School of Automation, Guangdong University of Petrochemical Technology, Maoming 525000, PR China.
Neural Netw. 2020 Oct;130:165-175. doi: 10.1016/j.neunet.2020.07.002. Epub 2020 Jul 10.
In this study, we consider the fixed-time synchronization problem for stochastic memristor-based neural networks (MNNs) via two different controllers. First, a new stochastic differential equation is established using differential inclusions and set-valued maps. Next, two kinds of control protocols are designed, including a nonlinear delayed state feedback control scheme and a novel adaptive control strategy, by which fixed-time synchronization of MNNs can be achieved. Then based on stochastic analysis techniques and a Lyapunov function, some sufficient criteria are obtained to ensure that stochastic MNNs achieve stochastic fixed-time synchronization in probability. In addition, the upper bound of the settling time is estimated. Finally, simulation results are provided to demonstrate the validity of the proposed schemes.
在这项研究中，我们通过两个不同的控制器考虑了基于随机忆阻器的神经网络（MNN）的固定时间同步问题。首先，使用微分包含和集值映射建立了一个新的随机微分方程。接下来，设计了两种控制协议，包括非线性延迟状态反馈控制方案和新颖的自适应控制策略，通过这些协议可以实现 MNN 的固定时间同步。然后，基于随机分析技术和 Lyapunov 函数，得到了一些充分条件，以确保随机 MNN 以概率实现随机固定时间同步。此外，还估计了稳定时间的上界。最后，提供了仿真结果以验证所提出方案的有效性。
