Self-Tuned NFC and Adaptive Torque Hysteresis-Based DTC Scheme for IM Drive

Abstract

This paper presents a simplified self-tuned Sugeno-type neuro-fuzzy controller (NFC)-based direct torque control (DTC) scheme for induction motor (IM) drive. A simple tuning algorithm is developed based on the reference and actual acceleration of the motor. The online tuning strategy adjusts the parameters of the consequent layer of the NFC in order to minimize the square of the error between the actual and the reference acceleration of the motor. In a conventional DTC scheme, the band limits of flux and torque hysteresis controllers remain fixed, and hence, the torque and flux ripples are high. In order to minimize the torque ripples of the IM, the hysteresis band limits are adjusted online for the torque hysteresis controller. Thus, the proposed NFC is used to achieve high dynamic performance of the drive, and the variable hysteresis controller is used to reduce the steady-state torque ripple. The proposed NFC-based DTC scheme is successfully implemented in real time using the DSP board DS1104 for a prototype 1-hp squirrel-cage IM. The performance of the proposed drive is tested at different operating conditions in both simulation and experiment. The proposed NFC-based DTC scheme is found superior to the conventional DTC scheme in both steady-state and transient conditions.