ThermoPore: Predicting Part Porosity Based on Thermal Images Using Deep Learning

TL;DR

This paper introduces a deep learning framework that predicts and localizes porosity in additive manufacturing parts using thermal images, aiming to enable real-time quality assessment and reduce post-processing inspections.

Contribution

It develops a novel combination of CNN and Video Vision Transformer models for in-situ porosity detection and localization during the manufacturing process.

Findings

Porosity quantification model achieved R^2 of 0.57.

Localization model achieved IoU of 0.32 on average.

Framework supports real-time monitoring to facilitate digital twins.

Abstract

We present a deep learning approach for quantifying and localizing ex-situ porosity within Laser Powder Bed Fusion fabricated samples utilizing in-situ thermal image monitoring data. Our goal is to build the real time porosity map of parts based on thermal images acquired during the build. The quantification task builds upon the established Convolutional Neural Network model architecture to predict pore count and the localization task leverages the spatial and temporal attention mechanisms of the novel Video Vision Transformer model to indicate areas of expected porosity. Our model for porosity quantification achieved a $R^2$ score of 0.57 and our model for porosity localization produced an average IoU score of 0.32 and a maximum of 1.0. This work is setting the foundations of part porosity "Digital Twins" based on additive manufacturing monitoring data and can be applied downstream to reduce time-intensive post-inspection and testing activities during part qualification and certification. In addition, we seek to accelerate the acquisition of crucial insights normally only available through ex-situ part evaluation by means of machine learning analysis of in-situ process monitoring data.