Semester of Graduation

Spring 2024

Degree

Master of Science in Computer Science (MSCS)

Department

Computer Science

Document Type

Thesis

Abstract

Following the current trends for minimizing human intervention in training intelligent architectures, this thesis proposes a self-supervised method for quality control of Additive Manufacturing (AM) parts. An Inconel 939 sample is fabricated with the Laser Powder Bed Fusion (L-PBF) method and scanned using X-ray Computed Tomography (XCT) to reveal the internal cracks. The presented workflow employs three modules that generate crack-like features for training a CycleGAN network within a self-supervised framework. By incorporating a combination of uniform and normal random variables, the method generates random cracks, exhibiting superior performance in fine-grain crack detection and capturing narrow tips compared to other approaches. Furthermore, a preliminary investigation into the training process reveals the algorithm's capability to predict crack propagation direction. The project introduces the IN939Crack dataset, a pioneering public dataset comprising typical internal features in Inconel 939 parts fabricated through L-PBF.

Date

1-25-2024

Committee Chair

Hao Wang

Available for download on Monday, January 25, 2027

Share

COinS