Principal Component Analysis in Used Oil Data for Condition Based Maintenance

Hussin , Burairah and Shibghatullah, Abdul Samad (2008) Principal Component Analysis in Used Oil Data for Condition Based Maintenance. In: In Proc Seminar on Data Mining & Optimization (DMO 08), 3-4 December 2008, UKM, Bangi, Malaysia..

[img]
Preview
PDF (ImageMagick conversion from application/pdf to application/pdf)
BurairahDMO-2008.pdf

Download (1MB)

Abstract

Thls paper reports on a study using the available oil monitoring lntormation that is ohtained from the Speeunmetric Oi] Analysis Programme (SOAP), to predict the residual life oi‘ a set of ship engines. The analysis oi‘ oil samples talaen from an engine git-es an indication of the suitability oi’ the oil for continued use and provides important information about the condition of the engine. This could allnw the ident.i.l‘ioation of wearing components heiore severe failure could occur without dismantling the engine. Given this condition-monitoring data, maintenance decisions may be taken as required and, moat importantly, maintenance may he done in an effective and eflicie-at way. This paper starts with some analysis. assumptions and techniques necessary to gain insight intn SOAP data that will he nseful to our modelling development. Several issues regarding the oonsistency, incompleteness and dimensions of the data used are discussed, which include the implementation oi‘ principal component analysis technique. This research proposed an approach called the ‘total metal concentrations’ ealculatjon, which is used to esplain the relationship hetvveen the residual life and the total wear concentrations which are avaiiahle irom SOAP data. Unce the ‘clean‘ data is attained, the next modelling steps can he eshlhiished tn rccnmmend the optimal maintenance actions in terms of cost, availability or any criterion of interest.

Item Type: Conference or Workshop Item (Paper)
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Information and Communication Technology > Department of Industrial Computing
Depositing User: Dr Abdul Samad Shibghatullah
Date Deposited: 05 Nov 2014 12:04
Last Modified: 28 May 2015 04:29
URI: http://eprints.utem.edu.my/id/eprint/13026
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item