights after each flight mission is completed.
4. The range extender control system for a drone of claim 3 wherein said multi-objective optimal scheduling module comprises: The cost function construction unit is used for mapping fuel consumption rate penalty items according to universal characteristic curves of the internal combustion engine, calculating power supply switching smoothness reward items according to target rotating speed differences of adjacent control period range extenders, and calculating battery health loss factors according to charge and discharge multiplying power, depth and temperature data uploaded by the battery management system; And the rolling optimization solving unit is used for adopting an improved dynamic programming algorithm to solve the range extender target generating power sequence of a plurality of step sizes in the future, which minimizes the comprehensive cost function, in each control period.
5. The range extender control system for a unmanned aerial vehicle according to claim 4, wherein in the cost function construction unit, the fuel consumption rate penalty term, the power supply switching smoothness bonus term and the battery health loss factor adopt a normalized weighted summation form, and weight coefficients thereof are preset according to the type of the flight mission and stored in a nonvolatile memory.
6. The range extender control system for a drone of claim 5, wherein the closed loop feedback execution module comprises: The target rotating speed conversion unit is used for converting the range extender target generated power sequence into a real-time target rotating speed signal according to an electromechanical coupling characteristic equation of the generator; And the self-adaptive PID control unit is used for adjusting the proportional gain and the integral time constant on line according to the current air inlet temperature, the atmospheric pressure and the engine warmup state so as to accurately track the real-time target rotating speed signal.
7. The range extender control system for the unmanned aerial vehicle of claim 6, wherein the closed loop feedback execution module is further provided with a fail-safe strategy unit for starting an emergency response program when detecting that the actual rotational speed of the range extender deviates from the target value by more than a preset threshold value and the duration reaches a preset duration, cutting off a power supply loop of the range extender, improving the output power of the power battery, and simultaneously sending a power-reduction flight command to the flight control system.
8. The range extender control system for the unmanned aerial vehicle of claim 7, wherein the electric energy conversion and distribution module is provided with a hot plug management unit for controlling the unloading and the connection of the power battery pack through a soft switching mechanism under the normal operation state of the range extender so as to avoid abrupt change of the voltage of the main direct current bus.
9. The range extender control system for the unmanned aerial vehicle of claim 8, wherein the starting process of the range extender power source module is completed by the reverse rotation of the power generator driven by the energy extracted from the power battery by the electric energy conversion and distribution module, and the power generation working condition is automatically switched after the starting is completed.
10. A controller for a range extender control system for an unmanned aerial vehicle, comprising: the flight state sensing unit is configured to acquire flight state parameters of the unmanned aerial vehicle in real time and form a flight condition identification vector; the load power prediction unit is in communication connection with the flight state sensing unit and is configured to predict the total load power of the unmanned aerial vehicle in a future time period through a recurrent neural network model based on historical flight data and the flight condition identification vector; The multi-target optimization scheduling unit is in communication connection with the load power prediction unit and is configured to receive the prediction result of the total load power, the flight condition identification vector, the bus voltage deviation amount and the power battery state of charge value, and solve an optimal range extender target power generation power sequence based on a comprehensive cost function comprising a fuel consumption rate penalty term, a power supply switching smoothness reward term and a battery health loss factor; The closed-loop feedback execution unit is in communication connection with the multi-target optimal scheduling unit and is configured to receive the range extender target generated power sequence and convert the range extender target generated power sequence into a real-time target rotating speed signal which can be executed by the range extender electronic control unit so as to drive the range extender power source module; it is characterized in that the method comprises the steps of, The flying state sensing unit further comprises an inertial measurement unit, an air pressure altimeter, a global positioning system receiver, an airspeed sensor and a flying control instruction interface, and is used for acquiring acceleration, angular velocity, altitude, airspeed, attitude angle change rate and remote control instruction input characteristics; The recurrent neural network model in the load power prediction unit is subjected to two-stage training before deployment, wherein the first stage uses real flight logs acquired offline for pre-training, and the second stage uses newly accumulated flight data to carry out fine adjustment updating on model weights after each flight task is finished; the multi-objective optimization scheduling unit further comprises a cost function construction subunit, a power supply switching smoothness rewarding item is calculated according to the target rotating speed difference of the range extender in the adjacent control period, and a battery health loss factor is calculated according to the charge and discharge multiplying power, depth and temperature data uploaded by the battery management system; The closed loop feedback execution unit further comprises a target rotating speed conversion subunit, an adaptive PID control subunit and an on-line regulation proportional gain and integral time constant, wherein the target rotating speed conversion subunit is used for converting the target generating power sequence of the range extender into a real-time target rotating speed signal according to an electromechanical coupling characteristic equation of the generator, and the adaptive PID control subunit is used for accurately tracking the real-time target rotating speed signal according to the current air inlet temperature, the atmospheric pressure and the engine warmup state.

