Combined Notch and Size Effect Modeling in a Local Probabilistic Approach for LCF

TL;DR

This paper enhances a probabilistic low cycle fatigue model by integrating notch support based on stress gradients, validated across various materials and geometries, reducing testing efforts compared to deterministic methods.

Contribution

It introduces a combined notch and size effect model within a local probabilistic framework, including FEA-based calibration for non-homogeneous stress fields.

Findings

Model accurately predicts fatigue life across different geometries.

Incorporating notch support improves prediction accuracy.

Reduced testing effort compared to deterministic models.

Abstract

In recent years a local probabilistic model for low cycle fatigue (LCF) based on the statistical size effect has been developed and applied on engineering components. Here, the notch support extension based on the stress gradient effect is described in detail, as well as an FEA-based parameter calibration. An FEA is necessary to simulate non-homogeneous stress fields in non-smooth specimens which exhibit gradients and determine size effects. The hazard density approach and the surface integration over the FEA stress lead to geometry-independent model parameters. Three different materials (superalloys IN-939, Rene80, steel 26NiCrMoV14-5) and three different geometry types (smooth, notch, cooling hole specimens) are considered for a more comprehensive validation of the probabilistic LCF model and to demonstrate its wide application range. At the same time, a reduced testing effort is needed compared to deterministic model predictions with notch support.