Back to Search Start Over

Numerical Sensitivity Analysis for Supercritical CO 2 Radial Turbine Performance and Flow Field

Authors :
Teemu Turunen-Saaresti
Alireza Ameli
Jari Backman
Antti Uusitalo
Source :
Energy Procedia. 129:1117-1124
Publication Year :
2017
Publisher :
Elsevier BV, 2017.

Abstract

The predominant advantage of a supercritical CO2 (SCO2) Brayton cycle is it’s diminished compression work when compared to an ideal working fluid such as helium, due to low compressibility factor. For flows where the density approaches the critical density, molecular interactions get stronger and the ideal gas assumption is no longer appropriate. As a means to investigate the accuracy of real gas models and sensitivity of the performance of a turbomachine on the numerical accuracy in the supercritical region, numerous unsteady simulations of a radial turbine have been performed. Real Gas Properties (RGP) table has been generated to overcome difficulties of instabilities in simulations in the supercritical region. Four Equation of States (EOS) models with different RGP table resolutions have been studied and results are compared with the experimental measurement performed at the Sandia National Laboratory. The studied unshrouded radial impeller has 11 blades while the nozzle has 10 vanes. The unsteady simulation results show the dependency of the predicted radial turbine performance on the RGP table resolution as well as on the implemented EOS models. In general, the results indicate that by using appropriate EOS model and a look up table with high resolution, the CFD can predict turbine performance with high accuracy while the use of some EOS and low resolution look up tables lead to significant deviations between the simulated and measured results. As a result of this study, an appropriate EOS model and sufficient resolution of the look-up table for turbines operating at supercritical region are suggested.

Details

ISSN :
18766102
Volume :
129
Database :
OpenAIRE
Journal :
Energy Procedia
Accession number :
edsair.doi...........dde128f48bffd6d1ea2914c5fbb3edc6