Fluid flow analysis behind heliostat using LES and RANS: A step towards optimized field design in desert regions

Abstract

Heliostats are used for concentrating beam radiation onto a receiver. The flow induced dust deposition on these reflectors will lead to failure of the receiver. For this purpose, the wake behind a heliostat is analyzed at 25° of inclination and at a Reynolds number of 60000. In this paper the Reynolds Averaged Navier-Stokes (RANS) and the Large Eddy Simulation (LES) approaches are used for analyzing the air-flow behind a heliostat. LES and RANS are performed with a wall-resolved grid. For the purpose of validation, the horizontal velocity is measured in a wind-tunnel with a model heliostat using laser Doppler velocimetry technique. RANS and LES approaches are found to qualitatively predict the statistical quantities, like the mean horizontal-velocity in comparison to experiment. RANS under-predicts root-mean-square of the horizontal-velocity and even failed to capture the flow features behind heliostat. Thus, it is concluded that RANS will suffice with well-resolved grid for analyzing mean flow features. For analyzing wake and to understand the induced dust deposition LES is required. Further, the analysis reveals that the wake-affected region is up to three times the length of the heliostat’s mirror. This can be recommended as the minimum distance between any two aligned heliostats in Jodhpur.Heliostats are used for concentrating beam radiation onto a receiver. The flow induced dust deposition on these reflectors will lead to failure of the receiver. For this purpose, the wake behind a heliostat is analyzed at 25° of inclination and at a Reynolds number of 60000. In this paper the Reynolds Averaged Navier-Stokes (RANS) and the Large Eddy Simulation (LES) approaches are used for analyzing the air-flow behind a heliostat. LES and RANS are performed with a wall-resolved grid. For the purpose of validation, the horizontal velocity is measured in a wind-tunnel with a model heliostat using laser Doppler velocimetry technique. RANS and LES approaches are found to qualitatively predict the statistical quantities, like the mean horizontal-velocity in comparison to experiment. RANS under-predicts root-mean-square of the horizontal-velocity and even failed to capture the flow features behind heliostat. Thus, it is concluded that RANS will suffice with well-resolved grid for analyzing mean flow features. ...