Noise induced escape from stable invariant tori

Author:

Zhao Feng,Li Yang,Liu Xian-bin

Abstract

Abstract In the present paper, the noise-induced escape from stable invariant tori with the frame of general Langevin dynamics is investigated under weak random perturbations. Based on the large deviation theory, the quasi-potential, a quantity exponentially dominating the mean first escape time and the stationary or quasi-stationary probability density, is explored both analytically and numerically. The results show that whether the stochastic trajectories are ergodic on a torus has a crucial influence on the behavior of the quasi-potential. Specifically, there are two sources of ergodicity. One is the ergodicity of the deterministic flow, and the other requires the nondegeneracy of the noise in the tangential directions of a torus. It is found that if the ergodicity holds, the quasi-potential will be independent of the initial position on a torus, but not when the ergodicity is broken. In particular, it indicates that, for nonlinear systems driven by combined Gaussian white noise and multiple harmonic excitations, the mean first exit time varies discontinuously with respect to the frequency vector of these harmonic excitations as the noise intensity approaches zero. Adding noise to the phase of these harmonic excitations will eliminate the dependence on the initial position and thus, make the systems more robust. It reminds us that great care must be taken when dealing with noise-induced problems involving systems that possess unstable geometric structures.

Publisher

IOP Publishing

Subject

Statistics, Probability and Uncertainty,Statistics and Probability,Statistical and Nonlinear Physics

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3