Back to search

Georgia Institute of Technology

Federated learning for process monitoring in directed energy deposition

Abstract

dc:description.abstract

The usage of machine learning for process monitoring of additive manufacturing processes is becoming more popular. Researchers have developed and applied various machine learning methods to improve the efficiency and quality of the additive manufacturing process. A research domain in machine learning of current interest is federated learning, which is a type of model training, where the local training data remain on different devices without being exposed to other devices. Federated learning has been developed and applied by researchers in areas such as telecommunication, pharmaceutics and IoT. This research work focuses on federated learning for process monitoring in directed energy deposition. Several deep learning models with varying key parameters such as number of clients and number of communication rounds are trained using the federated learning approach. Those models are then compared to the centralized approach to examine the impact of federated learning on training time, performance metrics and communication cost.

Degree

thesis:*
Level thesis:degree_level
Masters
Department dc:contributor.department
Mechanical Engineering
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2021

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Krug, Fabian L.
Advisor dc:contributor.advisor
  • Saldaña, Christopher J.
Committee members dc:contributor.committeemember
  • Fu, Katherine
  • Tarin, Cristina
  • Sawodny, Oliver

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
http://hdl.handle.net/1853/70056

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Krug, Fabian L.. Federated learning for process monitoring in directed energy deposition. Masters thesis, Georgia Institute of Technology, 2021. http://hdl.handle.net/1853/70056