在本文中,我们讨论了一类带时间延迟的Cohen-Grossberg神经网络,并研究了这个系统平衡点的全局鲁棒稳定性。利用Lyapunov函数,我们得出了全局鲁棒收敛性的几个充分条件。这些条件以线性矩阵不等式(LMI)的形式表达。因此,从计算的角度出发他们是高效的。另外,这些条件不依赖于时间延迟和神经网络的激发函数。 In this paper, we discuss a class of Cohen-Grossberg neural networks with time delay and study the global robust stability of the system equilibrium point. Using Lyapunov function, we obtain several sufficient conditions for global robust convergence. These conditions are expressed in terms of Linear Matrix Inequalities (LMIs). Therefore, they are efficient from a computational point of view. In addition, these conditions do not depend on the time delay and the excitation function of the neural network.