引用本文: | 关焕新,王占山,张化光.不确定双向联想记忆神经网络的稳定性分析[J].控制理论与应用,2008,25(3):421~426.[点击复制] |
GUAN Huan-xin,WANG Zhan-shan,ZHANG Hua-guang.Stability analysis of uncertain bi-directional associative memory neural networks with variable delays[J].Control Theory and Technology,2008,25(3):421~426.[点击复制] |
|
不确定双向联想记忆神经网络的稳定性分析 |
Stability analysis of uncertain bi-directional associative memory neural networks with variable delays |
摘要点击 1617 全文点击 1114 投稿时间:2006-09-07 修订日期:2007-05-09 |
查看全文 查看/发表评论 下载PDF阅读器 |
DOI编号 10.7641/j.issn.1000-8152.2008.3.006 |
2008,25(3):421-426 |
中文关键词 双向联想记忆神经网络 时变时滞 不确定性 鲁棒稳定 线性矩阵不等式 Lyapunov-Krasovskii函数 |
英文关键词 bi-directional associative memory neural networks time varying delays uncertainty robust stability linear matrix inequality (LMI) Lyapunov-Krasovskii functional |
基金项目 国家自然科学基金资助项目(60534010, 60572070, 60774098, 60774093); 辽宁省自然科学基金资助项目(20072025); 东北大学博士后资助项目(20080314). |
|
中文摘要 |
对双向联想记忆神经网络研究了平衡点的鲁棒稳定性. 该网络的参数不确定, 并且有时变时滞. 当神经网络的激励函数满足Lipschitz连续性条件时, 通过选取合适的Lyapunov-Krasovskii函数, 建立了两个全局鲁棒稳定判据. 由于这些判据考虑了神经元激励作用和抑制作用对网络的影响, 他们和时变时滞的数值无关, 并且易于使用内点算法进行检验. 在注释中和已有的结果进行了对比. 两个数值例子展示了所得结果的有效性. |
英文摘要 |
The robust stability of equilibrium point is studied for bi-directional associative memory neural networks with parameter uncertainties and time-varying delays. When the activation function satisfies the condition of Lipschitz continuity, two sufficient conditions are established for the globally robust stability of the equilibrium point by suitably choosing Lyapunov-Krasovskii functional. The obtained results, which take account of the effects of neural inhibitory and excitatory on neural networks, are independent of the sizes of the time-varying delays and are easy to be checked by the interior-point algorithms in MATLAB toolbox. They are compared with prior results in a remark, and are demonstrated by two numerical examples for their effectiveness. |
|
|
|
|
|