Department of Applied Mathematics, University of Waterloo, Waterloo, ON, Canada N2L 3G1.
Neural Netw. 2022 Jan;145:374-385. doi: 10.1016/j.neunet.2021.10.025. Epub 2021 Nov 6.
This paper investigates the synchronization problem of complex-valued neural networks via event-triggered pinning impulsive control (ETPIC). A time-delayed pinning impulsive controller is proposed based on three levels of event-triggered conditions. By employing the Lyapunov functional method and differential inequality technique, sufficient delay-dependent synchronization criteria are derived under the proposed ETPIC scheme. The obtained result shows that synchronization of master and slave complex-valued neural networks can be achieved even if the sizes of delays exceed the length of intervals between any two consecutive impulsive instants determined by Lyapunov-based event-triggered conditions in the proposed control strategy. Moreover, the linear matrix inequality approach is utilized to exclude Zeno behavior. Numerical examples are provided to illustrate the effectiveness of the theoretical results.
本文研究了通过事件触发钉扎脉冲控制（ETPIC）实现复值神经网络同步的问题。提出了一种基于三级事件触发条件的时滞钉扎脉冲控制器。通过李雅普诺夫泛函方法和微分不等式技术，在提出的 ETPIC 方案下推导出了时滞相关的同步判据。所得结果表明，即使主从复值神经网络的延迟大小超过由基于李雅普诺夫的事件触发条件确定的任意两个连续脉冲时刻之间的间隔长度，也可以实现同步。此外，利用线性矩阵不等式方法排除了零行为。数值例子验证了理论结果的有效性。
