Abstract
In this paper, deep belief networks were used for localization of acoustic emission events on ship hull structures. In order to avoid complex and time consuming implementations, the proposed approach uses a simple feature extraction module, which significantly reduces the extremely high dimensionality of the raw signals/data. In simulation experiments, where a stiffened plate model was partially sunk into the water, the localization rate of acoustic emission events in a noise-free environment is greater than 94 %, using only a single sensor.
Original language | English |
---|---|
Title of host publication | Vibroengineering PROCEDIA |
Publisher | JVE International Ltd. |
Publication date | 2016 |
Pages | 56‑61 |
Publication status | Published - 2016 |
Event | 23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures - Istanbul, Turkey Duration: 7. Oct 2016 → 9. Oct 2016 Conference number: 23 http://www.jveconferences.com/about-istanbul-conference |
Conference
Conference | 23rd International Conference of Vibroengineering on Modeling, Identification and Fault Detection in Oil and Gas Equipment and Infrastructures |
---|---|
Number | 23 |
Country/Territory | Turkey |
City | Istanbul |
Period | 07/10/2016 → 09/10/2016 |
Internet address |
Series | Vibroengineering Procedia |
---|---|
Volume | 9 |
ISSN | 2345-0533 |
Keywords
- Acoustic emission
- Deep belief networks
- Deep learning
- Ship hull
- Source localizations