Description

Range extender control system and controller for unmanned aerial vehicle Technical Field The invention belongs to the technical field of aircraft control, and particularly relates to a range extender control system for an unmanned aerial vehicle and a controller. Background Unmanned aerial vehicle technology is an important branch of modern aviation science and technology, has been widely used in a plurality of fields such as military reconnaissance, commodity circulation transportation, environmental monitoring, agricultural plant protection. Along with the continuous expansion of application scenes, the requirements on the endurance capacity of the unmanned aerial vehicle are increasingly improved, and particularly in long-distance and long-time operation tasks, the traditional battery power supply mode is difficult to meet the actual requirements. Under the background, the range extender is used as a technical scheme capable of continuously supplying power to the unmanned aerial vehicle through an internal combustion engine or an auxiliary power source such as a fuel cell and becomes one of key means for improving the endurance performance of the unmanned aerial vehicle. The control strategy of the range extender for the unmanned aerial vehicle directly influences the starting time, the running efficiency and the overall energy utilization rate of the range extender. The ideal control system needs to dynamically match the load requirement in the flight process, ensure smooth switching between the main power supply and the range extender, and maintain the stability of the power supply system, thereby maximally prolonging the flight time and guaranteeing the task continuity. In the prior art, although part of unmanned aerial vehicles are integrated with range extender modules to extend range, the control system of the unmanned aerial vehicle has the problems of lag response, unreasonable energy distribution and the like. For example, range extender start logic relies on a fixed threshold determination, and cannot be adaptively adjusted according to flight conditions, load changes, or environmental factors, resulting in a risk of fuel waste due to premature start or power interruption due to delayed start. Meanwhile, a real-time monitoring and fault pre-judging mechanism for the running state of the range extender is lacking, so that the reliability of the system is reduced under a complex working condition. In addition, the current control architecture often neglects the cooperative optimization of the whole energy flow, and the dynamic matching of the battery discharging strategy and the output power of the range extender is not realized, so that the further improvement of the whole energy efficiency is restricted. Therefore, there is a need for a range extender control system for an unmanned aerial vehicle with intelligent decision making capability to solve the above technical problems. Disclosure of Invention The invention aims to provide a range extender control system for an unmanned aerial vehicle, which aims to solve the technical problems of limited unmanned aerial vehicle endurance, reduced flight stability and insufficient task adaptability caused by discontinuous energy supply, delayed power response and difficult cooperative control of a multi-mode power source in the prior art. Along with the development of a vertical take-off and landing fixed-wing unmanned aerial vehicle, a long-endurance reconnaissance platform and a heavy logistics unmanned aerial vehicle, the requirement on a continuous high-energy-density energy supply system is increasingly urgent. While the conventional battery driving system is limited by the upper limit of specific energy and is difficult to meet long-time flight requirements, the conventional power-increasing Cheng Qiduo adopts constant-rotation-speed power generation or simple start-stop control strategies, and cannot effectively match the power requirements of the unmanned aerial vehicle dynamically changing in different flight stages, so that fuel oil waste, emission increase and vibration disturbance are easily caused, and the overall performance and reliability of the aircraft are further affected. The technical scheme of the invention is that the intelligent range extender comprises a range extender power source module, an electric energy conversion and distribution module, a flight state sensing module, a load power prediction module, a multi-target optimal scheduling module and a closed loop feedback execution module. The power source module of the range extender consists of an internal combustion engine and a generator which is coaxially connected, so as to form a hybrid power generation unit, and the output alternating current of the hybrid power generation unit is connected to a main direct current bus of the unmanned aerial vehicle after being rectified. The electric energy conversion and distribution module is responsible for carrying out grid-connecte