Modeling and Minimization of Speed Ripple of a Faulty Induction Motor With Broken Rotor Bars

Abstract

This paper presents a technique of modeling and minimization of speed ripples of a vector-controlled faulty induction motor (IM) with broken rotor bars. First, the performance of the faulty IM is investigated in terms of speed ripples under the open-loop condition. Then, a new IM model is developed, incorporating the speed ripples. Consequently, a new neuro-fuzzy controller (NFC) is proposed to tolerate the fault effect under an indirect field-oriented control scheme. The proposed NFC can compensate the motor rotation imperfections by minimizing the supply frequency-related speed ripples instead of directly working on the low-frequency speed ripples exhibited by a faulty IM. A novel self-tuning algorithm is proposed to adjust the weight of the developed NFC, and its training convergence is investigated by simulation. The effectiveness of the developed NFC is verified by both simulation and experimental tests.