Abstract
In this paper, six different SA models and their combinations with the QCR2000 and QCR2013 methods are used for simulating the low-speed hump flow. Referring to the experiment and LES data, the performance of these turbulence models in predicting two-dimensional separated flow is evaluated. It is found that all SA models over predict the separation size thus the recovery of axial velocity loss is delayed. Also, the peak value of Reynolds shear stress is largely underestimated. Although the separation structure given by tested models basically keeps the same, the original SA model fits the best with the experiment in terms of separation size among them. After applying QCR corrections to the linear eddy viscosity models, the length of separation region gets even larger while the Reynolds shear stress is simulated more accurately.
| Original language | English |
|---|---|
| Pages (from-to) | 865-874 |
| Number of pages | 10 |
| Journal | Kung Cheng Je Wu Li Hsueh Pao/Journal of Engineering Thermophysics |
| Volume | 42 |
| Issue number | 4 |
| State | Published - Apr 2021 |
Keywords
- Hump flow
- SA turbulence model
- Two-dimensional separation
Fingerprint
Dive into the research topics of 'Performance of SA Model With Various Modifications in Predicting Two-Dimensional Separated Flow'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver