The compound kriging-based importance sampling (IS) strategy is proposed for the efficient estimation of failure probability of systems with multiple failure modes. The proposed method is based on the IS probability density function of each failure mode constructed by the kriging model, where the probabilistic classification function is treated as a surrogate model for the actual failure indicator function. The whole algorithm of the proposed method can be divided into two stages. First, the kriging model is constructed to estimate the component augmented failure probabilities and obtain quasi-optimal IS samples. Secondly, the constructed kriging model is further refined based on these quasi-optimal IS samples to estimate the correction factor. Finally, the system failure probability is estimated by the product of the component augmented failure probabilities and the correction factor. The system reliability analysis results of the presented examples illustrate the feasibility of the proposed method.

提出了基于复合克里金法的重要性采样（IS）策略，用于有效估计具有多种故障模式的系统的故障概率。该方法基于克里金模型构建的每种失效模式的IS概率密度函数，其中概率分类函数被视为实际失效指示函数的代理模型。该方法的整个算法可以分为两个阶段。首先，构建克里金模型以估计组件增强失效概率并获得准最优IS样本。其次，在这些准最优IS样本的基础上进一步细化所构建的克里金模型，以估计校正因子。最后，系统故障概率由组件增强的故障概率和校正因子的乘积来估计。给出的例子的系统可靠性分析结果说明了该方法的可行性。