Scatter factor analysis for sample fatigue life based on a three parameter birnbaum-saunders distribution

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

The Birnbaum-Saunders (B-S) distribution was derived as a lifetime model for specimens subjected to cyclic patterns of stresses and strains. A new model of three-parameter Birnbaum-Saunders distribution was derived by proposing a dependence assumption on account of the assumption of independence of crack growth from cycle to cycle being unrealistic. A method for reliable fatigue life analysis and assessment based on three-parameter Birnbaum-Saunders distribution was studied, and fatigue life scatter factor formulae of three-parameter Birnbaum-Saunders distribution based on median value was derived, and based on the order statistic theory, the scatter factor formulae corresponding to maximum and minimum test life were presented. The results of calculating examples show that life scatter factors of three-parameter Birnbaum-Saunders distribution are usually better than that of the lognormal distributions as well as that of the original Birnbaum-Saunders distribution.

Original languageEnglish
Title of host publicationRAMS 2019 - 2019 Annual Reliability and Maintainability Symposium, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538665541
DOIs
StatePublished - Jan 2019
Event2019 Annual Reliability and Maintainability Symposium, RAMS 2019 - Orlando, United States
Duration: 28 Jan 201931 Jan 2019

Publication series

NameProceedings - Annual Reliability and Maintainability Symposium
Volume2019-January
ISSN (Print)0149-144X

Conference

Conference2019 Annual Reliability and Maintainability Symposium, RAMS 2019
Country/TerritoryUnited States
CityOrlando
Period28/01/1931/01/19

Keywords

  • Fatigue life
  • Life scatter factor
  • Order statistic
  • Three-parameter B-S distribution

Fingerprint

Dive into the research topics of 'Scatter factor analysis for sample fatigue life based on a three parameter birnbaum-saunders distribution'. Together they form a unique fingerprint.

Cite this