DMCA
Investigating Computational Geometry for Failure Prognostics in Presence of Imprecise Health Indicator: Results and Comparisons
| Citations: | 2 - 2 self |
BibTeX
@MISC{Ramasso_investigatingcomputational,
author = {Emmanuel Ramasso},
title = {Investigating Computational Geometry for Failure Prognostics in Presence of Imprecise Health Indicator: Results and Comparisons},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Prognostics and Health Management (PHM) is a multidisci-plinary field aiming at maintaining physical systems in their optimal functioning conditions. The system under study is assumed to be monitored by sensors from which are obtained measurements reflecting the system’s health state. A health index (HI) is estimated to feed a data-driven PHM solution developed to predict the remaining useful life (RUL). In this paper, the values taken by an HI are assumed imprecise (IHI). An IHI is interpreted as a planar figure called polygon and a case-based reasoning (CBR) approach is adapted to estimate the RUL. This adaptation makes use of computational geom-etry tools in order to estimate the nearest cases to a given testing instance. The proposed algorithm called RULCLIP-PER is assessed and compared on datasets generated by the NASA’s turbofan simulator (C-MAPSS) including the four turbofan testing datasets and the two testing datasets of the PHM’08 data challenge. These datasets represent 1360 test-ing instances and cover different realistic and difficult cases considering operating conditions and fault modes with un-known characteristics. The problem of feature selection, health index estimation, RUL fusion and ensembles are also tackled. The proposed algorithm is shown to be efficient with few pa-rameter tuning on all datasets. 1.
Keyphrases
computational geometry imprecise health indicator failure prognostic difficult case case-based reasoning health index estimation computational geom-etry tool planar figure system health state nasa turbofan simulator optimal functioning condition fault mode data challenge health management multidisci-plinary field un-known characteristic physical system rul fusion feature selection test-ing instance health index data-driven phm solution useful life
