Closed-loop control of nonlinear neural networks: The estimate of control time and energy cost. (September 2019)

• Record Type: Journal Article
• Authors: Chen, Chongyang; Zhu, Song; Wei, Yongchang
• Is Part Of: Neural networks. Volume 117(2019)
• Page Start: 145
• Page End: 151
• Publication Date: 2019-09
• Subjects: Closed-loop control -- Nonlinear neural networks -- Finite-time -- Energy cost; Neural computers -- Periodicals; Neural networks (Computer science) -- Periodicals; Neural networks (Neurobiology) -- Periodicals; Nervous System -- Periodicals; Ordinateurs neuronaux -- Périodiques; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux neuronaux (Neurobiologie) -- Périodiques; Neural computers; Neural networks (Computer science); Neural networks (Neurobiology); Periodicals; 006.32
• DOI: 10.1016/j.neunet.2019.05.016
• Languages: English
• ISSNs: 0893-6080

Abstract

Abstract: This paper concentrates on an estimate of the upper bounds for control time and energy cost of a class of nonlinear neural networks (NNs). By constructing the appropriate closed-loop controller u S and utilizing the inequality technique, sufficient conditions are proposed to guarantee achieving control target in finite time of the considered systems. Then, the estimate of the upper bounds for the control energy cost of the designed controller u S is proposed. Our results provide a new controller which can ensure the realization of finite time control and energy consumption control for a class of nonlinear NNs. Meanwhile, the obtained results contribute to qualitative analysis of some nonlinear systems. Finally, numerical examples are presented to demonstrate the effectiveness of our theoretical results.