一种基于无源性理论的模糊Hopfield神经网络学习律设计方法  

Passivity-based learning law design for a class of fuzzy Hopfield neural networks

作　　者：王婧[1] 柏建军 薛安克 WANG Jing;BAI Jian-jun;XUE An-ke(Mechanical and Automotive Engineering Department of Laiwu Vocational and Technical College,Laiwu Shandong 271100,China;Key Lab for IOT and Information Fusion Technology of Zhejiang,Hangzhou Zhejiang 310018,China;Institute of Information and Control,Hangzhou Dianzi University,Hangzhou Zhejiang 310018,China)

机构地区：[1]莱芜职业技术学院机械与汽车工程系,山东莱芜271100 [2]浙江省物联感知与信息融合技术重点实验室,浙江杭州310018 [3]杭州电子科技大学信息与控制研究所,浙江杭州310018

出　　处：《控制理论与应用》2020年第2期405-410,共6页Control Theory & Applications

基　　金：国家自然科学基金项目(61773146,61333009,U1509205,61703132,61427808)资助.

摘　　要：本文研究了一类模糊Hopfield神经网络系统的稳定性问题.首先,基于无源性理论,设计了一种新的权重学习律,并通过构造的模糊Lyapunov函数证明了系统从输入到输出是无源的.在此基础上,证明了系统在该学习律下是输入到状态稳定的.相比于传统的公共Lypaunov函数,本文所提的模糊Lyapunov函数能保证系统具有更好的性能.最后,通过数值仿真验证了所提方法的有效性.The stability problem of a class of fuzzy Hopfield neural networks is investigated in this paper.Based on the passivity theory,a new learning law is proposed to guarantee the system to be input-to-output passive by constructing a new fuzzy Lyapunov function,which will allow the system a better performance compared to the common Lyapunov function.Then the system is proved to be input-to-state stable by using the new learning law.Finally,a numerical example is given to show the effectiveness of the proposed approach.

关 键 词：模糊HOPFIELD神经网络 无源性 学习律 输入–状态稳定 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]