This thesis presents a probabilistic model, specifically a nonhomogeneous Markov model, for the description of fatigue crack growth. This model is capable of predicting the statistics of crack growth, the mean crack size and the variance in the crack size, at future times given two empirically determined system parameters, $ lambda$ and $ kappa$, and an initial crack size distribution. Prior to the mathematical development of this model a brief review of existing crack growth models is presented. After the mathematical development of the model has been presented the results from a previous application in the study of pitting corrosion are presented. Following this, the results from the modelling of several existing fatigue data sets are examined in order to determine the necessary empirical material parameters and also to verify the model's applicability. Once the model has been verified and the empirical parameters determined one data set is chosen in order to illustrate the uses of this model for component reliability predictions. These uses include reliability at a future time, inspection optimization and the effect of changes in repair policy on reliability.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.59827 |
| Date | January 1990 |
| Creators | Bohn, Steven R. (Steven Raymond) |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Master of Engineering (Department of Mechanical Engineering.) |
| Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
| Relation | alephsysno: 001171822, proquestno: AAIMM66437, Theses scanned by UMI/ProQuest. |
Page generated in 0.0021 seconds