BROWSE
Detail
Conference Paper
A Data Mining Approach for Bearing Failure Prediction Using Multiple Non-linear Features
Cited - time in 
Share 
Share
- Authors
- Heon Gyu Lee, Hoon Jung
- Issue Date
- 2016-04
- Citation
- International Conference on Frontiers of Information Technology, Applications and Tools (FITAT) 2016, pp.1-6
- Language
- English
- Type
- Conference Paper
- Abstract
- The main objective of this paper is to suggest a novel method for fault diagnosis of bearing using data mining technique. We also develop and then propose a novel methodology useful in developing the various non-linear features helpful in diagnosing bearing condition. Various function-based prediction models are applied in order to detect and extract those which provide the better differentiation between normal and abnormal data. In our experiments, all non-linear features are used for constructing the diagnosis model. As a result, MDA method outperformed the other models.
- KSP Keywords
- Bearing failure, Data mining(DM), Diagnosis model, Failure prediction, Fault diagnosis, Non-linear features, abnormal data, data mining approach, data mining techniques, function-based, novel method