A statistical comparison of neuroclassifiers and feature selection methods for gearbox fault diagnosis under realistic conditions

Fannia Pacheco, José Valente de Oliveira, René Vinicio Sánchez, Mariela Cerrada, Diego Cabrera, Chuan Li, Grover Zurita, Mariano Artés

Research output: Contribution to journalArticlepeer-review

46 Scopus citations

Abstract

Gearboxes are crucial devices in rotating power transmission systems with applications in a variety of industries. Gearbox faults can cause catastrophic physical consequences, long equipment downtimes, and severe production costs. Several artificial neural networks, learning algorithms, and feature selection methods have been used in the diagnosis of the gearbox healthy state. Given a specific gearbox, this study investigates how these approaches compare with each other in terms of the typical fault classification accuracy but also in terms of the area under curve (AUC), where the curve refers to the precision-recall curve otherwise known as receiver operating characteristic (ROC) curve. In particular, the comparison aims at identifying whether there are statistically significant (dis)similarities among six feature selection methods, and seven pairs of neural nets with different learning rules. Genetic algorithm based, entropy based, linear discriminants, principal components, most neighbors first, and non-negative matrix factorization are the studied feature selection methods. Feed forward perceptrons, cascade forward, probabilistic nets, and radial basis function neural nets are evaluated. Six supervised and one unsupervised learning rules are considered. Both parametric and nonparametric statistical tests are employed. A ranking process is defined to elect the best approach, when available. An experimental setup was especially prepared to ensure operating conditions as realistic as possible.

Original languageEnglish
Pages (from-to)192-206
Number of pages15
JournalNeurocomputing
Volume194
DOIs
StatePublished - 19 Jun 2016

Bibliographical note

Funding Information:
The work was sponsored in part by the GIDTEC project called “Desarrollo de una herramienta computacional basada en modelos de computación inteligente para el monitoreo en maquinaria rotativa” No. 017-007-2015-11-05 , and the Prometeo Project of the Secretariat for Higher Education, Science, Technology and Innovation (SENESCYT) of the Republic of Ecuador . The experimental work was developed at the GIDTEC research group lab of the Universidad Politécnica Salesiana de Cuenca, Ecuador .

Publisher Copyright:
© 2016 Elsevier B.V.

Keywords

  • Classification
  • Fault diagnosis
  • Feature selection
  • Gearbox
  • Neural networks
  • Statistic tests

Fingerprint

Dive into the research topics of 'A statistical comparison of neuroclassifiers and feature selection methods for gearbox fault diagnosis under realistic conditions'. Together they form a unique fingerprint.

Cite this