Neural network-based discrete-time command filtered adaptive position tracking control for induction motors via backstepping

Abstract

Considering the problems of parameter uncertainties and load disturbance appeared in induction motor drive systems, a discrete-time command filtered adaptive position tracking control method based on neural networks is proposed in this paper. First, Euler method is used to describe the discrete-time dynamic mathematical model of induction motors (IMs). Next, the neural networks technique is employed to approximate the unknown nonlinear functions. Furthermore, the “explosion of complexity” problem and noncausal problem, which emerged in traditional backstepping design, are eliminated by command filtered control technique. Simulation results prove that tracking error converges to a small neighborhood of the origin and the effectiveness of the proposed approach is illustrated.