System reliability assessment of deteriorating structures subjected to time-invariant loads based on improved moment method. (September 2017)

• Record Type: Journal Article
• Authors: Zhou, Qingyu; Li, Zhengliang; Fan, Wenliang; Ang, Alfredo H-S; Liu, Runyu
• Is Part Of: Structural safety. Volume 68(2017)
• Page Start: 54
• Page End: 64
• Publication Date: 2017-09
• Subjects: System reliability -- Deteriorating structure -- Complete system failure process -- Point estimate method (PEM) -- Adaptive dimensional decomposition -- Saddlepoint approximation; Structural stability -- Periodicals; Safety factor in engineering -- Periodicals; Reliability (Engineering) -- Periodicals; Constructions -- Stabilité -- Périodiques; Coefficient de sécurité en ingénierie -- Périodiques; Fiabilité -- Périodiques; 620.86
• DOI: 10.1016/j.strusafe.2017.05.006
• Languages: English
• ISSNs: 0167-4730

Abstract

Highlights: The single equivalent performance function for time-variant system is acquired. Performance function expressed by standard normal variables and time is proposed. A sixth-moment method combining adaptive PEM and SPA is proposed. The proposed method is more robust than the existing fourth moment methods. Abstract: The time-variant system reliability analysis is a significant topic in the field of reliability engineering. Although progress has been made, there remain multiple challenges in the existing methods including the explosive number of possible failure modes, requiring correlation information, and excessive computational efforts. In the present work, an improved moment method with high accuracy, efficiency and robustness is proposed for system reliability analysis of deteriorating structures. First, by introducing the complete system failure process, an equivalent time dependent performance function describing a deteriorating structural system is obtained. Second, a point estimate method based on the adaptive trivariate dimensional decomposition is adopted to calculate the first six moments of the performance function. And third, a saddlepoint approximation involving the first six moments is developed to estimate the system failure probability. Several examples are investigated to verify the accuracy, efficiency and stability of the proposed method.