Data-Driven Fault Diagnostics and Prognostics for Predictive Maintenance: A Brief Overview^*

摘要

Predictive Maintenance (PdM) is a maintenance strategy which predicts equipment failures before they occur and then performs maintenance in advance to avoid the occurrence of failures. A PdM system generally consists of four main components: data acquisition and preprocessing, fault diagnostics, fault prognostics and maintenance decision-making. Recently, massive condition monitoring data of equipment, also known as the industrial big data, has shown explosive growth. A large number of research works, including theoretical studies and industrial applications, have focused on implementing PdM with industrial big data analytics. This paper aims to provide a brief overview on the PdM system in the era of big data, with a particular emphasis on models, methods and algorithms of data-driven fault diagnostics and prognostics. In addition, a conclusion with a discussion on possible future trends in the research field of PdM is also given.