Abstract

This paper presents a simple graphic method for detecting and classifying faults in point mechanisms based on the study of some statistical parameters of the force and current signals of the point machine. Principal Components Analysis (PCA) employed in order to reduce the number of these parameters. PCA is utilised in this paper for modifying the parameter dataset, and reducing the coordinate system by linear transformation. It is then possible to plot the new coordinate system in 2 or 3 dimensions, where the faults can be detected and identified. In this work most of the faults could be detected, but only a few experiments could be identified.

Details

Title
A New Method for Maintenance Management Employing Principal Component Analysis
Author
Fausto Pedro García Márquez
Pages
89-100
Section
ARTICLE
Publication year
2010
Publication date
2010
Publisher
Tech Science Press
ISSN
19302983
e-ISSN
19302991
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2397245562
Copyright
© 2010. This work is licensed under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.