This paper aims to investigate the global exponential stability of a class of inertial Cohen-Grossberg neural networks with parameter uncertainties and time-varying delays. By constructing a modified delay-dependent Lyapunov-Krasovskii functional, delay-dependent criteria stated with simple algebraic inequalities are given in order to ensure the global exponential stability for the addressed neural network model. In sharp contrast to the existed reduced order method used to and delay-independent criteria derived for the neural networks with inertial terms, the model proposed and results established of this paper are more general and rigorous. Finally, numerical examples with simulations are presented to illustrate the main results.

本文旨在研究一类具有参数不确定性和时变延迟的惯性 Cohen-Grossberg 神经网络的全局指数稳定性。通过构造一个改进的延迟相关 Lyapunov-Krasovskii 泛函，给出了用简单代数不等式表示的延迟相关标准，以确保所解决的神经网络模型的全局指数稳定性。与已有的用于惯性项神经网络的降阶方法和推导的时延无关准则形成鲜明对比的是，本文提出的模型和建立的结果更具有一般性和严谨性。最后，给出了带有模拟的数值例子来说明主要结果。