Research On Modeling And Control Method Of Giant Magnetostrictive Actuator Based On FBG

Giant magnetostrictive material is a kind of new functional material with high strain,high energy conversion rate,low response time,large carrying capacity and simple driving mode.It is widely used in aerospace,ultra precision machining,optical microprocessing,micro-displacement drive and other fields.However,there are nonlinear,hysteresis and the inherent characteristics of the magnetostrictive material,resulting in the input voltage and output displacement of the giant magnetostrictive,which made of giant magnetostrictive material,has serious hysteresis nonlinear,this has certain influence on the positioning accuracy and stability,so it is of great practical significance to study the hysteresis characteristic modeling and compensation control of the giant magnetostrictive actuator.In this paper,the dynamic displacement of the giant magnetostrictive actuator is measured by a grating fiber sensor,and then the hysteresis nonlinear model is established.The effect of the hysteresis nonlinearity of the giant magnetostrictive actuator is eliminated by the inverse compensation control method.The main research contents are as follows:(1)Based on the study of the existing methods of measuring the displacement of the giant magnetostrictive actuator,the measurement principle and characteristics of the fiber grating sensor are analyzed,and an experimental measurement platform based on the fiber grating sensor is built.The displacement signal of the giant magnetostrictive actuator under the 10-100 Hz sinusoidal excitation is analyzed by Fourier transform.According to the characteristics of low noise,the original signal is processed by bartworth filter algorithm to obtain the effective signal.The experimental data and simulation results show that the fiber grating sensor can be used to measure the displacement feasibility of GMA,and provide data support for model parameter identification.(2)The basic characteristics of the giant magnetostrictive material and the structure and working principle of the giant magnetostrictive actuator are studied,and Bouc-Wen model is established for its nonlinear,hysteresis and asymmetry.The traditional two degrees of freedom Bouc-Wen model is analyzed,and point out its disadvantages,on this basis,a six-degree-of-freedom Bouc-Wen model with quadratic term is proposed.In order to identify the improved Bouc-Wen model parameters,the particle optimization algorithm and the differential evolution algorithm are studied,and the error and stability of the two algorithms are compared and analyzed.The experimental results show that the differential evolution algorithm has a stronger stability than the particle optimization algorithm.Finally,the parameters identified by the differential evolution algorithm are brought back into the model,and the improved model is compared with the four-degree-of-freedom Bouc-Wen model.The experimental results show that the improved model can describe the hysteresis nonlinearity of the giant magnetostrictive actuator more accurately at 10-100 Hz drive frequency.(3)The principle of hysteresis inverse compensation of giant magnetostrictive actuator is studied.A feedforward inverse compensation control method based on support vector machine is proposed,and the inverse model of Bouc-Wen model is constructed by using support vector machine.The inverse model is connected with the giant magnetostrictive actuator in series,and the direct compensation control of the hysteresis nonlinearity of the giant magnetostrictive actuator is realized.Aiming at the problem that the stability of direct open loop control is weak and the control precision is not high enough,the PID controller is proposed as the feedback to integrate with the feedforward inverse compensation controller,and the feedback inverse compensation control system is established.The simulation results show that the control system stability and control accuracy are improved,and the positioning error caused by GMA hysteresis nonlinearity can be effectively suppressed.