A Novel Method for Evaluating the Averaged Drift and Diffusion Coefficients of High DOF Quasi-Non-integrable Hamiltonian Systems

Sun Jiaojiao,Huan Ronghua,Deng Maolin,Zhu Weiqiu
DOI: https://doi.org/10.1007/s11071-021-06952-9
IF: 5.741
2021-01-01
Nonlinear Dynamics
Abstract:Most multi-degree-of-freedom (MDOF) strongly nonlinear vibratory systems driven by Gaussian white noise can be modeled as quasi-non-integrable Hamiltonian systems. The prominent advantage of the stochastic averaging method of quasi-non-integrable Hamiltonian systems is that the averaged Itô stochastic differential equation (SDE) for Hamiltonian is only one-dimensional and the exact stationary solution of Fokker–Planck–Kolmogorov equation associated with the averaged Itô SDE can be readily obtained. However, for n -DOF quasi-non-integrable Hamiltonian systems with n>2 , one challenge to apply the stochastic averaging method for quasi-non-integrable Hamiltonian systems is how to evaluate ( 2n-1 )-dimensional domain integrals in determining the averaged drift and diffusion coefficients. In the present paper, a novel method using two-step generalized elliptical coordinate transformations is proposed to convert the ( 2n-1 )-dimensional domain integrals into ( n-1 )-fold repeated integrals and then n -fold repeated integrals. For quasi-non-integrable Hamiltonian systems with nonlinear momentum polynomial dampings and excitation coefficients, the first ( n-1 )-fold repeated integrals can be completed easily. Thus, the ( 2n-1 )-dimensional domain integrals are actually reduced to much simple n -fold repeated integrals. A 5-DOF strongly nonlinear stochastic system is taken as an example to show the procedure and Monte Carlo simulation is used to show the validity of the proposed method.
What problem does this paper attempt to address?