AK-ARBIS: An improved AK-MCS based on the adaptive radial-based importance sampling for small failure probability

• Published in: Structural safety Vol. 82; p. 101891
• Main Authors: Yun, Wanying, Lu, Zhenzhou, Jiang, Xian, Zhang, Leigang, He, Pengfei
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Ltd 01.01.2020; Elsevier BV
• Subjects: Adaptive radial-based importance sampling; Adaptive sampling; Candidate sampling pool; Computer simulation; Convergence; Failure analysis; Importance sampling; Iterative methods; Kriging interpolation; Kriging model; Monte Carlo simulation; Nonlinear analysis; Probability; Reliability analysis; Reliability aspects; Sampling; Size reduction; Small failure probability; Training time; Turbine disks; Turbines; Wing boxes; Small failure probability; Kriging model; Adaptive radial-based importance sampling; Candidate sampling pool; Training time

•An improved AK-MCS method is proposed to estimate the small failure probability.•An optimal β-sphere is searched without any extra model evaluations.•Kriging model is ceaselessly updated layer by layer outside the current β-sphere.•The adaptive Kriging model is finished until an optimal β-sphere is...