Neural network intelligent reconf igurable control for nonlinear system with uncertainty

A new type of robust model-following reconfigurable control strategy based on the neural network compensation is presented for a class of nonlinear systems with uncertainties. By using the neural network online compensator, this method can eliminate the effect of unmodeled dynamics caused by faults, and the system output is able to accurately track the output of an ideal model even when there exist uncertainties. Stability of the closed-loop system is rigorously established under certain assumptions . Both the theoretical analysis and the computer simulation reveal that the presented scheme is effective and the closed - loop system has a good reconfiguration performance.