<p>Inspection of aircraft structures usually requires the disassembly of structures. The intrusive nature of the disassembly process is a major cause of unreliability. Thus, inspection without disassembly is favored, but it causes decreased structural reliability due to increased uncertainties in crack size measurement. This paper proposes and validates a Bayesian method for estimating the initial size of cracks hidden by the fastener head. A shifted probability of detection method was used to provide a reasonable likelihood function for inspection with a fastener. Combined with Monte Carlo simulation and backpropagation, the initial size of the hidden crack was estimated. The resultant crack size distribution was applied to structural reliability analysis, and the result was compared with experimental results. The method yielded slightly conservative risk, but a reasonable inspection interval. The decreased uncertainty benefits the operator, who enjoys an increased inspection and maintenance schedule.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Bayesian Method for Estimating Equivalent Initial Size Distribution of Hidden Crack Under Fastener Head

  • Youngchan Kim,
  • Seonhwa Jung,
  • Dooyoul Lee

摘要

Inspection of aircraft structures usually requires the disassembly of structures. The intrusive nature of the disassembly process is a major cause of unreliability. Thus, inspection without disassembly is favored, but it causes decreased structural reliability due to increased uncertainties in crack size measurement. This paper proposes and validates a Bayesian method for estimating the initial size of cracks hidden by the fastener head. A shifted probability of detection method was used to provide a reasonable likelihood function for inspection with a fastener. Combined with Monte Carlo simulation and backpropagation, the initial size of the hidden crack was estimated. The resultant crack size distribution was applied to structural reliability analysis, and the result was compared with experimental results. The method yielded slightly conservative risk, but a reasonable inspection interval. The decreased uncertainty benefits the operator, who enjoys an increased inspection and maintenance schedule.