Premium
Fatigue Test Analysis of Automotive Key Parts Based on Censored Data and Small Sample Setting
Author(s) -
Wang Minlong,
Liu Xintian,
Wang Xiaolan,
Wang Yansong
Publication year - 2017
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.2089
Subject(s) - reliability (semiconductor) , quantile , amplitude , standard deviation , curve fitting , statistics , test data , confidence interval , dispersion (optics) , mathematics , automotive industry , sample (material) , reliability engineering , engineering , structural engineering , physics , optics , power (physics) , software engineering , quantum mechanics , thermodynamics , aerospace engineering
The study discusses how some challenges from a small sample fatigue test can be analysed statistically and resolved. In terms of data dispersion, fatigue data are initially processed by confidence, and then, the fatigue life curve is replaced by the parabolic reliability approximation. Fatigue data under various reliability levels are obtained. Instead of the traditional single S–N curve, the mean value curve and mean square deviation curve of the S–N curve are computed based on the processed fatigue data. The S–N curve with reliability is deduced by applying quantile theory. The S–N curve is improved by considering the effects of low‐amplitude load strengthening. The benefits of the modification are visible in the comparison of the fatigue life before and after low‐amplitude load strengthening. Copyright © 2016 John Wiley & Sons, Ltd.