A novel statistical approach to model size effects in very high cycle fatigue (VHCF) is proposed in the paper. The statistical distributions of the VHCF life and of the size of the defect leading to VHCF failure (critical defect) are identified in the paper through the weakest‐link principle. The statistical distributions are able to account for the stress gradients that are present in a loaded material volume. An efficient procedure for estimating the parameters involved in the statistical distributions is also shown. The proposed statistical models are finally validated through experimental data taken from the literature. The experimental validation proves the fitting capability of the proposed models that outperform the traditional models based on the 90% risk‐volume.

中文翻译：

---

极高的循环疲劳寿命和关键缺陷尺寸：统计尺寸效应的建模

本文提出了一种新的统计方法，用于对超高循环疲劳（VHCF）中的模型尺寸影响。本文通过最弱链接原理确定了VHCF寿命和导致VHCF失效（严重缺陷）的缺陷尺寸的统计分布。统计分布能够说明加载的材料体积中存在的应力梯度。还显示了一种用于估计统计分布中涉及的参数的有效过程。最后，通过从文献中获得的实验数据对所提出的统计模型进行了验证。实验验证证明了所提出模型的拟合能力优于基于90％风险量的传统模型。