Articles
COMPARISON OF THE ACCURACY OF DIFFERENT CFD TURBULENCE MODELS FOR THE PREDICTION OF THE CLIMATIC PARAMETERS IN A TUNNEL GREENHOUSE
Article number
719_32
Pages
287 – 294
Language
English
Abstract
This paper presents a comparative study of numerical simulations and experimental determination of dynamic and climatic parameters in a greenhouse.
The objective of this work is the determination of the accuracy of different turbulence models (k-ε Standard, k-ε RNG and k-ε Realizable) included in the Fluent® CFD package and used to predict velocity and climatic fields in large scale domains and for low characteristic velocity.
For that purpose simulations have been carried out for an empty tunnel-type greenhouse with lateral vents, located in Avignon, southern France.
The equations of flow are solved with the Fluent® CFD package inside and outside the greenhouse.
Several series of experimental measurements have been previously carried out using sonic anemometers and a comparative study between measured and calculated values at 48 nodal points of the greenhouse is presented.
It is shown that for the three turbulence models, errors on obtained results for temperature as well as humidity remain relatively weak.
However, accuracy on velocity modulus as well as on turbulent kinetic energy is more significant in high circulation areas than in other low ones. Neglecting the fluctuations induced by the instability of the measured boundary conditions (intensity and direction of the wind) and the intrusive nature of the measurement tools (sonic anemometers) the experimental data allow us to easily compare the three models.
For this tunnel greenhouse arrangement, the k-ε Standard turbulence model comes out to be the more accurate among the three considered ones.
The objective of this work is the determination of the accuracy of different turbulence models (k-ε Standard, k-ε RNG and k-ε Realizable) included in the Fluent® CFD package and used to predict velocity and climatic fields in large scale domains and for low characteristic velocity.
For that purpose simulations have been carried out for an empty tunnel-type greenhouse with lateral vents, located in Avignon, southern France.
The equations of flow are solved with the Fluent® CFD package inside and outside the greenhouse.
Several series of experimental measurements have been previously carried out using sonic anemometers and a comparative study between measured and calculated values at 48 nodal points of the greenhouse is presented.
It is shown that for the three turbulence models, errors on obtained results for temperature as well as humidity remain relatively weak.
However, accuracy on velocity modulus as well as on turbulent kinetic energy is more significant in high circulation areas than in other low ones. Neglecting the fluctuations induced by the instability of the measured boundary conditions (intensity and direction of the wind) and the intrusive nature of the measurement tools (sonic anemometers) the experimental data allow us to easily compare the three models.
For this tunnel greenhouse arrangement, the k-ε Standard turbulence model comes out to be the more accurate among the three considered ones.
Publication
Authors
R. Nebbali, J.C. Roy, T. Boulard, S. Makhlouf
Keywords
greenhouse, tunnel, ventilation, CFD, velocity, temperature, humidity, sonic anemometer
Online Articles (73)