切换时滞神经网络的动力学行为

Dynamical Behavior of Switched Delay Neural Networks

文章研究时滞切换神经网络的动力学行为。运用非负矩阵谱半径及M-矩阵的性质,建立了一个新的切换不等式。在此不等式的基础上,研究了中立型切换时滞神经网络的指数稳定性,得到了其稳定性的充分判据,去掉了时滞函数的可微性及其导数的有界性,从而改进了一些现有文献的结果,并通过一个实例论证了所获结果的有效性。

This paper studies the dynamical behavior of the switched neural networks with time-delays.By using the properties of the non-negative matrix spectral radius and M-matrix,a new switching inequality is established. Based on the new inequality,this paper investigates the exponential stability of neutral switched neural networks with delays,and obtains the sufficient conditions for ensuring the exponential stability of the considered system. The results is that it no longer requires the differentiability of the delay function and boundedness of its derivative,thus improving the results of some of the existing literature. The validity of the obtained results is demonstrated by an example.