Hybrid data augmentation method for combined failure recognition in rotating machines

Dionísio H. C. S. S. Martins, Amaro A. Lima (), Milena F. Pinto (), Douglas de O. Hemerly, Thiago de M. Prego (), Fabrício L. e Silva (), Luís Tarrataca (), Ulisses A. Monteiro (), Ricardo H. R. Gutiérrez () and Diego B. Haddad ()
Additional contact information
Dionísio H. C. S. S. Martins: Federal Center for Technological Education of Rio de Janeiro
Amaro A. Lima: Federal Center for Technological Education of Rio de Janeiro
Milena F. Pinto: Federal Center for Technological Education of Rio de Janeiro
Douglas de O. Hemerly: International Business Machines Corporation
Thiago de M. Prego: Federal Center for Technological Education of Rio de Janeiro
Fabrício L. e Silva: Federal Center for Technological Education of Rio de Janeiro
Luís Tarrataca: Federal Center for Technological Education of Rio de Janeiro
Ulisses A. Monteiro: Federal University of Rio de Janeiro
Ricardo H. R. Gutiérrez: Federal University of Rio de Janeiro
Diego B. Haddad: Federal Center for Technological Education of Rio de Janeiro

Journal of Intelligent Manufacturing, 2023, vol. 34, issue 4, No 17, 1795-1813

Abstract: Abstract Rotating machines are frequently subject to a wide range of rough conditions, resulting in mechanical failures and performance degradation. Thus, it is important to apply proper failure detection and recognition techniques, such as machine learning algorithms, to prevent these issues early. In industrial environments, little data exists regarding failure conditions, which hinders the training stage of the classification algorithms responsible for classifying the failures. Therefore, this work proposes a hybrid method of data augmentation to increase the number of minority class instances in order to improve classifier performance. The approach combines the synthetic minority over-sampling and the additive white Gaussian noise techniques to create a set of artificial signals. The results show that the proposal is able to achieve better results than applying those techniques separately and also when using an undersampling strategy. For comparison purposes, four machine learning classification methods were analyzed alongside our data augmentation proposal, namely, support vector machines, K-nearest neighbors, random forest and stacked sparse autoencoder. The proposed hybrid data augmentation method associated with stacked sparse autoencoder outperformed the other models obtaining an accuracy of 100% and a processing time of 0.13 s.

Keywords: Data augmentation; Combined failures recognition; Imbalance; Misalignment; Rotating machines; Predictive maintenance (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s10845-021-01873-1 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:joinma:v:34:y:2023:i:4:d:10.1007_s10845-021-01873-1

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10845

DOI: 10.1007/s10845-021-01873-1

Access Statistics for this article

Journal of Intelligent Manufacturing is currently edited by Andrew Kusiak

More articles in Journal of Intelligent Manufacturing from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().