Research On High Performance Control Method Of Electromechanical Transmission System Based On Backlash Effect Compensation

Backlash is a common nonlinear factors,it will reduce the control performance of the electromechanical system,which mainly expresses in the tracking error of the servo control system tracking the desired motion trajectory,which resulted from control time delay during backlash working stage,and oscillation and noise caused by impact between the gear teeth when changing motion direction.If we do not take effective measures to eliminate or reduce the influence of backlash nonlinearity,the above problems will greatly weaken the dynamic and static performance of the system.Or even worse,it may result in system instability and uncontrollability in serious cases.Therefore,it is of great practical value and broad application prospect to carry out a continuous and in-depth study of the backlash problem.In this paper,I will be devoted to the research of compensation control strategy for the backlash nonlinear system.This paper article first focuses on the introduction of three common models of backlash,analyses the characteristics of each other and makes classification and brief analysis of existing compensation control strategies for backlash nonlinearity.Then I build the mathematical model of the motor servo system based on the deadzone model of backlash.Apply the backstepping method to design the controller.Based on Lyapunov stability theory,design the parameter adaptive law for uncertain torque transfer parameters.Thus,an adaptive controller based on state feedback has been designed,which can realize asymptotic stability control and on-line identification of uncertain parameters of the motor servo system with backlash nonlinearity.On the basis of the classic deadzone model,a simplified deadzone model is built.Design the deadzone model inverse function to approximate and replace the actual deadzone inverse.Apply the RBF neural network observer to compensate the deadzone inverse error online.A continuous nonlinear robust feedback term is designed to suppress the unmodeled dynamics and disturbances of the system.The weight adaptive law of the neural network and the system controller are designed to realize the asymptotic stability control of the system.At last,use Matlab/Simulink software for simulation,and the permanent magnet brushless DC motor servo system experimental platform to carry out the experiment.Compared with PID control,the effectiveness and superiority of the proposed control strategies are verified.