Neural Networks-Based Adaptive Finite-Time Fault-Tolerant Control for a Class of Strict-Feedback Switched Nonlinear Systems

• Published in: IEEE transactions on cybernetics Vol. 49; no. 7; pp. 2536 - 2545
• Main Authors: Liu, Lei, Liu, Yan-Jun, Tong, Shaocheng
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.07.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Actuators; Adaptive control; Adaptive systems; Basis functions; Computer simulation; Control systems design; Dynamic stability; Fault tolerance; Fault tolerant control (FTC); finite-time stability; Neural networks; neural networks (NNs); Nonlinear systems; Radial basis function; Stability analysis; switched nonlinear systems; Switched systems; Switches; Switching theory; System effectiveness; Uncertainty
• ISSN: 2168-2267; 2168-2275
• DOI: 10.1109/TCYB.2018.2828308

This paper concentrates upon the problem of finite-time fault-tolerant control for a class of switched nonlinear systems in lower-triangular form under arbitrary switching signals. Both loss of effectiveness and bias fault in actuator are taken into account. The method developed extends the traditional finite-time convergence from nonswitched lower-triangular nonlinear systems to switched version by designing appropriate controller and adaptive laws. In contrast to the previous results, it is the first time to handle the fault tolerant problem for switched system while the finite-time stability is also necessary. Meanwhile, there exist unknown internal dynamics in the switched system, which are identified by the radial basis function neural networks. It is proved that under the presented control strategy, the system output tracks the reference signal in the sense of finite-time stability. Finally, an illustrative simulation on a resistor-capacitor-inductor circuit is proposed to further demonstrate the effectiveness of the theoretical result.