PSI - Issue 38

Robin Hauteville et al. / Procedia Structural Integrity 38 (2022) 507–518 Robin Hauteville, Xavier Hermite, Fabien Lefebvre / Structural Integrity Procedia 00 (2021) 000 – 000

518 12

Figure 8: Bastenaire median and design curves

7. Conclusion and perspectives The approach for the post-processing of fatigue data, for obtaining a S-N curve according to a defined model with the consideration of runouts data, has shown that it is reliable through several statistical tests. For the example treated in this document, the dataset did not have a very large dispersion because the process studied during these tests was quite well known and under control. However, the question may be raised about the need for statistical analyses when dealing with new materials whose process is not well under control. Indeed, some statistical anomalies are no physical anomalies: different failure mode due to defects, very large dispersion due to material heterogeneity (not detectable at each level and could lead to reject the Bartlett test), etc. Therefore, the tool developed by Cetim should allow the user to move in a more trivial way towards more advanced analyses such as fracture surface analyses to confirm, for example, an outlier when the failure is due to a defect. This tool is also required to evolve allowing analysis according to other models of S-N curve such as Weibull, Wöhler or Manson-Coffin-Basquin for EN curves. An advanced module for the determination of the endurance limit (Staircase with automated test points sorting) will be added as well as module specific to the multiaxial fatigue criteria (Dang Van, Crossland, etc.) and mean stress correction (Söderberg, Goodman, Gerber, etc.). References [1] H. J. Grover; Battelle Memorial Institute, 1954, Fatigue of metals and structures. [2] A. Brand et al, 1999, Données technologiques sur la fatigue [3] OH. Basquin, 1910, The exponential law of endurance tests. [4] CE. Stromeyer, 1914, The determination of fatigue limits under alternating stress conditions. [5] F. Bastenaire, 1960, Etude statistique et physique de la dispersion des résistances et des endurances à la fatigue. [6] E. Castillo, A. F. Canteli, 2009, Springer, A unified statistical methodology for modeling fatigue damage. [7] Norme Française A03-405, 1991, Essais de fatigue – Traitement statistiques des données. [8] P. Bonnet, X. Hermite, I. Huther, F. Lefebvre, 2016. Guide pour le choix d’une méthode d’essais de fatigue et de l’analyse statistique associée. [9] ISO 12107, 2012, Metallic materials - Fatigue testing - Statistical planning and analysis of data [10] NF ISO 5725-2 - Accuracy (trueness and precision) of measurement methods and results - Part 2: basic method for the determination of

repeatability and reproducibility of a standard measurement method [11] M. S. Bartlett, 1937, Properties of sufficiency and statistical test. [12] G. Saporta, 2006, Probabilités analyse des données et statistique, 2 ème édition, P339-341 [13] A. Hobbacher, IIW-1823-07, 2016, Recommendations for Fatigue Design of Welded Joints and Components.