Sliding Mode Tracking Differentiator And Its Application In Mechatronic Control System

It is an important issue to calculate the derivative information of sensor signal quickly and accurately in the interference environment.However,the simple difference estimation method is easy to amplify the noise in the signal,so that the correct information is submerged in the noise.Although the state observer can effectively suppress noise,it is necessary to establish an accurate state model of the target signal that limits its scope of application and universality.In recent years,due to the strong anti-jamming ability,and do not need to establish the target signal state model.TD(tracking differentiator)is widely studied in the academic field.It has been widely used in electromechanical control system.In this paper,the subject is sliding mode tracking differentiator and its application in electromechanical control system.Based on the analysis of the convergence of TD state variables in the state space,the problems of TD are studied.The theory of sliding mode tracking differentiator is extended to solve the specific problems.The specific work is as follows:Firstly,in order to solve the problem that the state convergence rate of the system is too slow when the traditional sliding mode tracking differentiator is far away from the equilibrium position,this paper proposes a fast response sliding mode tracking differentiator td-e(tracking differentiator enhanced by empirical reaching term).By adding exponential convergence term,td-e can accelerate the convergence speed of the system without sacrificing the filtering performance.In addition,in order to avoid high frequency chattering of td-e in discrete-time realization,the discrete-time algorithm of td-e is derived by implicit Euler method.The numerical simulation results show that the proposed algorithm can achieve accurate convergence without chattering.Finally,an experimental platform of DC motor position control is built to verify the effectiveness of td-e in the closed-loop electromechanical control system.In addition,in view of the trade-off between the tracking performance and the filtering performance of the traditional sliding mode tracking differentiator when the input is not constant,this paper proposes a new differentiator based on the first derivative feedforward.The proposed differentiator is an extension of td-e,which consists of two different sliding mode tracking differentiators.A tracking differentiator is used to obtain the smooth estimation of the first derivative of the input signal.Another tracking differentiator redesigns the sliding surface based on the derivative estimation of the input signal and the input signal to improve the tracking performance.Through the combination of two different tracking differentiators,the proposed new differentiator can better balance the tracking performance and filtering performance.In addition,an implicit Euler method is used to derive the chattering free discrete-time algorithm of the new differentiator,and its effectiveness is verified by numerical simulation.Finally,a simulation example of variable length inverted pendulum control is given to verify the effect of the new differentiator in the closed-loop system.