时变时滞神经网络系统的时滞依赖的鲁棒稳定性准则  

Delay-dependent Robust Stability Criterion of Neural Networks with Time-varying Delays

作　　者：苏卫卫[1] 陈一鸣[1] 张拥萍[1] 崔玉环[1] 李霞[2] 

机构地区：[1]燕山大学理学院,河北秦皇岛066004 [2]河北科技师范学院,河北秦皇岛066004

出　　处：《计算技术与自动化》2007年第2期1-4,共4页Computing Technology and Automation

基　　金：河北省自然科学基金项目(E2007000381)

摘　　要：基于Lyapunov稳定性理论和线性矩阵不等式技术,针对一类带有范数有界不确定性的时变时滞神经网络系统,给出时滞依赖的鲁棒稳定性准则。系统稳定的充分条件是在激励函数满足一类更为通用的条件下得到的,即激励函数不必是单调可微的,并且消除对时变时滞导数的限制。所给的准则可用Mat-lab中的线性矩阵不等式控制工具箱进行验证。仿真结果进一步证明结论的有效性。Based on Lyapunov stability theory and linear matrix inequality, delay- dependent robust stability is derived for a class of neural networks with time - varying delays and norm - bounded uncertainties. The result is obtained under mild conditions, assuming neither differentiability nor monotonicity for activation function. And the restriction of the derivative of the time - varying delay is removed. The proposed stability criterion can be checked by the LMI Control Toolbox in Matlab. Finally, the effectiveness of the result is demonstrated by a example.

关 键 词：时滞依赖鲁棒稳定性 时变时滞 范数有界不确定性 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]