Neural Networks-Based Adaptive Fault-Tolerant Control For Two Classes Of Uncertain Nonlinear Systems

In this paper,the problem of adaptive neural networks fault-tolerant control for two classes of uncertain nonlinear systems is studies.The fault information is obtained indirectly according to the change of system states parameters caused by the faults,and then the system parameters and control inputs are automatically adjusted to ensure the stability of the system.According to the good approximation ability to smooth functions of neural networks,radial basis function neural networks are utilized to approximate unknown nonlinear dynamics functions in the system.Firstly,this paper investigates the adaptive fault-tolerant control in switched stochastic systems with the actuator faults(including loss of effectiveness and bias faults),selecting common Lyapunov functions to analyze the stability of the systems under arbitrary switching signals,and using It(?) differential equation and backstepping technology.The finite-time fault-tolerant control strategy based on adaptive neural networks is proposed.It is proved that the closed-loop system is semi-global practical finite-time stable in probability and has great tracking effect.Secondly,this paper researches fault-tolerant control under the full state constraints in uncertain nonlinear systems,choosing barrier Lyapunov functions to analyze systems stability,the adaptive fault-tolerant control strategy based on neural networks makes tracking error asymptotically converge to zero and the system states are not violated the constraints.Finally,the effectiveness of the proposed methods is further verified by simulation examples.