A V-Model Framework to Support Qualification of Metal Additive Manufacturing Parts Made with Laser Powder Bed Fusion
Restricted (Penn State Only)
Author:
Roh, Byeong Min
Graduate Program:
Mechanical Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
August 09, 2022
Committee Members:
Daniel Connell Haworth, Program Head/Chair Soundar Kumara, Outside Unit & Field Member Timothy Simpson, Chair & Dissertation Advisor Allison Beese, Major Field Member Guhaprasanna Manogharan, Major Field Member Paul Witherell, Special Member
Keywords:
V-ModelOntology Sensors V-Model Quality Assurance Metal Additive Manufacturing Ontology Data-Driven Modeling In-Situ Monitoring Sensor Selection Printable Zone Ontology-Driven Physics-Informed Neural Network Monitoring and Quality Networks Complex Systems
Abstract:
This dissertation presents a novel framework for monitoring metal additive manufacturing (AM) processes for quality assurance using process signatures occurring during the laser powder bed fusion process. The framework offers a structured approach based on the Systems Engineering V-model to translate part requirements into process signatures that can be monitored using in situ sensing technologies. A method for selecting sensors based on the process signatures of interest is also developed to address a related shortcoming in the field. Information gathered from the selected sensors during the build is used to ascertain build quality with respect to the microstructure and mechanical properties of the resulting metal AM part. This information can then be used for quality assurance, validating that the requirements are met by the metal AM part. The V-model framework and sensor selection method for laser powder bed fusion are demonstrated and tested using existing Ti-6Al-4V and IN625 test data from the literature to validate traceability and requirements for density and ultimate tensile strength. The contributions of the work are discussed in this dissertation, along with limitations and future work.
