Adaptive neural network decentralized fault-tolerant control for nonlinear interconnected fractional-order systems

Abstract

This paper studies the neural network (NN) decentralized observer-based fault-tolerant control (FTC) design problem on the nonlinear interconnected fractional-order systems. The considered fractional-order systems have the non strict-feedback form and are subject to immeasurable states and intermittent actuator faults. Neural networks (NN) are employed to identify the unknown dynamics, and a NN decentralized observer is formulated to estimate unknown states. By constructing appropriate Lyapunov functions and using the backstepping dynamic surface control (DSC) technique, a NN observer-based decentralized FTC method is developed. It is proved that the developed NN decentralized FTC scheme can guarantee that the controlled interconnected fractional-order system is stable and the track errors can be made smaller. Finally, a practical simulated example is provided to check the validity of the NN decentralized FTC algorithm.