An application of machine learning approach to fault detection of a synchronous machine
Authors:
- Jose Gregorio Ferreira,
- Adam Warzecha
Abstract
The paper focuses on experimental procedures to develop a multi-criteria methodology to classify up to ten machine conditions. Using machine learning for signal processing techniques, any deviation from a normal steady state might be categorized as an abnormal behavior and, when demonstrated, a fault. To demonstrate the procedure the authors examined a synchronous machine. The authors recorded currents and voltages primarily, in stator and rotor winding, well as rotational speed and electromechanical torque. The collected signals were filtered and pre-processed, and to 5038 features were calculated and transformed into a tidy dataset. The sparse Linear Discriminant Analysis algorithm was used to extract the most important defined features. The results are shown in 3D scatter plots; in which each machine condition is represented
- Record ID
- CUT074d474748a94349b46e25700f1bf1a3
- Publication categories
- ; ;
- Author
- Pages
- [1-6]
- Other elements of collation
- rys.; tab.; wykr.; Bibliografia (na s.) - [6]; Bibliografia (liczba pozycji) - 18; Oznaczenie streszczenia - Abstr.
- Book
- 2017 International Symposium on Electrical Machines (SME), Naleczow, Poland, 18-21 June 2017, 2017, [S.l.], Institute of Electrical and Electronics Engineers, IEEE, [6] p., ISBN 978-1-5386-0359-8
- Keywords in English
- machine learning, synchronous machine, classificatio, fault diagnostics
- DOI
- DOI:10.1109/ISEM.2017.7993548 Opening in a new tab
- URL
- http://ieeexplore.ieee.org/abstract/document/7993548/ Opening in a new tab
- Related project
- Metody i algorytmy monitoringu i diagnostyki maszyn elektrycznych. . Project leader at PK: , ,
Działalność statutowa - Language
- eng (en) English
- Score (nominal)
- 20
- Additional fields
- Indeksowana w: Web of Science, Scopus
- Uniform Resource Identifier
- https://cris.pk.edu.pl/info/article/CUT074d474748a94349b46e25700f1bf1a3/
- URN
urn:pkr-prod:CUT074d474748a94349b46e25700f1bf1a3
* presented citation count is obtained through Internet information analysis, and it is close to the number calculated by the Publish or PerishOpening in a new tab system.