arXivDaily arXiv每日学术速递 周一至周五更新
arXiv math/0610920math.DS
收藏

一类时滞动力系统殆周期解的全局指数稳定性

Global Exponential Stability of Almost Periodic Solution for A Large Class of Delayed Dynamical Systems

Wenlian Lu, Tianping Chen

摘要:

具有可变自抑制、可变连接权值和可变输入的时滞神经网络的研究是一个重要的课题。在现实世界中,自我抑制、互连权重和输入应该随时间而变化。在本文中,我们讨论了一类具有概周期自抑制、关联权和输入的时滞动力系统。该模型具有普适性,包括时变时滞系统、分布时滞系统以及两者的组合。证明了在适当的条件下,该系统存在唯一的全局指数稳定的殆周期解。本文提出了一种新的方法,它独立于现有的关于动力系统的殆周期解存在性的理论。

英文摘要:

Research of delayed neural networks with variable self-inhibitions, inter-connection weights, and inputs is an important issue. %In the real world, self-inhibitions, %inter-connection weights, and inputs should vary through time. In In this paper, we discuss a large class of delayed dynamical systems with almost periodic self-inhibitions, inter-connection weights, and inputs. This model is universal and includes delayed systems with time-varying delays, distributed delays as well as combination of both. We prove that under some mild conditions, the system has a unique almost periodic solution, which is globally exponentially stable. We propose a new approach, which is independent of existing theory concerning with existence of almost periodic solution for dynamical systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