Scholarlink

<p>This paper addresses the issue of exponential stability for discrete-time switched neural networks (DSNNs) with time-varying delays. The considered system switches from one mode to another according to the mode-dependent average dwell-time (MDADT), which is more practical than classical average dwell-time (ADT) switching. Moreover, the specific goal is to use some improved zero equations with unrestricted weight matrices, which are provided to obtain more free variables. In addition, the augmented Lyapunov-Krasovskii functionals (LKFs) containing some new state-related vectors are proposed under MDADT switching signal, which effectively obviate the appearance for high-order polynomials. Furthermore, an augmented delay-product-type LKF is also established for enhancing the stability conditions of DSNNs. Then, based on the above methods and some classic inequalities, less conservative stability conditions are provided for DSNNs with time-varying delays. Finally, two numerical examples are offered to show the validity of obtained results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Exponential Stability of DSNNs via Improved Zero Equations With Unrestricted Weight Matrices

  • 期刊论文

  • 发表日期：

    2025年4月12日

查看全文
  • Da Chen,
  • Kaibo Shi,
  • Xingwen Liu,
  • Jinde Cao,
  • Shiping Wen,
  • Xiangxiang Wang

摘要

This paper addresses the issue of exponential stability for discrete-time switched neural networks (DSNNs) with time-varying delays. The considered system switches from one mode to another according to the mode-dependent average dwell-time (MDADT), which is more practical than classical average dwell-time (ADT) switching. Moreover, the specific goal is to use some improved zero equations with unrestricted weight matrices, which are provided to obtain more free variables. In addition, the augmented Lyapunov-Krasovskii functionals (LKFs) containing some new state-related vectors are proposed under MDADT switching signal, which effectively obviate the appearance for high-order polynomials. Furthermore, an augmented delay-product-type LKF is also established for enhancing the stability conditions of DSNNs. Then, based on the above methods and some classic inequalities, less conservative stability conditions are provided for DSNNs with time-varying delays. Finally, two numerical examples are offered to show the validity of obtained results.