MDL and Signal Change Detection in Machine Condition Monitoring
A minimum description length (MDL) based sequentially normalized maximum likelihood (SNML) approach combined with an autoregressive (AR) model is proposed for signal change detection in machine condition monitoring. The results showed the success of the method to detect signal changes and distinguish different ball bearing failures.
HULKKONEN Jenni;
HEIKKONEN Jukka;
2009-01-14
Tampere International Center for Signal Processing (TICSP)
JRC49471
http://sp.cs.tut.fi/WITMSE08/Proceedings/index.html,
http://sp.cs.tut.fi/WITMSE08/Proceedings/Posters/poster_Hulkkonen.pdf,
https://publications.jrc.ec.europa.eu/repository/handle/JRC49471,
Additional supporting files
| File name | Description | File type | |
|---|
