Stochastic Stability Analysis for Neutral-Type Markov Jump Neural Networks with Additive Time-Varying Delays Via a New Reciprocally Convex Combination Inequality.
Haiyang Zhang,Zhipeng Qiu,Lianglin Xiong,Guanghao Jiang
DOI: https://doi.org/10.1080/00207721.2019.1586005
IF: 2.648
2019-01-01
International Journal of Systems Science
Abstract:This paper investigates the stochastic stability problem for a class of neutral-type Markov jump neural networks with additive time-varying delays. Firstly, to derive a tighter lower bound of the reciprocally convex quadratic terms, a new reciprocally convex combination inequality is established by using parameters transformation approach. Secondly, by fully considering the peculiarity of various time-varying delays and Markov jumping parameters, an eligible stochastic Lyapunov-Krasovskii functional is constructed. Then, by employing the new reciprocally convex combination inequality and other analytical techniques, some novel stability criteria are provided in the forms of linear matrix inequalities. Finally, four illustrated examples are given to verify the effectiveness and feasibility of the proposed methods.