Turbulence Modeling Comparative Analysis For Vertical Axis Wind Turbines | ||||
Engineering Research Journal (Shoubra) | ||||
Volume 44, Issue 1, April 2020, Page 71-79 PDF (1.94 MB) | ||||
Document Type: Research articles | ||||
DOI: 10.21608/erjsh.2020.290027 | ||||
View on SCiNiTO | ||||
Authors | ||||
M. R. Rashed1, 2; O. E. Abdellatif2; M. F. Abd Rabbo2; E. E. Khalil3; I. Shahin2 | ||||
1Mechanical Engineering Dept., Faculty of Engineering-MTI University, Cairo, Egypt | ||||
2Mechanical Engineering Dept., Faculty of Engineering at Shoubra-Benha University, Cairo, Egypt | ||||
3Mechanical Power Engineering Dept., Faculty of Engineering-Cairo University, Giza, Egypt | ||||
Abstract | ||||
The impact of turbulence model selection on the accuracy of CFD simulations of VAWTs is highly considered. In this paper 2D CFD simulations were carried out on a three-bladed H-Darrieus VAWT with NACA 0021 airfoil. Two turbulence models, two equations k-ε Realizable and four equations Transition-SST, were selected to validate the power coefficient calculated from simulations with numerical and experimental data from the literature. The power coefficient calculated from the simulations performed by the k-ε Realizable turbulence model was slightly underestimated, while Transition-SST turbulence model results were slightly overestimated. However, in terms of root mean square error of the power coefficient calculated from the simulations performed by the k-ε Realizable turbulence model gives a value of 0.0503, while the Transition-SST model gives a value of 0.0655. Moreover, in terms of mean average percentage error of the power coefficient calculated from the simulations performed by the k-ε Realizable turbulence model gives a value of 35.15%, while the Transition-SST model gives a value of 42.34%. Another validation with a two straight bladed numerical model with NACA 0018 airfoil was performed using 2D CFD simulations. Two equations k-ε Realizable, two equations kω SST and four equations Transition-SST were selected to conduct these simulations, the results showed that the k-ε Realizable turbulence model was the best candidate among the other ones. | ||||
Statistics Article View: 41 PDF Download: 91 |
||||