Delay-dependent H∞ state estimation of neural networks with mixed time-varying delays
Delay-dependent H∞ state estimation of neural networks with mixed time-varying delays
复制标题
DOI：
10.1016/j.neucom.2013.09.020
复制
发表时间：
2014-04
期刊：
影响因子：
6
通讯作者：
S. Lakshmanan;K. Mathiyalagan;Ju H. Park;R. Sakthivel;Fathalla A. Rihan
中科院分区：
文献类型：
--
作者：
S. Lakshmanan;K. Mathiyalagan;Ju H. Park;R. Sakthivel;Fathalla A. Rihan
研究了具有混合时变时滞的神经网络的时滞相关H∞状态估计问题。通过构造一个合适的具有三重积分项的Lyapunov-Krasovskii泛函，利用詹森不等式和线性矩阵不等式（LMI）框架，给出了时滞相关的判据，使得误差系统全局渐近稳定且具有H∞性能.假设激活函数满足扇形非线性。延迟神经网络的增益估计矩阵可以通过求解线性矩阵不等式得到，这可以很容易地使用一些标准的数值软件包。最后给出了一个数值算例，仿真结果表明了所得结果的实用性和有效性。
In this paper, the delay-dependent H∞ state estimation of neural networks with a mixed time-varying delay is considered. By constructing a suitable Lyapunov–Krasovskii functional with triple integral terms and using Jensen inequality and linear matrix inequality (LMI) framework, the delay-dependent criteria are presented so that the error system is globally asymptotically stable with H∞ performance. The activation functions are assumed to satisfy sector-like nonlinearities. The estimator gain matrix for delayed neural networks can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. Finally a numerical example with simulation is presented to demonstrate the usefulness and effectiveness of the obtained results.