Research On Adaptive Pseudo Inverse Control Strategies For Several Kinds Of Hysteresis Nonlinear Systems

Smart material actuators include piezoelectric ceramics,giant magnetostriction and shape memory alloy,etc.,which have the advantages of fast response speed,high working frequency,strong reliability and so on.It overcomes the shortcomings of such as slow response speed and poor reliability of hydraulic and mechanical traditional actuators,which makes it widely used in ultra-high precision machining,precision instrument manufacturing,micro/nano-manipulator and other fields.However,the hysteresis nonlinearity of smart materials has the characteristics of multi value,strong nonlinearity and memory,which seriously affects the control performance of the controller.Therefore,it is of great theoretical and practical value to study the modeling,identification and control of hysteresis.In the field of nonlinear control,the introduction of low-pass filter in dynamic surface solves the complexity explosion problem caused by the repeated differentiation of some nonlinear signals in traditional backstepping method,and reduces the complexity of control law design.These research results lay a good foundation for dealing with hysteresis nonlinear problems.In this paper,an adaptive dynamic surface pseudo inverse control strategy is proposed for several kinds of hysteretic nonlinear systems,the effectiveness of the proposed control scheme is verified by MATLAB simulation and hardware in the loop simulation platform.The main research contents of this paper include:(1)An adaptive dynamic surface pseudo inverse control method is proposed for a class of asymmetrically saturated Prandtl ishilinskii(PI)hysteretic nonlinear systems to compensate for the influence of hysteretic nonlinearity.Among them,"pseudo inversion" means to develop an on-line calculation mechanism of approximate control signal by optimizing the designed temporary control signal containing real control signal.The specific contributions are as follows:due to the complexity of the hysteresis model itself,it is very difficult to construct the real saturated hysteresis inverse model,which is the first time to use the hysteresis pseudo inverse compensator to compensate the asymmetric saturated hysteresis;when designing the saturated hysteresis pseudo inverse compensator is not necessary to construct the explicit hysteresis inverse and its corresponding unknown parameters when dealing with the saturated hysteresis;The adaptive dynamic surface control is applied to the error correction The combination of transfer function overcomes the problem of "complexity explosion" in traditional backstepping method and achieves the predetermined performance index.Stability analysis and experimental results on hardware in the loop simulation platform show the effectiveness of the proposed adaptive pseudo inverse control scheme.(2)For a class of nonlinear systems with PI hysteresis input and time-delay state,an adaptive neural network control method is proposed to solve the unknown time-delay and time-delay nonlinear problems in the state feedback system.The main contributions are as follows: The density function of the unknown PI hysteresis model is estimated on-line in real time without establishing an accurate inverse model;The unknown nonlinear function in the system is approximated by radial basis function neural network,and the real control signal is coupled to the temporary control signal,the optimal control signal is obtained by hidden inverse method to compensate the hysteresis nonlinearity in the system,which overcomes the difficulty of constructing inverse model.The stability analysis shows that the system is semi-globally uniform ultimately bounded.The simulation results verify the effectiveness of the method.