Adaptive Neural Fractional-Order Pseudo-Inverse Control for a Giant Magnetostrictive Actuator

In this paper, an adaptive fractional-order pseudo-inverse control strategy utilizing radial basis function neural networks (RBFNNs) is proposed for a giant magnetostrictive actuator (GMA). Firstly, the hysteresis nonlinearity inherent in GMA is characterized by using the Prandtl-Ishlinskii (PI) model and addressed through a pseudo-inverse control approach, circumventing the need for direct inverse model construction. Secondly, a fractional-order model is derived from the open-loop experimental data, capturing the essential dynamics of the GMA. Thirdly, an adaptive fractional-order control technique leveraging RBFNNs is developed to enhance the tracking performance. Finally, the effectiveness of the proposed control strategy is validated through simulations.