The present study deals with natural convection flow in a vertical open-ended channel with wall constant heat flux. The experimental and numerical investigations are both conducted using water as the working fluid. Th...The present study deals with natural convection flow in a vertical open-ended channel with wall constant heat flux. The experimental and numerical investigations are both conducted using water as the working fluid. The numerical code is developed using finite differences scheme to solve the Navier-Stokes equations under the Boussinesq assumption. Concerning the experimental apparatus, it consists of two heated walls immersed in water. Temperature and velocity measurements are provided for different modified Rayleigh numbers based on the walls spacing b Rab = 1.67 x 10 6,3.6 x 10 6,8.97 x10 6,1.69 x 10 7,4.29 x 10 7. The numerical code is first validated with a numerical benchmark. Then, comparison between experimental and numerical results is performed. The code provides a satisfactory prediction of main quantities compared to the experimental results but only for the lowest Rayleigh numbers. For higher modified Rayleigh numbers, the flow becomes three-dimensional and turbulent. Therefore, 2D numerical simulations fail to predict flow and heat transfer for this range of modified Rayleigh number.展开更多
文摘The present study deals with natural convection flow in a vertical open-ended channel with wall constant heat flux. The experimental and numerical investigations are both conducted using water as the working fluid. The numerical code is developed using finite differences scheme to solve the Navier-Stokes equations under the Boussinesq assumption. Concerning the experimental apparatus, it consists of two heated walls immersed in water. Temperature and velocity measurements are provided for different modified Rayleigh numbers based on the walls spacing b Rab = 1.67 x 10 6,3.6 x 10 6,8.97 x10 6,1.69 x 10 7,4.29 x 10 7. The numerical code is first validated with a numerical benchmark. Then, comparison between experimental and numerical results is performed. The code provides a satisfactory prediction of main quantities compared to the experimental results but only for the lowest Rayleigh numbers. For higher modified Rayleigh numbers, the flow becomes three-dimensional and turbulent. Therefore, 2D numerical simulations fail to predict flow and heat transfer for this range of modified Rayleigh number.
