Modeling And Internal Model Control Of Piezoceramic Actuator Based On Generalized Bouc-Wen Model

Piezoceramic actuator has the advantages of better piezoelectric effect,fast response and high resolution.It plays an increasingly important role in the precision positioning field of optics,electronics and micro electro mechanical system.However,the defects such as hysteresis and creep weaken the control precision of the system,even make the system lose stability,which has caused serious impact on its application in the field of engineering.The hysteresis nonlinearity is the main factor on the positioning accuracy,while in practical applications the hysteresis nonlinearity of the piezoceramic actuator shows asymmetry and rate-dependent characteristics.Therefore,taking the micro positioning system experimental platform of piezoceramic actuator as the research object,the following aspects are studied in the paper:1、Aimed at the asymmetric hysteresis,a generalized Bouc-Wen model is established which has a higher flexibility in describing asymmetric characteristic.And the model parameters are identified based on the differential evolution algorithm.Compared with the classical Bouc-Wen model,the generalized Bouc-Wen model is more suitable for the modeling of actual piezoelectric actuators.2、Based on the established generalized Bouc-Wen model,constructing its inverse model and designing a filter which can enhance the robustness of the system,then proposing an internal model control method.Finally,the experiment is carried on the piezoceramic actuator.It is verified that the proposed internal model control method can eliminate the influence of hysteresis on actuator positioning effectively.3、Aimed at the rate-dependent dynamic hysteresis,a Hammerstein model is proposed.The static nonlinear function and the linear dynamic subsystem are described by generated Bouc-Wen model and autoregressive exogenous model,respectively.And the generalized Bouc-Wen model parameters are identified by Differential Evolution(DE)algorithm and the linear dynamic model parameters are identified by recursive least squares(RLS)algorithm.Finally,the accuracy of Hammerstein model in different single frequencies and compound frequencies is validated.4、Based on the established rate-dependent Hammerstein model,designed three dynamic tracking control schemes:(1)a compound control scheme with a feed-forward hysteresis compensation and a feed-back PID controller;(2)internal model control based on inverse Hammerstein model;(3)internal model control based on hysteresis linearization.Finally,the tracking control experiment with different single frequencies and compound frequencies reference signals is carried out on the dSPACE platform,and the effectiveness of the proposed control schemes are verified.