Fatigue Life Prediction Based on the Advanced Fatigue Crack Growth Model and the Monte-Carlo Simulation Method
Loading...
Date
2015-01-05
Authors
Bogdanov, Sergey
Advisor
Journal Title
Journal ISSN
Volume Title
Publisher
University of Waterloo
Abstract
It is well-known that loss of the structural stability due to propagation of the dominant
crack is one of the main sources of fatigue failures. Therefore, computationally feasible
fatigue crack growth model is essential for advanced fatigue life analysis. Appropriate
model should be able to calculate the time required for crack to grow: from the initial crack of any size to the final critical length; in any structure; under applied variable loading.
The UniGrow two-parameter total driving force was proposed ten years ago. Since then, the fatigue crack growth model based on it, was enhanced with a set of memory rules and
resulted in a sophisticated fatigue crack growth software package which was extensively
validated on the basis of available experimental data. Nevertheless, in its previous form
the UniGrow model had several serious limitations: it was not able to model the fatigue
crack growth under plane-strain conditions; the Neuber rule was the only method used
for elastic-plastic stress-strain analysis; the fatigue crack growth prediction was limited
to macro-cracks and the variability of the material response to cyclic loading has not
been considered. In addition to these shortcomings, there were no preferred method for
estimation of the material block size , which is one of the main parameters required for
the analysis. Therefore, further modeling and validation of the UniGrow concept were
required before it could be coupled with the "Monte-Carlo" simulation method.
The main research goal pursued in this work was to improve and remove limitations of
the existing fatigue crack growth model and to combine this model with "Monte-Carlo" simulations. Such combination enables the assessment of the reliability of predicted fatigue lives. Thus, the result of this work is an extensive probabilistic fatigue crack growth model, which is able to perform the analysis of structures in the plane-strain or plane-stress condition. Comprehensive set of the fatigue crack growth data for aluminum and steel alloys generated under the constant and variable amplitude loading was used for the validation.
Description
Keywords
Fatigue Crack Growth, Monte-Carlo simulation, UniGrow, Stress-strain analysis