Stability criteria for memristor-based delayed fractional-order Cohen–Grossberg neural networks with uncertainties

Abstract

This paper investigates the Mittag-Leffler (ML) stability analysis problem of memristor-based fractional-order Cohen–Grossberg neural networks (MFCGNNs) with time delays and uncertainties. The voltage type memristor is taken into consideration. The Cohen–Grossberg neural networks (CGNNs) are described by using fractional-order systems (FOS) unified by memristive circuit elements. Sufficient conditions are derived on the basis of the fractional-order (FO) Lyapunov direct approach, differential inclusion theory, and the Filippov solution. Besides that, the conditions are formed in terms of linear matrix inequalities (LMI), which ensure ML stability for MFCGNNs with time delays. Finally, the validity and efficacy of the obtained theoretical results are demonstrated by appropriate numerical example and simulation results.