The doctoral course “Advanced methodologies for the assesment of the fatigue response” given to doctoral students at Politecnico di Torino covered the application of Machine Learning models in the field of fatigue of metals. I have had the pleasure to teach the class about “Fatigue & Machine Learning”, where I could give an overview of machine learning methods for science has been given.

In this jupyter notebook you will find an introduction to Physics-Informed Neural Networks (PINN) with application to a toy problem. Following, a dataset of fatigue experiments conducted on Additively Manufactured metals is provided to exercise with NN and PINN.