Prediction of impeller rotating stall onset using numerical simulations of a centrifugal compressor. Part 1: Detection of rotating stall using fixed-flow transient simulations
Rotating stall is an unsteady flow phenomenon that appears in both axial and centrifugal compressors. It is detrimental to the performance of the compressor, significantly narrowing its operating range. Numerical modeling of this phenomenon has been a major area of investigation for axial compressor...
Gespeichert in:
Veröffentlicht in: | Proceedings of the Institution of Mechanical Engineers. Part A, Journal of power and energy Journal of power and energy, 2013-06, Vol.227 (4), p.403-414 |
---|---|
Hauptverfasser: | , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Rotating stall is an unsteady flow phenomenon that appears in both axial and centrifugal compressors. It is detrimental to the performance of the compressor, significantly narrowing its operating range. Numerical modeling of this phenomenon has been a major area of investigation for axial compressors with some success. While stall occurs less often in centrifugal compressors than axial ones, it can be much harder to predict. Some preventive measures are known but are mostly rules of thumb developed through experimental experience. This work focuses on the detection of impeller rotating stall using computational fluid dynamics. A compressor was chosen that has demonstrated rotating stall instabilities with different diffuser lengths and return vanes. Unsteady numerical simulations were performed on full 360° models of this geometry. The transient simulations were conducted using distributed processing on high-performance servers. The objective was to determine the accuracy in which rotating stall can be captured in simulations. For this purpose, simulation results were compared to experimental results for the same compressor and show a good correlation between the experimental and numerical tests. |
---|---|
ISSN: | 0957-6509 2041-2967 |
DOI: | 10.1177/0957650912474386 |