Hysteresis Modeling for Magnetic Shape Memory Alloy Actuator via Pi-Sigma Neural Network with Backlash-Like Operator

Wei Pan, Rui Xu, Yewei Yu, Chen Zhang, Miaolei Zhou Department of Control Science and Engineering, Jilin University, Changchun 130022, China

Full Text PDF

Magnetic shape memory alloy materials have a bright prospect in micro-nano intelligent actuators, however, the hysteresis in these materials is prevalent and damages the positioning precision of these actuators. In this paper, a Pi-Sigma neural network model with a modified backlash-like operator is developed to capture the dynamic hysteresis of the magnetic shape memory alloy material actuator. The modified backlash-like operator is designed for the multi-value-mapping and asymmetrical hysteresis of the magnetic shape memory alloy material actuator. The rate-dependent hysteresis of the magnetic shape memory alloy material actuator is described via the Pi-Sigma neural network model. Experimental results show that the effectiveness of the proposed model outperforms the Krasnosel'skii-Pokrovskii model.

DOI:10.12693/APhysPolA.137.634 topics: hysteresis model, magnetic shape memory alloy, backlash-like operator