微分包含
惯性参考系
控制理论(社会学)
沉降时间
理论(学习稳定性)
数学
转化(遗传学)
人工神经网络
时间导数
计算机科学
应用数学
控制(管理)
数学优化
数学分析
工程类
控制工程
人工智能
阶跃响应
物理
机器学习
基因
化学
量子力学
生物化学
作者
Fanchao Kong,Quanxin Zhu,Tingwen Huang
标识
DOI:10.1109/tsmc.2021.3096261
摘要
In this article, a class of discontinuous inertial neural networks (DINNs) with parameter uncertainties and time delays is studied. The main aim is to investigate the new fixed-time stability (FTS). In order to achieve the targets, first, by introducing the generalized variable transformation and differential inclusions theory, two kinds of drive–response differential inclusion systems are established. Based on the definition of FTS and inequality technologies, by constructing the Lyapunov–Krasovskii functional (LKF), whose derivative is allowed to be indefinite, new delay-dependent criteria shown by some simple inequalities are derived for the purposing of achieving the FTS based on the designed discontinuous control strategies. Moreover, the new settling time (ST) is given. Compared to the previous stability results on INNs, the results established and the approaches applied are absolutely new. Finally, examples are given to show the effectiveness of the established results.
科研通智能强力驱动
Strongly Powered by AbleSci AI