Advances in deep learning and IIoT toward industry-scale machine health monitoring
Date
2023-05
Authors
Lu, Hao
Major Professor
Advisor
Hu, Chao
Laflamme, Simon
Ramamoorthy, Aditya
Li, Beiwen
Darr, Matthew
Sarkar, Soumik
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In modern manufacturing systems, the increased availability of sensor data is contributing significantly to the enhancement of the health monitoring field. Intelligent fault diagnosis and prognosis in health monitoring have been remarkably improved with the introduction of advanced deep learning approaches; however, several concerns still hinder the full implementation of deep learning health monitoring algorithms in real-world applications. The developed algorithms' ability to generalize over diverse operating conditions is still unclear. In addition, the data-driven model's performance is directly related to the quality and quantity of training datasets. However, it can be costly to gather sufficient data for model training. And data privacy is a significant concern when multiple agencies/systems are connected via these algorithms. This dissertation focuses on developing new methods and algorithms that address some of the abovementioned concerns and could be used for industry-scale machine health monitoring. The research contributions of this dissertation are in the areas of post-design fault diagnostics and prognostics, which focuses on applications within rotating machinery.
Series Number
Journal Issue
Is Version Of
Versions
Series
Academic or Administrative Unit
Type
dissertation