Fault Prognostics under Data Scarcity: Data Augmentation using Transfer Learning

Description

In this project we will develop methods for fault prognostics of process sensors. In particular, we will focus on the transferability and generalization of these methods for various types of process sensors in different field applications of these sensors and under diverse operatiive conditions. The methods combine physical models with data-driven approaches such as machine learning and deep leanring algorithms, and will be validated and tested on lab data as well as field data.