Neural Networks ( IF 7.8 ) Pub Date : 2020-07-21 , DOI: 10.1016/j.neunet.2020.07.019 Xiao-Zhen Liu 1 , Kai-Ning Wu 1 , Weihai Zhang 2
Cohen–Grossberg neural networks (CGNNs) play an important role in many applications and the stabilization of this system has been well studied. This study considers the exponential stabilization for stochastic reaction–diffusion Cohen–Grossberg neural networks (SRDCGNNs) by means of an aperiodically intermittent boundary control. Both SRDCGNNs without and with time-delays are discussed. By employing the spatial integral functional method and Poincare’s inequality, criteria are derived to ensure the controlled systems achieve mean square exponential stabilization. Based on these criteria, the effects of diffusion item, control gains, the minimum control proportion and time-delays on exponential stability are analyzed. Examples are given to illustrate the effectiveness of the obtained theoretical results.
中文翻译:
随机反应扩散Cohen-Grossberg神经网络的间歇边界稳定。
Cohen-Grossberg神经网络(CGNN)在许多应用中都起着重要作用,并且对该系统的稳定性进行了深入研究。这项研究考虑了通过非周期性间歇边界控制的随机反应-扩散Cohen-Grossberg神经网络(SRDCGNNs)的指数稳定。讨论了没有时间延迟和有时间延迟的SRDCGNN。通过采用空间积分泛函方法和庞加莱不等式,导出了标准,以确保受控系统达到均方指数稳定。基于这些标准,分析了扩散项,控制增益,最小控制比例和时延对指数稳定性的影响。举例说明了所获得理论结果的有效性。