Modeling rate-dependent hysteresis in piezoelectric actuators using neural networks based on expanded input space method

A neural network-based approach of identification for rate-dependent hysteresis is proposed. In this method, a novel hysteretic operator is proposed to describe the change tendency and extract the dynamic property of rate-dependent hysteresis. Then an expanded input space is constructed to transform the multi-valued mapping into one-to-one mapping so that the neural networks are capable of implementing identification for the rate-dependent hysteresis. Finally, the experimental results are presented to illustrate the potential of the proposed modeling technique.