Fatigue and fracture reliability and maintainability process for structural systems.
AuthorKung, Chieh Julius
AdvisorWirsching, Paul H.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractStructures subject to oscillatory stresses are vulnerable to fatigue and/or fracture failure. But some fatigue and fracture design factors are subject to considerable uncertainty. Therefore, probabilistic and statistical methods are appropriate as tools for making design decisions and reliability assessments of structural systems. System reliability can be improved by a maintenance program of periodic inspection and repair or replacement of damaged members. But there are also uncertainities introduced by the inspection process. In summary, the fatigue/fracture reliability and maintainability (FRM) process for a structural system is a "replacement/repair renewal" process complicated by a large number of random variables and small failure probabilities for a typical system. The goal of this research is to develop an efficient numerical method to estimate the reliability and performance of structural systems subject to the FRM process. Reliability methods such as first order reliability methods (FORM) and second order reliability methods (SORM) are not appropriate as tools to analyze the FRM process. As a continuation of the FRM study by Torng (1989), an equivalent member approach is introduced. A sampling simulation procedure is employed. The "equivalent member approach" considers each member (a series system) as a single member having one single stress concentration representing the "weakest" component. Elementary extreme value theory is applied to approximate the distribution of ultimate strength and fatigue strength of the member. It is demonstrated that significant savings in computer (CPU) time is achieved without sacrificing accuracy in reliability and performance estimates. Typically, the sampling scheme employed herein will result in an improvement in computer time by a factor of 1000 relative to the direct Monte Carlo method. One goal of this research is to study the relationship between the design factors of the tension-leg platform (TLP) tendon system. Effects on system reliability and maintenance performance (repair and replacement rates) are studied as a function of (a) number of joints, J, (b) number of members, M, (c) inspection frequency, (d) inspection sensitivity as defined by the POD (probability of detection) curve, (e) ultimate tensile strength, (f) repair policy, etc. The performance of an initially damaged or flawed tendon system is investigated. The reliability of a tendon system that uses pressurized tendon to detect through thickness cracks is studied as is the vulnerability of the tendon system before replacement of broken tendon.
Degree ProgramAerospace and Mechanical Engineering