Research on data-driven modeling method of giant magnetostrictive actuator based on LS-SVM
DUAN Lijun1， TIAN Hao2， HAN Ping3
1.School of Computer， Hubei University of Education， Wuhan 430205， China；2.School of Information and Communication Engineering， Hubei University of Economics， Wuhan 430205， China；3.School of Information Engineering， Wuhan University of Technology， Wuhan 430070， China
Abstract：Aiming at the strong nonlinear characteristics of Giant Magnetostrictive Material (GMM)， a new experimental system of Giant Magnetostrictive Actuator (GMA) and its data-driven modeling method are proposed. The measured data in the experiment were taken from grating sensors， and the nonlinear modeling of GMA was realized based on the Least Squares Support Vector Machine (LS-SVM) using the data-driven principle. The performance of the model is evaluated experimentally， the dynamic characteristics of the GMM rod are predicted， and the influence of driving voltage on the output characteristics is also discussed. The experimental results show that the model can predict the braking output of GMA well， and the prediction error is within 0.05%.
2019, Vol. 53 
