Department of Mathematics, Anna University-Regional Centre, Coimbatore 641047, India.
Department of Mathematics, Sri Ramakrishna Institute of Technology, Coimbatore 641010, India; Department of Mathematics, Sungkyunkwan University, Suwon 440-746, South Korea.
ISA Trans. 2014 Jul;53(4):1006-14. doi: 10.1016/j.isatra.2014.05.002. Epub 2014 Jun 3.
This paper focuses the issue of robust stochastic stability for a class of uncertain fuzzy Markovian jumping discrete-time neural networks (FMJDNNs) with various activation functions and mixed time delay. By employing the Lyapunov technique and linear matrix inequality (LMI) approach, a new set of delay-dependent sufficient conditions are established for the robust stochastic stability of uncertain FMJDNNs. More precisely, the parameter uncertainties are assumed to be time varying, unknown and norm bounded. The obtained stability conditions are established in terms of LMIs, which can be easily checked by using the efficient MATLAB-LMI toolbox. Finally, numerical examples with simulation result are provided to illustrate the effectiveness and less conservativeness of the obtained results.
本文关注了一类具有各种激活函数和混合时滞的不确定模糊马尔可夫跳跃离散时间神经网络（FMJDNN）的鲁棒随机稳定性问题。通过使用 Lyapunov 技术和线性矩阵不等式（LMI）方法，为具有不确定 FMJDNN 的鲁棒随机稳定性建立了一组新的时滞相关充分条件。更准确地说，假设参数不确定性是时变的、未知的和范数有界的。所得到的稳定性条件是用 LMI 表示的，可以使用高效的 MATLAB-LMI 工具箱很容易地进行检查。最后，通过数值实例和仿真结果说明了所得到的结果的有效性和较小的保守性。
