Disturbance Analysis And Control Of Electromechanical Actuator System

The electromechanical actuator(EMA)system is a high-precision position servo system,which is an important part of the aircraft's flight control system.Its performance directly determines the dynamic quality of the aircraft's flight control system.However,due to manufacturing process and installation accuracy,there are inevitably many nonlinear factors in EMA system,which seriously affects the dynamic and static performance of the EMA system,and even affects the performance of the aircraft.Therefore,it is very important to research the influence of disturbance factors such as backlash and friction on the dynamic and static characteristics of EMA systems,and to choose appropriate methods to reduce or eliminate the influence of these factors.This paper choose a certain type of aircraft EMA system as the research object,researches and analyzes the disturbance factors in the EMA system,and uses an improved sliding mode control based on PI(PI-ISM)and a sliding mode control based on the radial base function neural network(RBF-SMC)to eliminate or reduce the effects caused by friction and backlash to improve the tracking accuracy of the EMA system.This paper mainly includes the following six parts:(1)The overall scheme of the EMA system was determined based on the overall system indexes,including mechanical transmission scheme using ball screw and system control scheme using position-speed double closed loop.The detailed design of the EMA system is carried out,including load analysis and load matching analysis.The optimal transmission ratio of the system is determined,and the parameter design and selection of the servo motor and the parameter design of the reduction gear are introduced.(2)Considering the disturbance factors,such as backlash and friction,this paper studies the friction and backlash of the EMA system,and establishes the models of friction and backlash.Finally,the influence of friction and backlash on the performance of the EMA system is analyzed with actual engineering projects.(3)Aiming at the characteristics of non-linear and time-delay of the EMA system,this paper designs a sliding mode controller based on PI method.However,SMC has chattering problems,which may induce poor tracking performance and create undesirable oscillations.In contrast to traditional PID or traditional sliding mode controllers that missiles or aircrafts usually use,this paper utilizes improved sliding mode control based on novel reaching law to compensate static friction and eliminate the flat-top during the steering of the input signal,utilizes boundary layer and switching function to solve high-frequency chattering problem,and utilizes PID control to improve the performance index of EMA system.(4)In order to improve the robustness of EMAs,this paper proposes a sliding mode controller based on radial basis neural network(RBF-SMC).The determining part of the system is calculated by sliding mode control algorithm,and the uncertain part of the system is approximated by radial basis neural network(RBF),which can not only improve the anti-disturbance performance of the EMAs,but also reduce the chattering of sliding mode control.(5)Aiming at the problem that the network weight of RBF needs to be learned online and is not easy for engineering practice,this paper proposes a sliding mode controller of radial basis neural network based on minimum parameter method(MPRBF-SMC),which use the minimum parameter method to replace the network weight learning algorithm,and greatly simplifies the control algorithm.Then,we prove the stability of the control algorithm by use Lyapunov function.Finally,the feasibility of the control algorithm is verified through simulation analysis.The conclusion shows that the improved RBF sliding mode controller has good dynamic characteristics and robustness.(6)Finally,an experimental platform for the EMAs is built,and the PID controller,PI-ISM controller,and MPRBF-SMC controller are used to control the EMAs.Experiments indicate that the PI-ISM or MPRBF-SMC can evidently reduce the flattop time to 12 ms or 9ms when the PID controller has 64 ms.Besides,PI-ISM and MPRBF-SMC controllers can eliminate the trajectory limit circle oscillation.Compared with PID controller,PI-ISM controller and MPRBF-SMC controller provide better performance—less chattering,less flat-top,higher precision,and no oscillation.This research shows that the proposed EMA system and control schemes are feasible,and have good robustness and tracking accuracy.The research results of this paper have certain reference for the future research of EMA system,and also have certain reference value for further research in the future.