Publication Date
Spring 5-11-2024
Abstract
Understanding the reason for mechanical failures of manufactured parts in their operating environments is critical to prevention of future failures. However, in-situ post-mortem evaluation of physical properties, such as fracture toughness, is time consuming and alters the condition of the material, leading to potentially misleading findings. In this study, additively manufactured test coupons were produced over a wide range of process conditions to test the impact toughness of a material. The Charpy V-Notch toughness was measured on over 200 samples alongside corresponding optical images of both sides of the fracture surface. Convolutional neural network models were trained to correlate fracture toughness values and fractographic images. The models predicted the Charpy toughness for fractographic images spanning the range of the Charpy tester, with a best mean absolute percent error of 8.5%, while also identifying interpretable physical characteristics associated with changes in toughness, such as porosity, shear lips, and fracture surface edges.
Degree Name
Mathematics
Level of Degree
Masters
Department Name
Mathematics & Statistics
First Committee Member (Chair)
Jacob Schroder
Second Committee Member
Mohammad Motamed
Third Committee Member
Brad Boyce
Keywords
Charpy test, convolutional neural networks, machine learning, interpretability
Document Type
Thesis
Recommended Citation
Bianco, Nathan R.. "Robust Prediction of Charpy Toughness of Additively Manufactured Kovar Using Deep Convolutional Neural Networks." (2024). https://digitalrepository.unm.edu/math_etds/203
