An improved stability criterion for generalized neural networks with additive time-varying delays

This paper deals with the problem of stability analysis of generalized neural networks with time delays. It should be noted that additive time-varying delays are taken in the state of the neural networks. A novel augmented Lyapunov-Krasovskii (L-K) functional which involves more information on the activation function of the neural networks and upper bound of the additive time-varying delays is constructed. By introducing some zero equations and using the reciprocal convex combination technique and Finsler's lemma, an improved delay-dependent stability criterion is derived in terms of linear matrix inequalities (LMIs), which can be efficiently solved via standard numerical software. Finally, three numerical examples are provided to demonstrate the less conservatism and effectiveness of the proposed results.