基于马氏切换随机变时滞神经网络的稳定性研究

Study on stability of stochastic neural networks with time-varying delays and Markovian switching

为了研究具有马尔可夫切换的随机变时滞神经网络系统的零解稳定性,使用1种有别于线性矩阵不等式(linear matrix inequality,LMI)的方法,即M矩阵方法,讨论该系统的均方指数稳定性。在此基础上,利用泛函微分方程理论获得时滞依赖的稳定性判据,并通过一个数值例子验证所得结论的正确性和有效性。

To study the stability ot trivial solution of stocnasuc neural networks with time-varying delays and Markovian, a meth-od called M-matrix mathod, different to the linear matrix inequality （LMI） method, was used to discuss mean square exponential stability of the above system. Meanwhile, the delay-dependent stability criteria which were captured by using the theory of the functional differential equations. A numerical example was provided to examine the correctness and effectiveness of the theoretic results.