Teachers Login Form



Pupils Login Form


Fault Detection using Principal Component Analysisbased on Derived Variables

Auteurs : Hocine BENDJAMA, Kaddour Gherfi, Daoud Idiou, Jürgen Bast
Année : 2016
Domaine : Automatique
Type : Communication
Conférence: International Conference On Electrical Sciences and Technologies
Résumé en PDF : (résumé en pdf)
Fulltext en PDF :
Mots clés : fault detection; principal component analysis; casting process; squared prediction error

Résumé :

This paper presents a computational method forcasting faults detection. The proposed method is based onPrincipal Component Analysis (PCA) and it is applied to lowpressure lost foam casting process. To enhance the PCA matrixand improve early detection, we use the derived variables in theinput matrix to establish the statistical correlation among themeasured data in order to detect the abnormal situations and toprovide information about the process state. The obtained resultsdemonstrate the potential of the proposed fault detectionapproach.