论文部分内容阅读
反馈型神经网络由于具有极为丰富的动力学行为和整体计算能力（如优化、联想、振荡和混饨）而倍受关注，近几年的研究表明，当网络的时延足够小时，具有延时的对称Hopfield型神经网络和无时延情况一样也是全局稳定的．本文通过构造适当Lyapunov泛函的方法，对一类具有时延的反馈型神经网络平衡点的渐近稳定性进行了分析，得到了平衡点渐近稳定的充分条件：要检验一个有时间延迟的反馈型神经网络的稳定性，只要测试一个特定矩阵的定性性质或一个特定不等式即可．最后我们也提供了一种估计网络渐近稳定平衡点吸引域的方法．
Feedback neural network has attracted much attention due to its extremely rich dynamic behavior and overall computational ability (such as optimization, association, oscillation and chaos). In recent years, research shows that when the network delay is small enough, The symmetric Hopfield neural network is globally stable as well as without delay. In this paper, we construct the appropriate Lyapunov functional approach to analyze the asymptotic stability of a class of feedback neural networks with delay and obtain the sufficient condition for the asymptotic stability of the equilibrium point. To test a time delay Feedback neural network stability, as long as the test of a particular matrix of qualitative properties or a specific inequality can be. Finally, we also provide a method to estimate the asymptotic stable equilibrium point attracting domain.