α Regularization of the POD-Galerkin dynamical systems of the Kuramoto–Sivashinsky equation

摘要

In order to improve the dynamics and stability of the POD-Galerkin models of strongly-stiff systems, an α-like regularization is suggested and assessed in the present article. In this method, the POD eigenmodes of the non-linear terms are replaced by their Helmholtz filtered counterparts, while the other terms are remained unchanged. As an example, the method is applied to the POD-Galerkin models of the one-dimensional Kuramoto–Sivashinsky (KS) equation in a full chaotic regime; and the fidelity of the original and regularized models to the direct numerical simulations (DNS) are investigated. Moreover, the effects of regularization on the dynamics of various terms, and whole of the systems, are analyzed via eigenvalue analysis of each term separately, and the total dynamical system as a whole. The numerical experiments show definite effectiveness of the method and excellent improvements in the predicted dynamics and stability, by minimum number of free parameters.