Estimating Notch Fatigue Limits via a Machine Learning-Based Approach Structured According to the Classic Kf Formulas

Aug 3, 2026

This paper addresses estimating notch fatigue limits using machine learning models aligned with classical Kf formulas. The strategy leverages constitutive elements associated with foundational approaches by Peterson, Neuber, Heywood, and Topper. Algorithms are trained and evaluated on a literature dataset comprising 238 notch fatigue limits. Results support machine learning as a promising path for designing notched components against fatigue. Accuracy can improve by increasing dataset size and calibration quality, adding informative inputs such as grain size or hardness, and exploiting regression flexibility and generalization for predicting new materials.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Introduction

Fatigue behavior and stress concentrators in notched materials

Ferrous materials and the fatigue limit concept

Non-ferrous materials and endurance limit framing

This is a repository copy of Estimating notch fatigue limits via a machine learning-based approach structured according to the classic Kf formulas. White Rose Research Online URL for this paper: [https://eprints.whiterose.ac.uk/205168/](https://eprints.whiterose.ac.uk/205168/) Version: Published Version Article: Susmel, [L. orcid.org/0000-0001-7753-9176](L. orcid.org/0000-0001-7753-9176) (2024) Estimating notch fatigue limits via a machine learning-based approach structured according to the classic Kf formulas. International Journal of Fatigue, 179. 108029. ISSN 0142-1123 [https://doi.org/10.1016/j.ijfatigue.2023.108029](https://doi.org/10.1016/j.ijfatigue.2023.108029) Reuse This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the authors for the original work. More information and the full terms of the licence here: [https://creativecommons.org/licenses/](https://creativecommons.org/licenses/) Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [eprints@whiterose.ac.uk](eprints@whiterose.ac.uk) including the URL of the record and the reason for the withdrawal request. [eprints@whiterose.ac.uk](eprints@whiterose.ac.uk)[ ](eprints@whiterose.ac.uk)[https://eprints.whiterose.ac.uk/](https://eprints.whiterose.ac.uk/) International Journal of Fatigue 179 (2024) 108029 Contents lists available at ScienceDirect International Journal of Fatigue journal [homepage: www.elsevier.com/locate/ijfatigue](homepage: www.elsevier.com/locate/ijfatigue) | Estimating notch fatigue limits via a machine learning-based approach structured according to the classic Kf formulas
Luca Susmel
Department of Civil and Structural Engineering, The University of Sheffield, Mapping Street, Sheffield S1 3JD, UK | | | | --- | --- | --- | | A R T I C L E I N F O | | A B S T R A C T | | Keywords:
Notch fatigue limit Machine learning Kf
Critical distance | | This paper deals with the problem of estimating notch fatigue limits via machine learning. The proposed strategy is based on those constitutive elements that were used by the pioneers like Peterson, Neuber, Heywood, and Topper to devise their well-known formulas. The machine learning algorithms being considered were trained and tested using a database containing 238 notch fatigue limits taken from the literature. The outcomes from this study confirm that machine learning is a promising approach for designing notched components against fatigue. In particular, the accuracy in the estimates can easily be increased by simply increasing size and quality of the calibration dataset. Further, since machine learning regression models are highly flexible and can handle highdimensional datasets with many input features, they can capture complex relationships between input features and the target variable. This means that the accuracy in estimating notch fatigue limit can be increased by including in the analyses further input features like, for instance, grain size or hardness. Finally, machine learning’s generalization ability is crucial for regression tasks where the goal is to predict values for new materials. | 1. Introduction The prediction of the fatigue behaviour of materials containing notches (such as keyseats, fillets, or holes) is a topic of enduring interest in engineering and materials science. In structural components, notches act as stress concentrators and they are known to significantly reduce the fatigue strength of materials. Since the estimation of the fatigue strength of notched components is of primary importance in ensuring the structural integrity in various engineering applications, the development of reliable design methodologies has been a challenge for a large number of researchers, resulting in decades of dedicated in

FAQs

What is the main goal of the paper?

To estimate notch fatigue limits using machine learning models structured according to the classic Kf formulas.

Which dataset and methods are used to train the machine learning algorithms?

The models are trained and tested using a literature database containing 238 notch fatigue limits.

How can the estimation accuracy be improved?

By increasing the size and quality of the calibration dataset, and by adding additional input features such as grain size or hardness.

Aug 3, 2026

Upload by Lucas Martin