基于自适应事件触发牵制控制的多时滞随机耦合神经网络簇同步

Cluster synchronization of multi-delayed stochastic coupled neural networks via adaptive event-triggered pinning control

中文关键词  多时滞  事件触发  随机耦合神经网络  簇同步

英文关键词  multi-delay  event-triggering  stochastic coupled neural networks  cluster synchronization

中文摘要

      本文通过自适应事件触发牵制控制策略, 研究了多时滞的随机耦合神经网络在均方意义下以指数速率进 行簇同步的问题. 在耦合神经网络中, 同一簇中的节点只需与相应的孤立节点同步, 而对于不同簇中节点之间的同 步状态没有要求. 首先, 本文提出了一种事件触发牵制控制方法来解决耦合神经网络中节点数量众多、通讯复杂的 问题. 该方法不仅能减少耦合神经网络中控制器的数量, 还可以减少控制信号的传输次数、减轻网络传输压力. 然 后根据M矩阵方法, 建立了随机耦合神经网络均方指数稳定的充分条件. 同时, 利用自适应控制策略, 给出了反馈 增益的更新规律. 最后, 通过一个数值例子验证了所提出的自适应事件触发牵制控制策略的有效性和适用性.

英文摘要

      In this paper, the cluster synchronization of multi-delayed stochastic coupled neural networks at exponential rate in the sense of mean square is studied by the adaptive event-triggered pinning control strategy. In coupled neural networks, nodes in the same cluster only need to synchronize with the corresponding isolated nodes, but there is no requirement for the synchronization state between nodes in different clusters. Firstly, an event-triggered pinning control method is proposed to solve the problems of large number of nodes and complex communication in the coupled neural networks. This method not only reduce the number of controllers in the coupled neural networks, but also reduce the transmission times of control signals and the transmission pressure of the network. Then, according to the M-matrix method, a sufficient condition for the mean square exponential stability of stochastic coupled neural networks is established. At the same time, the update law of feedback gain is given by the adaptive control strategy. Finally, a numerical example is given to verify the effectiveness and applicability of the proposed adaptive event-triggered pinning control strategy.