Stability criterion for delayed neural networks via Wirtinger-based multiple integral inequality

详细信息    查看全文

• 作者：Sanbo Ding ^{dingsanbo@163.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Zhanshan Wang ; ^{zhanshan_wang@163.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Yanming Wu ^{yanmingwu@yeah.net" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Huaguang Zhang ^{hgzhang@ieee.org" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Delayed neural networks ; Wirtinger-based multiple integral inequality ; Global asymptotic stability ; Lyapunov&ndash ; Krasovskii functional with multiple integral terms
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：19 November 2016
• 年：2016
• 卷：214
• 期：Complete
• 页码：53-60
• 全文大小：718 K

文摘

This brief provides an alternative way to reduce the conservativeness of the stability criterion for neural networks (NNs) with time-varying delays. The core is that a series of multiple integral terms are considered as a part of the Lyapunov–Krasovskii functional (LKF). In order to estimate the multiple integral terms in the derivative of the LKF, a multiple integral inequality, named Wirtinger-based multiple integral inequality (WMII), is proposed. This inequality includes some recent related results as its special cases. Based on the multiple integral forms of LKF and the WMII, a novel delay dependent stability criterion for NNs with time-varying delays is derived. The effectiveness of the established stability criterion is verified by an open example.