Fitting fatigue test data with a novel S-N curve using frequentist and Bayesian inference

In design against fatigue, a lower bound stress range vs. endurance curve (S-N curve) is employed to characterize fatigue resistance of plain material and structural details. With respect to the inherent variability of the fatigue life, the S-N curve is related to a certain probability of exceedance, a percentile of the fatigue life. This paper is concerned with modelling and estimating uncertainties in fatigue resistance of welded joints under constant amplitude loading. A new S-N curve format is proposed and fitted to fatigue test data by using the Maximum Likelihood Method. The results have been compared with the Random Fatigue Limit Model and the Bilinear Random Fatigue Limit Model. The proposed S-N curve appears to be more accurate in describing the S-N relation in high-cycle fatigue: it presents a smooth transition from finite to infinite-life regions and, differently from previous non-linear S-N relations with fatigue limit, this transition is controlled by an independent model parameter. Thereby it provides more flexibility for statistical fitting. In addition, a Bayesian framework is defined to fit the proposed relation including informative and non-informative prior distributions. © 2017 Elsevier Ltd

[1]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[2]  Cetin Morris Sonsino,et al.  Fatigue assessment of welded joints by local approaches Second edition , 2007 .

[3]  M.B.A. van Asselt,et al.  Perspectives on Uncertainty and Risk: The PRIMA Approach to Decision Support , 2000 .

[4]  S. S. Wilks.  The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses , 1938 .

[5]  S. J. Maddox,et al.  Fatigue Testing of Full Scale Girth Welded Pipes Under Variable Amplitude Loading , 2012 .

[6]  Y. Pawitan.  In all likelihood , 2001 .

[7]  Ivo Babuska,et al.  Bayesian inference and model comparison for metallic fatigue data , 2015, 1512.01779.

[8]  E. Haibach,et al.  THE METHOD OF MAXIMUM LIKELIHOOD APPLIED TO THE STATISTICAL ANALYSIS OF FATIGUE DATA INCLUDING RUN-OUTS , 1979 .

[9]  David R. Anderson,et al.  Model Selection and Multimodel Inference , 2003 .

[10]  Maurizio Guida,et al.  A Bayesian analysis of fatigue data , 2010 .

[11]  H. Akaike,et al.  Information Theory and an Extension of the Maximum Likelihood Principle , 1973 .

[12]  J. Evans.  Straightforward Statistics for the Behavioral Sciences , 1995 .

[13]  Eugen Brühwiler,et al.  S-N-P Fatigue Curves using Maximum Likelihood , 2014 .

[14]  J. Lawless.  Statistical Models and Methods for Lifetime Data , 2002 .

[15]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[16]  G. Marquis,et al.  Analysis of Welded Structures with Failed and Non-Failed Welds Based on Maximum Likelihood , 2002 .

[17]  W. Meeker,et al.  Estimating fatigue curves with the random fatigue-limit model , 1999 .

[18]  A. Gelman,et al.  Weak convergence and optimal scaling of random walk Metropolis algorithms , 1997 .

[19]  Alain Nussbaumer,et al.  Estimation of fatigue S-N curves of welded joints using advanced probabilistic approach , 2017 .

[20]  W. Nelson.  Statistical Methods for Reliability Data , 1998 .

[21]  Klaus Hentschel,et al.  Weld geometry as a factor controlling the fatigue strength of butt welded joints , 1990 .

[22]  Elizabeth A. Peck,et al.  Introduction to Linear Regression Analysis , 2001 .

[23]  G. Edwards,et al.  A Bayesian method for establishing fatigue design curves , 1984 .

[24]  S. Sarkani,et al.  Runout analysis in fatigue investigation , 2007 .

[25]  M. E. Haddad,et al.  Prediction of non propagating cracks , 1979 .

[26]  C. E. Stromeyer.  The determination of fatigue limits under alternating stress conditions , 1914 .

[27]  C. Robert,et al.  Understanding Computational Bayesian Statistics , 2009 .

[28]  O. Basquin.  The exponential law of endurance tests , 1910 .

[29]  Michael Havbro Faber,et al.  On the Treatment of Uncertainties and Probabilities in Engineering Decision Analysis , 2005 .

[30]  L. Held,et al.  Applied Statistical Inference: Likelihood and Bayes , 2013 .

[31]  P. H. Wirsching,et al.  Design Curve to Characterize Fatigue Strength , 1996 .