Hysteresis model identification for piezoelectric positioning stage using improved moth-flame optimization

Abstract

The hysteresis existing in the piezoelectric positioning stage seriously affects the performance of the positioning stage, and even causes the instability of the system. However, due to the complexity of hysteresis, hysteresis modeling and parameter identification are still a challenging task. According to the hysteresis characteristic of the piezoelectric positioning stage, a hysteresis parameter identification method based on the improved moth-flame optimization (IMFO) algorithm is proposed. First, the diversity of IMFO is enhanced by differential evolution (DE) algorithm, then the subpopulation update strategy is used to balance the exploration and exploitation capability of the algorithm, so as to effectively improve the performance of the identification algorithm. Finally, for the identification problem of the enhanced Prandtl-Ishlinskii (PI) model in the piezoelectric positioning stage, the proposed identification algorithm is compared with five stochastic optimization algorithms. The experimental results show that the proposed algorithm has higher accuracy, which also demonstrates the effectiveness of the algorithm.