Event-Triggered Synchronization for Neutral-Type Semi-Markovian Neural Networks With Partial Mode-Dependent Time-Varying Delays
Event-Triggered Synchronization for Neutral-Type Semi-Markovian Neural Networks With Partial Mode-Dependent Time-Varying Delays
复制标题
DOI：
10.1109/tnnls.2019.2955287
复制
发表时间：
2019-12
影响因子：
10.4
通讯作者：
Haiyang Zhang;Zhipeng Qiu;Jinde Cao;M. Abdel-Aty;Lianglin Xiong
中科院分区：
文献类型：
--
作者：
Haiyang Zhang;Zhipeng Qiu;Jinde Cao;M. Abdel-Aty;Lianglin Xiong
研究了具有部分模相关加性时变延迟（ATDs）的中立型半马尔可夫跳（SMJ）神经网络的事件触发随机同步问题，其中考虑两个ATDs中的SMJ参数与系统连接权矩阵中的SMJ参数不完全相同。不同于多马尔可夫过程的弱无穷小算子，本文首次提出了双半马尔可夫过程的一个新的弱无穷小算子。为了降低稳定性判据的保守性，利用一种有趣的技术建立了广义的互凸组合不等式。然后，基于符合条件的随机Lyapunov-Krasovski泛函，采用新的RCCI并结合精心设计的事件触发控制方案，导出了所研究系统的三个新的稳定性判据。最后给出了三个数值算例和一个工程实例，验证了所提方法的有效性。
This article studies the event-triggered stochastic synchronization problem for neutral-type semi-Markovian jump (SMJ) neural networks with partial mode-dependent additive time-varying delays (ATDs), where the SMJ parameters in two ATDs are considered to be not completely the same as the one in the connection weight matrices of the systems. Different from the weak infinitesimal operator of multi-Markov processes, a new one for the double semi-Markovian processes (SMPs) is first proposed. To reduce the conservative of the stability criteria, a generalized reciprocally convex combination inequality (RCCI) is established by the virtue of an interesting technique. Then, based on an eligible stochastic Lyapunov–Krasovski functional, three novel stability criteria for the studied systems are derived by employing the new RCCI and combining with a well-designed event-triggered control scheme. Finally, three numerical examples and one practical engineering example are presented to show the validity of our methods.