东南大学学报（自然科学版）2025，Vol.55Issue(4)：955-964,10.DOI:10.3969/j.issn.1001-0505.2025.04.004
基于非线性概率模型与稀疏多项式混沌展开的结构失效概率分析
Structural failure probability analysis based on nonlinear probabilistic model and sparse polynomial chaos expansion
摘要
Abstract
A new method for calculating the probability of structural failure based on nonlinear probabilistic model and sparse polynomial chaos expansion(SPCE)was proposed.Firstly,the SPCE surrogate model with structural response was established based on the parameter recognition results of the nonlinear probabilistic model and the orthogonal matching pursuit algorithm.Secondly,the SPCE surrogate model was used to re-place the finite element simulation in the subset simulation,and the failure probability was obtained through the limit state function.Finally,using the reinforced concrete pier as a numerical example,the calculation effi-ciency of different surrogate models was compared,the failure probability under different seismic loads was calculated,and the influence of conditional failure probability,the number of event samples and measurement noise was explored.The feasibility of the proposed method was verified by the shaking table test of the scaled pier.The results show that the proposed method can quickly and accurately obtain the failure probability of the structure under dynamic load,and the error of the failure probability calculated by the proposed method com-pared with the Monte Carlo simulation method is 1.18%and 0.63%under the load conditions of 0.8g and 0.9g,while the calculation efficiency is 15.2 times and 12.6 times that of the latter,and it still maintains high accuracy and good noise immunity at 10%noise level.关键词
非线性概率模型/正交匹配追踪算法/稀疏多项式混沌展开/子集模拟/失效概率Key words
nonlinear probabilistic models/orthogonal matching pursuit algorithm/sparse polynomial chaos expansion/subset simulation/probability of failure分类
交通工程引用本文复制引用
王佐才,范弦扬,袁子青,辛宇..基于非线性概率模型与稀疏多项式混沌展开的结构失效概率分析[J].东南大学学报（自然科学版）,2025,55(4):955-964,10.基金项目
国家自然科学基金资助项目(52278301). （52278301）
