Finite-Time Synchronization of Fractional-Order Quaternion-Valued Neural Networks under Aperiodically Intermittent Control: A Non-Separation Method

Abstract

This article mainly focuses on investigating finite-time synchronization(FTS) of fractional-order quaternionvalued neural networks(FOQVNNs) with time-varying delay via aperiodically intermittent control. Firstly, a new group of fractional differential inequalities during aperiodically intermittent times is proposed, which provides a formula for calculating stable time \(\tilde{t}\) related to a parameter \(\varpi\) about aperiodically intermittent control strategy. It can make the stable time \(\tilde{t}\) even shorter than that of many existing literature results. Secondly, the theoretical analysis of the entire paper is conducted in a nonseparable method, and a quaternion-valued aperiodically intermittent controller based on 1-norm is directly designed to ensure the synchronization of FOQVNNs. By constructing a suitable Lyapunov function, the FTS criterion is derived. Finally, the validity of the proposed theoretical results is confirmed by numerical simulations.