Robust Adaptive Feedback Linearization Control Using Online Neural-Network Estimators for Uncertain Linear Induction Motor Drive System
Feedback Linearization， Neural- Network， Linear Induction Motor
Control methods for linear drives
Final Paper
Mahmoud F. Elmorshedy / Electrical Power and Machines Engineering Department; Faculty of Engineering; Tanta University
This paper presents a robust adaptive feedback linearization control (RAFLC) using Takagi-Sugeno-Kang (TSK)-type recurrent Petri fuzzy-neural-network (T-RPFNN) for accomplishing superior dynamic performance for the linear induction motor (LIM) drive system. The RAFLC includes a FLC, a T-RPFNN estimator and an adaptive PI controller. The FLC is used to stabilize the LIM drive and the T-RPFNN estimators are utilized to approximate the nonlinear functions of the LIM and the adaptive PI controller is utilized to reduce the chattering in the control inputs. Furthermore, the Lyapunov stability analysis is employed to ensure the RAFLC approach stability. The experimental results endorse the proposed RAFLC robustness even at uncertain dynamics existence and external disturbances.
