Abstract
A method based on signal analysis of incomplete cycles is studied for time reduction in flow regime recognition in pipeline-risers. Short signals of pressure and differential pressure in different stages of severe slugging are analyzed. Local flow characteristics are fused by SOM networks to describe the spatial feature of flow in the whole pipeline-riser system and distinguish the regime. Over 90% of stable and unstable regimes are correctly recognized within the velocity range of our experiment, thus the method is preliminary proven to be feasible.
| Original language | English |
|---|---|
| Pages (from-to) | 1966-1970 |
| Number of pages | 5 |
| Journal | Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics |
| Volume | 36 |
| Issue number | 9 |
| State | Published - 1 Sep 2015 |
Keywords
- Incomplete cycle
- Network
- Pipeline-riser
- Real-time recognition
- SOM neural
- Unstable flow regime
Fingerprint
Dive into the research topics of 'On-line flow regime identification in pipeline-risers based on signals of incomplete cycles'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver