A general physics-informed neural network framework for fatigue life prediction of metallic materials

Engineering Fracture Mechanics, vol. 322, pp. 111136

Abstract

The prediction of fatigue life in metallic materials using machine learning faces challenges related to interpretability and generalization. Motived by this, a physics-informed neural network framework for fatigue life prediction is proposed in this study. Physical constraints are incorporated into the framework through partial differential inequalities derived from experimental observations. Fatigue experiments with LZ50 steel and SLM 316L are newly designed, along with experimental data for nine different metallic materials taken from literature, to validate the predictive accuracy and generalization of the model. K-fold cross-validation and Bayesian methods were employed to optimize the model architecture and physical constraint weights. The results are compared with those obtained from a traditional artificial neural network . It is demonstrated that the physics-informed neural network achieves higher predictive accuracy, better consistency with physical principles, and improved generalization in most cases. More reliable predictions are observed in high-life regions, where artificial neural networks tend to exhibit greater variability. These findings highlight the versatility and robustness of the proposed framework, which is suggested as a promising general approach for fatigue life prediction through the integration of physical knowledge into machine learning-based modeling.

Authors 6

1. Shuwei Zhou corresponding Aachen Institute of Metal Forming

   RWTH Aachen University

   Affiliation as printed

   Institute of Metal Forming, RWTH Aachen University, Aachen, 52072, Germany

2. Manuel Henrich Aachen Institute of Metal Forming

   RWTH Aachen University

   Affiliation as printed

   Institute of Metal Forming, RWTH Aachen University, Aachen, 52072, Germany

3. Zhichao Wei Aachen Institute of Metal Forming

   Universität der Bundeswehr München · RWTH Aachen University

   Affiliation as printed

   Institute for Engineering Mechanics and Structural Analysis, University of the Bundeswehr Munich, Neubiberg, 85577, Germany

   Institute of Metal Forming, RWTH Aachen University, Aachen, 52072, Germany

4. Feng Feng

   Southwest Jiaotong University

   Affiliation as printed

   State Key Laboratory of Rail Transit Vehicle System, Southwest Jiaotong University, Chengdu, 610031, PR China

5. Bing Yang

   Southwest Jiaotong University

   Affiliation as printed

   State Key Laboratory of Rail Transit Vehicle System, Southwest Jiaotong University, Chengdu, 610031, PR China

6. Sebastian Münstermann Aachen Institute of Metal Forming

   RWTH Aachen University

   Affiliation as printed

   Institute of Metal Forming, RWTH Aachen University, Aachen, 52072, Germany