近似的理想重点サンプリング・シミュレーション法に基づく構造破損確率推定法(機械工学科)

概要

論文の詳細を見る
本文データの一部は、CiNiiから複製したものである。This paper describes a quasi ideal importance sampling simulation combined with the conditional expectation method in a simulation-based structural reliability estimation. The marginal probability densities of a quasi ideal importance sampling are constructed numerically. The ranges of the basic random variables are split into several segments and the conditional failure probabilities are calculated at the center of the respective segments and the conditional failure probabilities are determined and the structural failure probability is evaluated approximately by a piecewise integration. The marginal densities of the quasi ideal importance sampling are so determined as to be proportional to the conditional failure probabilities at the respective segments and the structural failure probability evaluated approximately. As the numerical examples, the quasi ideal importance sampling simulations combined with the conditional expectation are executed to estimate the failure probabilities of structures with various limit state functions and the results are compared with those estimated by other simulation approaches. The proposed method gives the failure probability estimations accurately with shorter processing time.
近畿大学の論文