Adaptive decentralised control for large‐scale non‐linear non‐strict‐feedback interconnected systems with time‐varying asymmetric output constraints and dead‐zone inputs. Issue 20 (2nd March 2021)

• Record Type: Journal Article
• Authors: Wang, Xiao‐Mei; Feng, Zhi‐Guang; Zhang, Guang‐Ju; Niu, Ben; Yang, Dong; Hayat, Tasawar; Alsaadi, Fuad E.
• Is Part Of: IET control theory & applications. Volume 14:Issue 20(2020)
• Page Start: 3417
• Page End: 3427
• Publication Date: 2021-03-02
• Subjects: feedback -- time‐varying systems -- stability -- decentralised control -- Lyapunov methods -- adaptive control -- interconnected systems -- closed loop systems -- control nonlinearities -- control system synthesis -- nonlinear control systems -- neurocontrollers; large‐scale nonlinear nonstrict‐feedback interconnected systems -- time‐varying asymmetric output constraints -- dead‐zone inputs -- neural‐network‐based adaptive decentralised controller design issue -- nonstrict‐feedback structure -- arbitrarily small time‐varying range -- predefined time‐varying range; Control theory -- Periodicals; Automatic control -- Periodicals; 629.8312
• DOI: 10.1049/iet-cta.2019.0283
• Languages: English
• ISSNs: 1751-8644

Abstract

Abstract : This study investigatesthe neural‐network (NN)‐based adaptive decentralised controller design issue for a class of large‐scale non‐linear non‐strict‐feedback interconnected systems with time‐varying asymmetric output constraints and dead‐zone inputs. The existences of the non‐strict‐feedback structure, time‐varying asymmetric output constraints, and dead‐zone inputs lead to the construction of the controller of each subsystem is very difficult. By applying the inherent property of Gaussian functions employed in radical basis function NNs, the non‐strict‐feedback structure is skillfully handled, and the adaptive backstepping method is used to construct the desired controllers of all subsystems. Furthermore, Lyapunov stability analysis shows that all the signals of the closed‐loop system are ultimately bounded, and each subsystem output can converge to an arbitrarily small and predefined time‐varying range with the corresponding constraint is always satisfied by employing a barrier Lyapunov function. Finally, simulation results based on a practical example prove the effectiveness of the proposed design strategy.