Adaptive real-time tracking controller for induction motor drives using neural designs

This paper presents a learning architecture for the identification and control of nonlinear induction motor dynamics with unknown parameters. The control and identification parameters are adjusted simultaneously in real-time using the dynamic backpropagation algorithm. Both identification and control are carried out at pre-specified (and possibly different) time intervals, as the system is in operation. The proposed architecture adapts and generalizes its learning to a wide variety of loads, and in addition provides the necessary abstraction when measurements are contaminated with noise. Extensive simulations reveal that neural designs are effective means of system identification and control for time-varying nonlinear systems, in the presence of uncertainty. The difficulties addressed by this article include incomplete system knowledge, nonlinearity, noise and delays.