RUL-RVE: Interpretable assessment of Remaining Useful Life

Abstract

This paper presents RUL-RVE, a Python tool for the assessment of Remaining Useful Life (RUL). Physical systems are normally subject to degradations that ultimately lead to failure, therefore prognostic technologies are crucial to estimate the lifetime of the system to be monitored. The problem with most existing data-driven approaches is that they lack an explanatory component to understand model learning and/or the nature of the data. RUL-RVE is a framework based on recurrent neural networks and variational inference that can achieve remarkable forecast accuracy while providing an interpretable assessment, which is highly valuable in real-world environments.

Publication
In Software Impacts
Nahuel Costa, PhD
Nahuel Costa, PhD
Machine Learning researcher 🤖 Assistant professor