High-Performance Control for a Bearingless Permanent-Magnet Synchronous Motor Using Neural Network Inverse Scheme Plus Internal Model Controllers

Abstract

This paper proposes a novel decoupling scheme for a bearingless permanent-magnet synchronous motor (BPMSM) to achieve fast-response and high precision performances and to guarantee the system robustness to the external disturbance and parameter uncertainty. The proposed control scheme incorporates the neural network inverse (NNI) method and 2-degree-of-freedom (DOF) internal model controllers. By introducing the NNI systems into the original BPMSM system, a decoupled pseudo-linear system can be constituted. Additionally, based on the characteristics of the pseudo-linear system, the 2-DOF internal model control theory is utilized to design extra controllers to improve the robustness of the whole system. Consequently, the proposed control scheme can effectively improve the static and dynamic performances of the BPMSM system, as well as adjust the tracking and disturbance rejection performances independently. The effectiveness of the proposed scheme has been verified by both simulation and experimental results.