Fractional-order discontinuous systems with indefinite LKFs: An application to fractional-order neural networks with time delays

Jinde Cao,K. Udhayakumar,Fathalla A. Rihan,R. Rakkiyappan
DOI: https://doi.org/10.1016/j.neunet.2021.10.027
IF: 7.8
2022-01-01
Neural Networks
Abstract:In this article, we discuss bipartite fixed-time synchronization for fractional-order signed neural networks with discontinuous activation patterns. The Filippov multi-map is used to convert the fixed-time stability of the fractional-order general solution into the zero solution of the fractional-order differential inclusions. On the Caputo fractional-order derivative, Lyapunov-Krasovskii functional is proved to possess the indefinite fractional derivatives for fixed-time stability of fragmentary discontinuous systems. Furthermore, the fixed-time stability of the fractional-order discontinuous system is achieved as well as an estimate of the new settling time.. The discontinuous controller is designed for the delayed fractional-order discontinuous signed neural networks with antagonistic interactions and new conditions for permanent fixed-time synchronization of these networks with antagonistic interactions are also provided, as well as the settling time for permanent fixed-time synchronization. Two numerical simulation results are presented to demonstrate the effectiveness of the main results.
computer science, artificial intelligence,neurosciences
What problem does this paper attempt to address?