A regularized deep clustering method for fault trend analysis

Yongzhi Qu, Yue Zhang, David He, Miao He, Zude Zhou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Effective fault feature extraction is the key of fault diagnosis. In previous works, it is shown that some embedding methods and unsupervised deep learning methods have the ability to extract fault features from raw signals directly, such as PCA and deep autoencoder. Particularly, deep autoencoder has been shown in relevant research that it can effectively extract the hidden 'trend' associated with machinery health states which can be used directly for online anomaly detection and prediction. However, in practical online fault diagnosis, the discrimination between successive signals is small due to the slow degradation progress and the external noise. Therefore, it is important to optimize the feature extraction process to achieve better online fault tracking. In this paper, a regularized deep clustering algorithm is proposed to guide the optimization process of feature extraction which combines embedding method and semi-guided learning. A regularization term for the cluster center points is proposed to make the feature optimization converge in a monotonic linear trend. In order to verify the effectiveness of the method, an accelerated gearbox run-to-failure experiment is carried out. The result shows that the feature optimization method can optimize the fault features on the basis of the deep autoencoder algorithm in two aspects: A better distinction of the fault features in short term and a more consistent trend of the gear wear in the long term.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual Conference of the Prognostics and Health Management Society, PHM
EditorsN. Scott Clements, Bin Zhang, Abhinav Saxena
PublisherPrognostics and Health Management Society
Edition1
ISBN (Electronic)9781936263059
DOIs
StatePublished - Sep 23 2019
Event11th Annual Conference of the Prognostics and Health Management Society, PHM 2019 - Scottsdale, United States
Duration: Sep 23 2019Sep 26 2019

Publication series

NameProceedings of the Annual Conference of the Prognostics and Health Management Society, PHM
Number1
Volume11
ISSN (Print)2325-0178

Conference

Conference11th Annual Conference of the Prognostics and Health Management Society, PHM 2019
CountryUnited States
CityScottsdale
Period9/23/199/26/19

Fingerprint Dive into the research topics of 'A regularized deep clustering method for fault trend analysis'. Together they form a unique fingerprint.

Cite this