Modeling And Control Of Piezoelectric Ceramic Actuators Based On Gaussian Process Regression

Piezoelectric ceramic actuators have the characteristics of low heat,high precision,large driving force,low noise and fast strain velocity,etc.,and have been widely used in the fields of ultra-precision manufacturing,micro-electromechanical system,active control and vibration and noise reduction.However,the nonlinear hysteresis reduces the control precision of the system to some extent,and may cause system disorder,so it is difficult to achieve high precision control of the system.In practical application,the piezoelectric actuators still have rate-dependent dynamic characteristics,which brings great difficulties to the traditional control.Therefore,it is of great theoretical significance and application value to study the high precision modeling and control methods of piezoelectric ceramic actuators.In this paper,the modeling and tracking control of piezoelectric ceramic actuators are studied based on Gaussian process regression model.The main research contents are as follows:1?A nonlinear hysteresis model of piezoelectric ceramic actuators based on Gaussian process regression is proposed.Particle swarm optimization(PSO)algorithm is used to identify the hyperparameters of the model,which solves the dependence between local optimum and initial value when using traditional conjugate gradient algorithm.The simulation results of the system under1?100Hz excitation show that the Gaussian process regression model can well describe the hysteretic nonlinear characteristics of the piezoelectric ceramic actuator,and the maximum root mean square error of the model is 0.5329?m,and the relative errors are all less than 4%.Compared with the traditional Bouc-Wen model,the model error of Gaussian process regression model is reduced by about 15% ? 72%.2?According to the established Gaussian process regression model,a PID compound tracking control method based on feedforward and inverse compensation is studied.The Gaussian process regression inverse model of the system is built and used as a feedforward hysteresis compensator for hysteresis nonlinear compensation.At the same time,considering the influence of uncertain factors such as noise,the introduction of PID feedback controller can achieve high precision tracking control of the system.The results show that in the frequency range of 1?100Hz,the feedforward and feedback composite control based on Gaussian process regression model can track the input signals well,so as to realize the compensation of the hysteresis nonlinear characteristics of the piezoelectric ceramic actuator,and achieve better control accuracy.3?Based on the established Gaussian process regression inverse model,a new internal model control strategy is proposed,and a filter is set to improve the robustness of the system.The simulation results show that the performance of internal model control is better than that of PID composite control.Moreover,compared with the PID controller,the internal model controller only needs to adjust one filter parameter during parameter selection,so the design is more convenient.