具有连续分布时变时滞神经网络时滞相关稳定性判据  

Delay-dependent stability criteria for continuous neural networks with time-varying delay

作　　者：刘燕 于传 黄永明[3] 熊晶晶 刘静 LIU Yan;YU Chuan;HUANG Yongming;XIONG JingJing;Liu Jing(Industrial Center,Nanjing Institute of Technology,Nanjing 211167,China;Training Center,State Grid Anhui Electric Power Company Limited,Hefei 230022,China;School of Automation,Southeast University,Nanjing 210096,China)

机构地区：[1]南京工程学院工业中心,南京211167 [2]国网安徽省电力有限公司培训中心,合肥230022 [3]东南大学自动化学院,南京210096

出　　处：《扬州大学学报（自然科学版）》2019年第1期17-22,共6页Journal of Yangzhou University：Natural Science Edition

基　　金：安徽省自然科学研究重点资助项目(KJ2017A740);国网安徽省电力有限公司培训中心群众性科技创新资助项目(2017QC06)

摘　　要：为研究具有连续分布时变时滞神经网络的全局稳定性条件,利用增广型Lyapunov-Krasovskii泛函(LKF)和运用多种积分不等式缩放技巧,推导了两种保守性相对较小的时滞相关稳定性判据.为进一步降低稳定性判据的保守性,通过改进增广型LKF,结合神经元激活函数的约束条件,得到了基于线性矩阵不等式形式的神经网络时滞相关渐近稳定性条件.结果表明,新的LKF方法具有更好的效果,且稳定性判据的运算负担更低,算例证实了该方法的有效性.This paper investigates the delay-dependent stability for continuous neural networks with time-varying delay.Two novel and less conservative stability criteria are derived by employing Lyapunov-Krasovskii functional(LKF)and several suitable integral inequalities.An augmented LKF which considers more information of the slope of neuron activation functions is developed for further reducing the conservatism of stability criteria.Moreover,it is found that the introduced LKF approach leads to better results and the derived stability criteria have lower computational burden.Finally,a numerical example is given to demonstrate the effectiveness of the proposed method.

关 键 词：神经网络 线性矩阵不等式 LYAPUNOV-KRASOVSKII泛函 时滞相关稳定性 时变时滞 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O175.13[自动化与计算机技术—控制科学与工程]