Induction motor identification using dynamic two-time scales neural networks with sliding mode learning
Zhijun Fu, Wenfang Xie, Weidong Luo · 2012
This paper presents a novel identification method of induction motor via Dynamic Neural Networks with two-time scales using sliding mode learning. Due to the fast adaptation and superb learning capability, Dynamic Neural Networks with two-time scales using sliding mode learning are used to identify the induction motor including the aspects of fast and slow phenomenon. The sliding mode technique and singularly perturbed theories are used to develop the on-line update laws for dynamic neural networks weights. The global convergence of the identification error to zero is analyzed by means of the Lyapunov function. Simulation results are presented confirming the validity of the above approach.