Data-Driven Gearbox Fault Severity Diagnosis Based on Concept Drift

Mario Peña, Laura Lanzarini, Mariela Cerrada, Diego Cabrera, René Vinicio Sánchez

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations


Condition-based maintenance aims to determine the machine state in real time, by monitoring the signals it emits. Such signals are potentially unlimited, generated at a high rate, and can evolve over time. These conditions tend to produce changes in the distribution of the data, known as concept drift. This phenomenon is analyzed and used to establish changes in the state of the machine. The present article proposes a methodological framework for the diagnosis of fault severity based on concept drift. A parsimonious unsupervised algorithm based on KNN is proposed to detect concept evolution. The results show that the algorithm is quite effective in declaring a concept evolution that is associated with a change in the failure condition of the machine. Finally, the results show that there is a high correlation between the displacement of the centroids of the emerging concepts and the % of deterioration of the machine.

Original languageEnglish
Title of host publicationETCM 2021 - 5th Ecuador Technical Chapters Meeting
EditorsMonica Karel Huerta, Sebastian Quevedo, Carlos Monsalve
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665441414
ISBN (Print)9781665441414
StatePublished - 12 Oct 2021
Event5th IEEE Ecuador Technical Chapters Meeting, ETCM 2021 - Cuenca, Ecuador
Duration: 12 Oct 202115 Oct 2021

Publication series

NameETCM 2021 - 5th Ecuador Technical Chapters Meeting


Conference5th IEEE Ecuador Technical Chapters Meeting, ETCM 2021

Bibliographical note

Funding Information:
ACKNOWLEDGMENT This paper is supported by the research group GIDTEC of the Universidad Politécnica Salesiana, sede Cuenca-Ecuador and the LIDI Computer Science Research Institute of the Universidad Nacional de La Plata.

Publisher Copyright:
© 2021 IEEE.


  • Concept drift
  • concept evolution
  • fault detection
  • Jensen-Shannon divergence
  • Kullback-Leibler divergence


Dive into the research topics of 'Data-Driven Gearbox Fault Severity Diagnosis Based on Concept Drift'. Together they form a unique fingerprint.

Cite this