In-situ Anomaly Detection in Additive Manufacturing with Graph Neural Networks

Sebastian Larsen, Paul A. Hooper

Keywords: Additive manufacturing, anomaly detection, geometric deep learning, graph neural networks, machine learning, materials science

TL;DR: A GNN was trained to predict features from high-speed video observations of laser melting based on laser input. The output prediction error then acted as an anomaly metric.

Abstract: Transforming a design into a high-quality product is a challenge in metal additive manufacturing due to rare events which can cause defects to form. Detecting these events in-situ could, however, reduce inspection costs, enable corrective action, and is the first step towards a future of tailored material properties. In this study a model is trained on laser input information to predict nominal laser melting conditions. An anomaly score is then calculated by taking the difference between the predictions and new observations. The model is evaluated on a dataset with known defects achieving an F1 score of 0.821. This study shows that anomaly detection methods are an important tool in developing robust defect detection methods.