关键词:
Fatigue
Inspection scheduling
Optimization
Bayesian updating
Crack growth
Life cycle
摘要:
Successful management of deteriorating structures requires the reliable prediction of damage occurrence as well as the time-dependent damage propagation under uncertainty. The reliability of the performance prediction process can be significantly improved by integrating information gained from inspection and monitoring actions. This integration leads to a more accurate prediction of the time-dependent damage level and, eventually, to a better supported decision-making process. In this paper, a probabilistic approach is proposed to find an optimum management plan for fatigue-sensitive structures by integrating the available information from inspection actions. The proposed approach utilizes a probabilistic time-dependent damage criterion, inspection cost, and failure cost to find the optimum inspection times under uncertainty. New information resulting from inspection actions performed during the lifetime of the structure is used to update the damage propagation parameters as well as the optimization procedure. This process results in an enhanced management plan which can provide managers the ability to make real-time decisions based on inspection results. The integration of this new information and its impact on the life-cycle management process are thoroughly investigated. In addition, a realistic fatigue critical detail is used to illustrate the proposed probabilistic approach. (C) 2014 American Society of Civil Engineers.