RTDS Implementation and Induction Motor Drive Performance Comparison with P-I, Sliding Mode and Iterative Learning Controller

Abstract

This paper implements an Iterative Learning Controller (ILC) for speed control of induction motor drive. First, state feedback linearization technique is applied to the induction motor drive for decoupling speed and flux control loop. It uses induction motor model in a stationary (α-β) reference frame with rotor flux and stator current components as state variables, and P-I control of rotor flux and speed. Since the induction motor drive system has model uncertainties, and it is also sensitive to parameter variation, and load disturbances, a robust control strategy based on sliding mode is designed. But, in sliding mode control, torque ripple is significant. To reduce the torque ripple and improve the transient response further, Iterative Learning Controller is implemented. ILC law consists of two parts. First part is a feedback P-I learning law with initial correction term and a forgetting factor. Second part is a Takagi Sugeno (TS) fuzzy logic based P-I learning controller connected in the feedforward path, to improve the dynamic performance. Three control schemes: P-I, sliding mode and ILC are simulated in SIMULINK environment and implemented with real time digital simulator, RT Lab. Results demonstrate that the performance of ILC is better than the other schemes.