Research On Control Method Of Piezoelectric Micro-Positioning Platform Based On Bouc-Wen Dynamic Model

As smart materials,piezoelectric ceramic materials are widely used in precision positioning and other fields.The piezoelectric ceramic actuator with piezoelectric material as the core material has the advantages of fast frequency response,good dynamic performance and high resolution.It is widely used in the field of micro/nano-level positioning technology.However,the inherent hysteresis nonlinearity of piezoelectric ceramic materials greatly restricts the positioning accuracy and even cause the system to oscillate.Therefore,reducing or even eliminating the effect of hysteresis nonlinearity is a problem that must be solved.This paper takes the piezoelectric micro-positioning platform as the research object,establishes an accurate hysteresis model to describe its hysteresis nonlinearity.By virtues of the controller design,the goal of reducing or even eliminating hysteresis nonlinearity is finally achieved,and the micron-level positioning control of the piezoelectric micro-positioning platform is finally realized.The main research contents of this paper are as follows:Establish a dynamic model that can accurately describe the hysteresis nonlinearity of the piezoelectric micro-positioning platform.This paper establishes a Bouc-Wen dynamic model,which is composed of two parts in series.One part is an asymmetric static Bouc-Wen model whose parameters are identified by the brainstorming optimization algorithm,The other part is a cerebellar model neural network that can describe dynamic characteristics,and its weights are adjusted online utilizing ? learning rules.The validity of the proposed dynamic model is verified by implementing experiments on the piezoelectric micro-positioning platform.Establish the controller based on the established hysteresis model.First,the asymmetric static Bouc-Wen inverse model is established by the direct method,and the inverse model is directly connected in series with the piezoelectric micro-positioning platform to realize the inverse compensation control.Under the excitation of sinusoidal signals with different frequencies,experiments have proved that the inverse compensation feedforward controller can effectively eliminate the effect of hysteresis nonlinearity under low frequency conditions.In order to improve the control accuracy under high frequency conditions,the recursive cerebellar model neural network controller is further designed as a composite controller.The weights of the recursive cerebellar model neural network are adjusted online by the gradient descent method,and the convergence of the weight adjustment algorithm is proved by the Lyapunov theory.Experiments have proved that the composite controller can effectively eliminate the effect of hysteresis nonlinearity under different frequencies and waveform driven signals.Establish a model-free discrete terminal sliding mode controller based on data-driven theory.In order to further improve the control accuracy and eliminate hysteresis nonlinearity,the compact format dynamic linearization method is used to describe the system firstly,which eliminates the influence of unmodeled dynamics on the control accuracy.Then the single step delay method is used to estimate the external disturbance of the system.Finally,the discrete terminal sliding mode controller is designed to eliminate the hysteresis nonlinearity and external disturbance of the system.Experimental results have proved that the constructed controller can better eliminate the effect of hysteresis nonlinearity on the positioning accuracy of the piezoelectric micro-positioning platform under different frequencies and waveform driven signals.