Research On Method To Suppress Surplus Torque Of Aircraft Rudder Electric Loading System

Aircraft rudder electric loading system is a ground simulation device,through simulating the changes in the various force load of rudder to test the performance of it in different flight conditions.It can provide a reliable data source for the research and design,improvement and modification of the ground simulation device of the flight control system.Due to electric loading system adopts passive torque loading,the active movements of rudder will produce surplus disturbance,seriously affecting static and dynamic performance of the system.Therefore,it is of great theoretical research value and engineering practical significance to study the method for the aircraft rudder electric loading system to suppress the surplus torque.Firstly,in this paper,the overall hardware structure of the system is designed with rubber-metal buffer spring and permanent magnet DC torque motor as the main loading mechanism,based on the thorough research of aircraft rudder electric loading system for medium and small torque loading.The work principle of the system is analyzed,and evaluation indicators and performance indicators are put forward.Secondly,according to the research needs and loading requirements,the mathematical models of the main executive agency are established,and something else is studied,which consists of the determination of the stiffness coefficient of the rubber-metal buffer spring and the influence of the loading gradient on the amplitude-frequency characteristics.Thirdly,the study about the mechanism and characteristics of surplus torque disturbance is executed,the definition of surplus torque is given and its disturbance ways under different motion states of rudder are analyzed.Then,the relationship between the motion instruction of rudder and disturbance states of surplus torque is studied,according to the mathematical model of surplus torque and its weights.Forthly,a conventional compensation method for suppressing torque is proposed that combing rudder position feedforward signal and the feedback of torque velocity,and to some extent improving the control performance of the system,but it can't meet the requirements of stability,real-time and high-precision loading.Therefore,an intelligent control compensation method based on cerebellar neural network is studied.On the one hand,one compensation method is put forward comprising new cerebellar neural network feedforward and incremental PID feedback,using the weight adjusting algorithm of variable balanced learning constant to improve the network.Which can effectively coordinate the relationship between network convergence speed and system steady-state performance.On the other hand,to improve the speed of network learning and solve the difficulty of incremental PID parameter tuning,another compensation method with dual control functions is proposed combining improved cerebellar neural network feedforward and the fractional PID feedback based on iterative learning control,supplying important guarantee for the system to suppress surplus torque disturbances.Finally,the validity and rationality of intelligent control compensation methods are verified through MATLAB.