Back to search

University of British Columbia

Integration of an AI-based self-correcting fused deposition modelling, composite sensor, and XR technology : an Industry 4.0 demonstration study

Abstract

dc:description

Additive manufacturing has become an essential part of the Industry 4.0 paradigm, but significant challenges remain in developing robust and automated process models that underpin the technology. In the first phase of this research, an AI-based robust method for an automatic stringing defect detection and its real-time self-correction during Fused Deposition Modelling (FDM) is proposed. The method is realized using a multi-stage’ progressive machine learning under a combination of a convolutional neural network (CNN) and an Extreme Gradient Boosting (XGB) decision tree classifier. The CNN is trained offline using camera images of observed stringing defects at various process conditions and during layer-by-layer material deposition, mapping the data onto three classes of print quality (low/medium/high). Using the output of this CNN and the process variables such as retraction rate, retraction distance, and nozzle temperature, the XGB classifier then identifies the optimal G-code that can yield low or no stringing in printing the new parts. Using real-time validation tests, the model demonstrated an accuracy of 97% in the print quality prediction, a specificity of 96%, as well as over 90% F-score, sensitivity, and precision.In the second phase of the research, it is aimed to combine the advantage of the above self-correcting FDM process with the emerging computer-mediated reality (XR) methods, and the Internet of Things (IoT), as an Industry 4.0 demonstration case study to safely monitor and control the temperature of manufactured parts in a simulated factory warehouse. To this end, a new Graphene- PLA sensor is 3D printed and calibrated, and subsequently, an XR mobile application is developed in Unity, to visualize and control the parts temperature in real-time. A user study is also conducted in the controlled lab environment, and the results showed that the XR mobile app clearly leads to faster (by 86.8%) and more accurate (97%) temperature control and monitoring compared to a conventional manual IR-camera application. The user experience (UX) was also significantly improved, as measured through a Holistic Presence Questionnaire (HPQ)

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Jayaraman Krishnamurthy, Rohith
Contributors dc:contributor
  • Milani, Abbas S.

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/84553

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Jayaraman Krishnamurthy, Rohith. Integration of an AI-based self-correcting fused deposition modelling, composite sensor, and XR technology : an Industry 4.0 demonstration study. master's thesis, University of British Columbia, 2023. http://hdl.handle.net/2429/84553