具有马尔可夫跳的多重定时滞随机神经网络的依分布渐近稳定性  

Asymptotic Stability in Distribution of Stochastic Neural Networks with Multiple Time Delays and Markovian Swithching

作　　者：王莉[1,2] 王玉春 WANG Li;WANG Yuchun(School of Liberal and Science,Suqian University,Suqian 223800,Jiangsu,China;School of Mathematics,China University of Mining and Technology,Xuzhou 221116,Jiangsu,China)

机构地区：[1]宿迁学院文理学院,江苏宿迁223800 [2]中国矿业大学数学学院,江苏徐州221116

出　　处：《山西师范大学学报（自然科学版）》2023年第3期9-15,共7页Journal of Shanxi Normal University(Natural Science Edition)

基　　金：国家自然科学基金(61873271);宿迁市科技计划项目(21SYB-33).

摘　　要：在神经网络的电子实现中往往伴有时滞现象,为了更好地应用,研究了一类带马氏切换的随机神经网络在多重延时下按分布的渐近稳定性.根据模型方程的特点,并结合按分布渐近稳定性的充分条件,构造了Lyapunov泛函,通过广义的Ito公式计算、不等式放大技巧,寻求到了系统相应的稳定性充分条件.最后举例做了数值拟合,验证了结果的可行性.The electronic implementation of neural networks often has time delays.For better application,the asymptotic stability in distribution of a class of stochastic neural networks with Markov switching under multi-ple delays is investigated.According to the characteristics of the model equation and the sufficient condition of as-ymptotic stability in distribution,the Lyapunov functional is constructed,and the corresponding sufficient condi-tion of stability of the system is obtained by means of generalized Ito formula calculation and inequality magnif-ying technique.Finally,a numerical example is given to verify the feasibility of the results.

关 键 词：马氏切换 多重定时滞 随机神经网络 依分布渐近稳定 LYAPUNOV泛函 

分 类 号：O175[理学—数学]