Research On Modeling And Robust Adaptive Control Of Magnetic Shape Memory Alloy Actuator

Magnetic Shape Memory Alloy(MSMA)is a new type of intelligent material with advantages such as high frequency response and large stroke,and has enormous application potential in modern medical and high-precision positioning fields.Under the continuous influence of the magnetic field,the inherent magnetostrictive properties of MSMA materials can generate deformation at the micro and nano level,which has been successfully applied to the field of high-precision positioning by MSMA actuators.However,due to the hysteresis nonlinearity,rate dependence,multivalued mapping,and memory characteristics of MSMA actuators,this greatly increases the difficulty of modeling and control for MSMA actuators.This article constructs an accurate comprehensive dynamic model of the MSMA actuator system,and designs an effective control scheme based on the dynamic model to achieve precise trajectory tracking control.The research content of this article is as follows:Firstly,in response to the rate dependent and asymmetric hysteresis nonlinearity of MSMA actuators,this paper proposes a Modified Prandtl Ishlinskii(MPI)model based on the traditional Prandtl Ishlinskii(PI)model.Weighting the weight and falling edge threshold of the traditional PI model’s play operator to improve the accuracy and flexibility of the model,concatenating the improved play operator with the dead zone operator,and introducing asymmetric polynomials and rate dependent terms to enhance the model’s description of asymmetry and rate dependence.The identification of MPI model parameters adopts adaptive genetic algorithm,and the experimental results verify the effectiveness of the constructed MPI model.Based on the MPI model,a comprehensive dynamic model of the MSMA actuator system is constructed through electromechanical coupling modeling.In addition,a step-by-step identification method is adopted,and on the basis of MPI model parameter identification,an adaptive genetic algorithm is used to identify unknown parameters in the dynamic model.The comprehensive dynamic model of the MSMA actuator system proposed in this article provides a foundation for the design of the controller.Then,in response to the problem of precise inverse model construction difficulties and modeling errors in inverse model feedforward control.Taking into account the impact of uncertain disturbances on the system,combined with the comprehensive dynamic model of the designed MSMA actuator system,a non inverse model robust adaptive controller based on the dynamic model is proposed to counteract the impact of hysteresis nonlinearity on the positioning accuracy of the MSMA actuator.The stability of the system is demonstrated using Lyapunov theory.Comparing the proposed control method with neural network control in literature,experimental results show that the proposed robust adaptive control has better control performance.Finally,although the control method proposed in Chapter 3 avoids the construction of hysteresis inverse models,its control accuracy is not completely free from the dependence on modeling accuracy.Considering the difficulty in measuring the internal state of the system in practical applications,this paper proposes a robust adaptive output feedback controller based on neural networks.RBF neural network state observer is used to observe the states that are difficult to measure inside the system.In addition,because the continuous hybrid Differentiator(CHD)has the advantages of avoiding chattering,finite time convergence,and good robustness,this chapter uses CHD to replace the first-order low-pass filter in traditional dynamic surface control,improving the control accuracy of the controller,The stability of the system was ultimately demonstrated using Lyapunov theory.The experimental results show that the robust adaptive output feedback controller proposed in this paper can ensure that the MSMA actuator accurately tracks the expected signal.