Abstract
This paper presents a statistical methodology to analyse damage accumulation in one part of a structure relative to another, and to determine how one crack population (e.g. pre-existing cracks) is behaving relative to another (e.g. fatigue induced cracks). The methodology is applied to damage accumulation in the early life of hip replacements where a metal prosthesis is interlocked into the bone using polymethylmethacrylate (PMMA). We have developed physical models of the hip replacement that allows visualisation of microcrack initiation and growth within the PMMA. Data was collected showing the growth of pre-load cracks (from pores and interfaces) and the initiation and growth of new cracks, in 10 regions of the cement mantle. Despite significant variability between the six specimens tested, it can be shown that highest damage growth rates can be correlated with specific sites in the structure. In conclusion, statistical analysis of the damage accumulation data allows us to make definite conclusions regarding the damage accumulation behaviour in structures where microcrack growth is disperse, and when it differs widely between specimens. Models for prediction of average crack behaviour in such structures are proposed.
Original language | English |
---|---|
Pages (from-to) | 581-593 |
Number of pages | 13 |
Journal | International Journal of Fatigue |
Volume | 20 |
Issue number | 8 |
DOIs | |
Publication status | Published - Sep 1998 |
Externally published | Yes |
Keywords
- Bayesian analysis
- Damage
- Implants