本文研究了网络化神经网络的稳定性问题.首先,为了利用网络系统的采样特征,定义了一个新的Lyapunov泛函;通过分析网络诱导时延和执行周期之间的关系,采用一个迭代凸组合技术,得到了一个包含较少保守性的稳定性判据.然后,给出一个基于采样数据的神经网络稳定性判据,减少了计算复杂性.最后,通过一个数例,验证了本文方法的有效性和优越性.
This paper investigates the problem of stability of network-based neural networks (NNs). To exploit the sampling characteristic of network systems, we define a new type of Lyapunov functional. By analyzing the relation between the network-induced delay and the executive duration, and employing an iterative convex combination technique, we develop a less conservative stability criterion for network-based NNs. To reduce the computational complexity, we also propose a stability criterion for sampled-data-based NNs. An illustrative example is given to show the effectiveness and the advantages of the proposed method.