Research On Control Technology Of High Performance Integrated Electromechanical Actuator(EMA)

The electromechanical actuator(EMA)is integrated by permanent magnet synchronous motor,harmonic driver,driver and controller.EMA has the characteristics of compact structure,high power density,high torque density and high control precision.It is widely used in the fields of manipulator,service robot,aerospace,precision instrument and so on.EMA is a multivariable,strongly coupled and nonlinear controlled object.There are time-varying parameters,friction,clearance,variable load and other factors in the operation process,which puts forward higher requirements for the control technology of EMA.Cascade PID control strategy is commonly used in EMA control system.This control method has simple structure and easy to realize,but there are also some problems,such as low position tracking accuracy,weak anti disturbance ability,slow response speed,incompatibility,fast and overshoot.Based on improving the control accuracy and anti disturbance of EMA,aiming at the shortcomings of classical control methods in the process of EMA control,this thesis puts forward a new control method,makes theoretical analysis and experimental research on this method,and verifies the feasibility and effectiveness of the new method through simulation and experiment.The research work of this thesis mainly includes the following five aspects:(1)The cascade three closed-loop control principle of current loop,speed loop and position loop of permanent magnet synchronous motor is analyzed,the mathematical model of permanent magnet synchronous motor is established,and the relationship between control current and torque is obtained,which provides a theoretical basis for the establishment of subsequent EMA friction model.The current loop and speed loop control models of permanent magnet synchronous motor are deduced,and the PI control parameters are obtained by type II system parameter tuning method and particle swarm optimization algorithm,and the PI parameter control performance obtained by these two methods is analyzed.(2)Aiming at the shortcomings of PI controller in EMA,a control scheme of ADRC instead of PI is proposed.Since fal(e,,α,δ)in ADRC has an inflection point near the origin,which is easy to cause buffeting problem,a new newfal(e,,α,δ)function is designed to realize the smooth transition of interval switching.Based on newfal(e,,α,δ),the improved first-order ADRC of speed loop and the improved second-order position position/speed loop second-order composite ADRC are designed.The Lyapunov stability of the improved ADRC is analyzed,and the sufficient conditions for the stability of the system are obtained.Through simulation,the action law of the main control parameters of TD,ESO and NLSEF in ADRC is analyzed in detail,and the improved particle swarm optimization algorithm is used to optimize the parameters of second-order ADRC.Taking the permanent magnet synchronous motor in EAM as the control object,the control performance of PI,ADRC and IADRC is verified by simulation and experiment,and the simulation and experimental results are evaluated and analyzed by using the error integral criterion.(3)The influence of two nonlinear factors of harmonic driver clearance and friction on the control accuracy in EMA is analyzed.Several common clearance models and friction models in mechanical system are described.The influence of different clearances on position tracking accuracy is analyzed by simulation.The dynamic model of EMA with friction is deduced,and Stribeck static friction model and Lu Gre dynamic friction model are established.In the process of friction model parameter identification,wavelet Kalman filter combined denoising method is used to process the measured data,and the optimal estimation of i_q is obtained.By measuring the control current i_q at different speeds,the torque speed model of EMA is obtained.The identified friction model is added to the control model by feedforward compensation.In order to realize the adaptive function of ADRC,the fuzzy controller is used to adjust the control parameters of IADRC in real time to realize the adaptive function.The effectiveness of fuzzy IADRC with friction feedforward compensation is verified by experimental method.(4)For sensorless control of permanent magnet synchronous motor,a new sensorless sliding mode control strategy is proposed.The rotor position estimation algorithm of sliding mode observer based on hyperbolic tangent function is derived,and the sufficient conditions for entering the sliding mode surface are obtained by using Lyapunov stability analysis method.The back EMF is filtered by Kalman filter,and the rotor position is estimated by PLL,which improves the observation accuracy of speed and position.The simulation model of the improved sliding mode observer is built on the Simulink platform,and compared with the switching function and saturation function.Compared with the traditional sliding mode observer,the start-up process of the proposed improved sliding mode observer is more stable,and the high-frequency chattering phenomenon in the steady-state operation is improved.(5)Finally,based on the above research,a new EMA is developed based on STM32F4.The mechanical structure and internal circuit of the EMA are analyzed,and the U-phase driving circuit and current sampling circuit are analyzed in detail.Based on the above EMA,the speed mode and position mode experiments are carried out using PI,ADRC and fuzzy IADRC with friction feedforward compensation.The position tracking results are evaluated by peak to peak and root mean square errors,and the experimental results are analyzed in detail.