It is proposed to develop and identify a probabilistic two-scale model for HCF that accounts for the failure of samples but also for the thermal effects during cyclic loadings in a unified framework.The probabilistic model is based on a Poisson point process. Within the weakest link theory, the model corresponds to a Weibull law for the fatigue limits. The thermal effects can be described if one considers the same hypotheses apart from the weakest link assumption. A method of identification is proposed and uses temperature measurements to identify the scatter in an S/N curve. The validation of the model is obtained by predicting S/N curves for different effective volumes of a dual-phase steel.

No abstract