Neurocomputing ( IF 6 ) Pub Date : 2021-05-11 , DOI: 10.1016/j.neucom.2021.05.018 Liguang Wan , Zhenxing Liu
This paper formulates new theoretical results concerning the multiple stability and instability for a class of time-varying delayed fractional-order Cohen-Grossberg neural networks (FoCGNNs) with Gaussian activation functions. With the aid of geometrical configurations obtained from the FoCGNNs model and Gaussian functions, the state space are partitioned into subspaces, where k is a nonnegative constant determined by the parameters of FoCGNNs model. By means of the Brouwer’s fixed point theorem as well as the contraction mapping, it is guaranteed that there exists a unique equilibrium point in each subspace. Sufficient conditions are achieved that equilibrium points are locally stable and equilibrium points are unstable. Several examples are rendered to demonstrate the feasible analysis of the theoretical results.
中文翻译:
多 具有高斯激活函数的时变时滞分数阶Cohen-Grossberg神经网络的稳定性和不稳定性
本文针对多重性提出了新的理论结果 一类具有高斯激活函数的时变时滞分数阶Cohen-Grossberg神经网络(FoCGNNs)的稳定性和不稳定性。借助从FoCGNNs模型和高斯函数获得的几何构型,将状态空间划分为子空间,其中k是由FoCGNNs模型的参数确定的非负常数。通过Brouwer的不动点定理和收缩映射,可以确保在每个子空间中都有一个唯一的平衡点。达到了足以满足以下条件的条件 平衡点是局部的 稳定且 平衡点不稳定。给出了几个例子来证明对理论结果的可行性分析。