A Neural-Network-Based Model of Hysteresis in Magnetostrictive Actuators

In this paper, a new neural-network-based hysteresis model is presented. First of all, a variable-power hysteretic operator is proposed via the characteristics of the motion point trajectory of hysteresis for magnetostrictive actuators. Based on the variable-power hysteretic operator, a basic hysteresis model is obtained. And then, a two-dimension input space of neural network is constructed based on the basic model, so that neural networks can be used to identify the mapping between the expanded input space and the output space. Finally, two experiments involved with a magnetostrictive actuator were implemented to validate the neural hysteresis model. The results of the experiments suggest that the proposed approach is effective.