Decoupled identification and compensation of nonlinear hysteresis cascading with linear dynamic in a moving magnet voice coil actuator

Abstract

The moving magnet voice coil actuator (MMVCA) is a promising choice for the long stroke nanopositioning stage with the advantage of low moving mass. However, the hysteresis observed in MMVCA limits further improvement on tracking performance. The hysteresis is cascading with the linear dynamic of the positioning stage, which makes common hysteresis identification inapplicable. In this paper, the cause and influence of hysteresis in MMVCA are analyzed, which reveal that the magnetic hysteresis leads to a hysteresis of force and causes motion accuracy to degrade. A modified rate-dependent Prandtl–Ishlinskii (P–I) model is proposed to describe the hysteresis in MMVCA. The decoupled method is implemented to identify the parameters of the linear dynamic model and nonlinear hysteresis model. The experimental results validate the feasibility of the proposed P–I model. Based on the hysteresis compensation, the peak-to-peak tracking errors are reduced by 30% and the root-mean-square (rms) tracking errors are decreased by 41% on average for the trajectories with amplitudes from 1 to 3 mm and frequencies from 1 to 5 Hz.