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

TL;DR

This paper presents a graph neural network-based in-situ anomaly detection method for metal additive manufacturing, predicting laser conditions and identifying defects to improve quality control and reduce inspection costs.

Contribution

It introduces a novel anomaly detection approach using GNNs trained on laser input data to identify defects in real-time during manufacturing.

Findings

Achieved an F1 score of 0.821 in defect detection.

Demonstrated the effectiveness of anomaly scores based on prediction deviations.

Highlighted the importance of anomaly detection in robust defect identification.

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.