Neural networks‐based adaptive practical preassigned finite‐time fault tolerant control for nonlinear time‐varying delay systems with full state constraints

• Published in: International journal of robust and nonlinear control Vol. 31; no. 5; pp. 1497 - 1513
• Main Authors: Wang, Xinjun, Niu, Ben, Song, Xinmin, Zhao, Ping, Wang, Zhenhua
• Format: Journal Article
• Language: English
• Published: Bognor Regis Wiley Subscription Services, Inc 25.03.2021
• Subjects: actuator fault; Actuators; Adaptive control; barrier Lyapunov functions; Control systems design; Controllers; Fault tolerance; full state constraints; Functionals; Liapunov functions; Neural networks; Nonlinear control; Nonlinear systems; nonlinear systems with time‐delay; practical preassigned finite‐time control
• ISSN: 1049-8923; 1099-1239
• DOI: 10.1002/rnc.5352

This article concentrates upon an adaptive practical preassigned finite‐time fault‐tolerant control problem for a class of time‐delay nonlinear systems in nonstrict‐feedback form with full state constraints (FSCs) and actuator fault. The completely unknown nonlinear functions exist in the system are identified by the neural networks (NNs). It is challenged to investigate finite‐time fault‐tolerant control problem for nonlinear systems while encountering the time‐delays, actuator faults and FSCs simultaneously, which increases the difficulty of design. The Lyapunov–Krasovskii functionals and the hyperbolic tangent functions are utilized to eliminate the effect of time‐varying delays. The actuator fault considered in this article contains the loss of effectiveness and the bias fault, simultaneously. By combining a modified barrier Lyapunov function with finite‐time performance function, the finite‐time fault‐tolerant controller is designed. It is demonstrated that the proposed adaptive controller guarantees that the system states converge to a preassigned zone at a finite‐time and all the signals of the closed‐loop system remain semiglobally practical finite‐time stable. Numerical examples are offered to illustrate the feasibility of the theoretical result.