Fault detection and localization with Neural Principal Component Analysis

Author

Khaled, O. ; Hedi, D. ; Lotfi, N. ; Messaoud, H. ; S-abazi, Zineb

Author_Institution

ATSI, Ecole Nat. d´´Ing. de Monastir, Monastir, Tunisia

fYear

2010

fDate

23-25 June 2010

Firstpage

880

Lastpage

885

Abstract

This paper presents a detection and diagnosis fault based on Neural Non Linear Principal Component Analysis (NNLPCA) and a Partial Least Square (PLS). This method is applied on a manufactured system, and the NNLPCA approach is used to estimate the non linear component. This NNLPCA model helps to estimate the prediction error and to define data classes with and without faults. The classes associated to data with faults are isolated by applying a PLS-2. Detecting faults is realized by SPE (square prediction error) statistics method, while locating them is realized by calculating contributions.

Keywords

fault diagnosis; least squares approximations; neural nets; principal component analysis; NNLPCA model; PLS; SPE statistics method; fault detection; fault diagnosis; manufactured system; neural nonlinear principal component analysis; partial least square; square prediction errors; Artificial neural networks; Covariance matrix; Data visualization; Manufacturing processes; Matrix decomposition; Principal component analysis; Training; Fault diagnosis; NIPALS algorithm; Neural Principal Component Analysis; PLS-2; Partial Least Square;

fLanguage

English

Publisher

ieee

Conference_Titel

Control & Automation (MED), 2010 18th Mediterranean Conference on

Conference_Location

Marrakech

Print_ISBN

978-1-4244-8091-3

Type

conf

DOI

10.1109/MED.2010.5547757

Filename

5547757