Eccentricity Reduction of Slotless Self- Bearing Motor System Based on Adaptive Neural Network and Sliding Mode Control

Abstract

Abstract This paper presents the effectiveness of a hybrid control method for the slotless self-bearing motor (SSBM), which combined by an improved adaptive radial basic function neural network (ARBFNN) and adaptive improved reaching law (AIRL) sliding-mode control (SMC) with the aim of softening the eccentricity of the rotor of SSBM. The source of eccentricity is considered from the disturbances and uncertainties and the self-vibration of the rotor. First, the mathematical model of the bearing motor system with fully embedded sources of the perturbations is represented. Second, the proposed control algorithms for SSBM system with the proof of stability via the Lyapunov theorem are given. Third, the illustrative examples of simulation and experimental studies are given to shown the effectiveness and power of the proposed control algorithms. The outcomes of the proposed control algorithms are small overshoots, small settling-times, and stable steady-states. The tested disturbances in the simulation study were mostly rejected by an ARBFNN. The performance of proposed controller with ARBFNN in experiment is better the case of controller without ARBFNN.