Combined Convex Technique on Delay-Dependent Stability for Delayed Neural Networks.

Ting Wang,Tao Li,Mingxiang Xue,Shumin Fei

DOI: https://doi.org/10.1109/tnnls.2013.2256796

IF: 1.4

2012-01-01

Discrete Dynamics in Nature and Society

Abstract:Together with the Lyapunov-Krasovskii functional approach and an improved delay-partitioning idea, one novel sufficient condition is derived to guarantee a class of delayed neural networks to be asymptotically stable in the mean-square sense, in which the probabilistic variable delay and both of delay variation limits can be measured. Through combining the reciprocal convex technique and convex technique one, the criterion is presented via LMIs and its solvability heavily depends on the sizes of both time-delay range and its variations, which can become much less conservative than those present ones by thinning the delay intervals. Finally, it can be demonstrated by four numerical examples that our idea reduces the conservatism more effectively than some earlier reported ones.