Study On Hysteresis Nonlinear Modeling And Control Of Piezoelectric Ultra Precision Positioning Table

Micro nano positioning technology has been developed into one of the key technologies in the field of high precision positioning,and the selection of core actuator materials has a very significant impact on positioning accuracy.A piezoelectric ceramic material is an intelligent material that can convert electrical energy into mechanical energy,and it has the advantages of fast response,high resolution,low price,light weight,has been focused by scholars in this field,and has been widely used in high precision positioning,the piezoelectric super precision positioning table developed from piezoelectric ceramic material is a hot and difficult point in recent years.However,due to the hysteresis nonlinearity and creep characteristics of piezoelectric ceramics,the factors that affect the positioning accuracy seriously hamper the development and application of the piezoelectric ultra precision positioning table.To solve this problem,it is the key and difficult point for every researcher to solve this problem.1)In order to solve the rate correlation characteristics of the piezoelectric ultra precision positioning table and improve the precision of the hysteresis modeling,this paper presents the establishment rate dependent hysteresis model.Because the Backlash-Like hysteresis model described in this paper can not describe the rate dependence of the piezoelectric ultra precision positioning table,so the Backlash-Like model and the ARX autoregressive model are connected in series to get the Hammerstein-like rate dependent hysteresis model.Considering the dynamic characteristics of partial piezoelectric ultra precision positioning platform,a Hammerstein-like like model(also called dynamic hysteresis model)is proposed.The hysteresis nonlinearity of the model is still Backlash-Like model,but the linear link becomes a spring damping two order system.Finally,the experimental results show that the two modeling methods respectively solve the rate dependent hysteresis characteristics and dynamic characteristics of the piezoelectric ultra precisionpositioning table.2)For linear control,this paper proposes three control strategies: first,Hammerstein-like inverse model feed-forward control strategy,the control strategy can effectively weaken the hysteresis,the hysteresis loop has been significantly weakened,and improve the control accuracy.The second is to combine Hammerstein-like inverse model feed-forward compensation control and PID feedback control to get a compound PID control,it further improves the control accuracy and anti-interference ability.However,because traditional PID control parameters need to be tried together,researchers need enough experience and are not always able to achieve optimal control.Third control strategies are proposed,the third is the combination of the Hammerstein-like inverse model feed-forward compensation control and the BP neural network PID feedback control,the control effect is better than the first and second strategies.The root mean square error ERMSE and relative error ERE are relatively smaller,and it can be adjusted online in real time,and basically achieve the optimal control.