Defect-based Physics-Informed Machine Learning Framework for Fatigue Prediction

I would like to draw your attention to our recently proposed predictive method based on a semi-empirical model (LEFM) and Neural Network, exploiting the Physics-informed Machine Learning concept. We show how the accuracy of state-of-the-art fatigue predictive models, based on defects present in materials, can be significantly boosted by accounting for additional morphological features via Physics-Informed Machine Learning. Although defect-based methodologies are widely employed in additively manufactured materials in these days, the methods can be effectively employed to other problems dealing with metallic materials presenting defects. I believe this is the way to go for the identification and quantification of “hidden” fatigue influencing factors while ensuring the soundness of the prediction. 

The paper published in Journal of Materials & Design (JMAD - IF 9.4) can be found here:

ResearchGate: https://www.researchgate.net/publication/362851602_A_defect-based_Physi…

Science Direct (Open Access): https://www.sciencedirect.com/science/article/pii/S0264127522007110?via…