Term of Award

Spring 2024

Degree Name

Master of Science, Mechanical Engineering

Document Type and Release Option

Thesis (open access)

Copyright Statement / License for Reuse

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Department

Department of Mechanical Engineering

Committee Chair

Hossain Ahmed

Committee Member 1

Haijun Gong

Committee Member 2

Mujibur Khan

Committee Member 3

Sourav Banerjee

Abstract

This study presents a novel approach to overcoming process reliability challenges in Material Extrusion (ME), a prominent additive manufacturing (AM) technique. Despite ME's advantages in cost, versatility, and rapid prototyping, it faces significant barriers to commercial-scale production, primarily due to quality issues such as overextrusion and underextrusion, which compromise final part performance. Traditional manual monitoring methods severely lack the capability to efficiently detect these defects and highlight the necessity for an efficient and real-time monitoring solution. Considering these challenges, an innovative and field-deployable infrared thermography-based in-situ real-time defect detection and feedback control system is proposed in this thesis. A novel experimental setup and data acquisition framework are developed and automated to collect consistent thermal data during the extrusion process. The study elaborates on the preprocessing of this data for deep learning model training, including sub-segmentation and temporal trimming. Six deep learning architectures are trained and evaluated for their ability to detect defects in real-time. One classification model incorporating CNN-SelfAttention architecture is selected for online implementation based on its performance. The proposed closed-loop control system demonstrates the dynamic adjustment of the relative flow rate within 3 seconds of process deviation. The developed methodology ensures consistent production of high-quality parts by reducing percentage volumetric change by factors of 10. This approach not only ensures the production of defect-free parts within tolerable ranges but also sets a foundation for future research in automated monitoring systems for additive manufacturing. The findings and methodologies detailed in this thesis contribute to the advancement of AM and offer a path toward overcoming one of its major barriers to commercial and industrial application.

OCLC Number

1432737146

Research Data and Supplementary Material

No

Download

Share

COinS