Nonlinear modeling of hysteresis in piezoelectric actuators

In this paper, a nonlinear modeling scheme of hysteresis in piezoelectric actuators is presented. In this modeling scheme, the extreme learning machine based model for hysteresis in piezoelectric actuators is proposed. In this method, a modified hysteresis operator extracting the main movement features of the hysteresis is introduced to construct an expanded space. Then, the multi-valued mapping of hysteresis can be transformed into a one-to-one mapping on this expanded input space. Thus, an extreme learning machine (ELM) based neural network model can be obtained to describe the behavior of hysteresis existing in piezoelectric actuators. Finally, the corresponding experimental results on a piezoelectric actuator are demonstrated.