A new structural uncertainty analysis method based on polynomial expansions

摘要

This paper proposes a new method based on the polynomial expansions for structural uncertainty analysis. A generalized finite difference method (GFDM) based on the Taylor expansion is adopted to compute the structural responses, which has good adaptabilities to the analysis domains due to its meshless property. With the help of the polynomial chaos expansions (PCE), random variables subjected to any probability distribution are implicitly quantified. The GFDMPCE method combines GFDM and PCE, is verified by the classical Monte Carlo method (MCM) in terms of calculation accuracy and efficiency. This method is non-intrusive, rigorous in mathematical theory, and shows bright prospects for the robust analysis of large-scale and complex structures.