Synchronization of memristive neural networks with mixed delays via quantized intermittent control
Synchronization of memristive neural networks with mixed delays via quantized intermittent control
复制标题
通过量化间歇控制实现具有混合延迟的忆阻神经网络的同步
DOI：
10.1016/j.amc.2018.08.009
复制
发表时间：
2018-12-15
影响因子：
4
通讯作者：
Cao, Jinde
中科院分区：
文献类型：
--
作者：
Feng, Yuming;Yang, Xinsong;Cao, Jinde
众所周知，如何处理时滞的影响以及如何确定控制宽度和休息宽度是间歇控制的主要难点。本文研究了具有有界时变离散延迟和无界分布延迟（混合延迟）的驱动-响应记忆神经网络（MNNs）的渐近同步问题，扩展了现有的间歇控制技术，揭示了控制宽度和休息宽度之间的新关系。为了节约信道资源和控制成本，减少传输信息量和信道阻塞，设计了量化间歇控制（QIC）。基于加权二重积分不等式，设计了具有负项的Lyapunov-Krasovskii泛函，大大降低了所得结果的保守性。用线性矩阵不等式（lmi）给出了保证渐近同步的充分条件。控制增益也可以通过求解lmi来设计。结果表明，QIC既不是周期性的，也不是控制宽度与静止宽度成比例的。此外，还明确给出了控制宽度、休息宽度和收敛速度之间的关系。最后，通过数值仿真验证了理论分析的有效性。(C) 2018爱思唯尔公司版权所有。
It is well known that how to deal with the effect of time delay and how to determine the control and rest widths are the main difficulties for intermittent control. This paper considers asymptotic synchronization of drive-response memristive neural networks (MNNs) with bounded time-varying discrete delay and unbounded distributed delay (mixed delays), which extends existing intermittent control techniques and reveals new relationship between control width and rest width. A quantized intermittent control (QIC) is designed to save both channel resources and control cost and reduce both the amount of transmitted information and channel blocking. Based on weighted double-integral inequalities, novel Lyapunov-Krasovskii functionals with negative terms are designed, which reduce the conservativeness of obtained results greatly. Sufficient conditions in terms of linear matrix inequalities (LMIs) are obtained to ensure the asymptotic synchronization. The control gains can also be designed by solving the LMIs. It is shown that the QIC can be neither periodic nor proportional between control width and rest width. Moreover, the relationships between control width, rest width, and convergence rate are explicitly given. Finally, numerical simulations are provided to illustrate the effectiveness of the theoretical analysis. (C) 2018 Elsevier Inc. All rights reserved.