A Plume On A Section-Triangular Roof

As one of the most ancient forms of roof, slope roof is widely applied in residential construction due to its variety of visual perception, better structure in spanning capacity, and strong ability in heat insulation and ventilation. Under the effect of solar radiation, the slope roof transports the hot air near the roof to the distance in the form of slope flow and plume. The slope flow and plume are easily found in the natural environment and industrial system, so the research on slope flow and plume formed on slope roof not only contributes to a better understanding of roofing heat transfer but also improve the living comfort. It is also of significance to solve the problem of urban pollution and predict climate in mountainous areas.In order to study the flow and heat transfer mechanism in the process of formation and subsequent development of the roof plume, scaling analysis and numerical simulation were adopted. Dynamics and heat transfer of the plume on the roof have been argued. Scaling relations under different dominances have been obtained and validated by numerical results. There exist good agreement between scaling predictions and numerical results.Further numerical results show that dependent on dimensionless governing parameters (such as Rayleigh number, aspect ratio and Prandt1 number), the development of the plume on the roof may be classified into three stages:an initial stage, a transitional stage and a fully developed stage. With the increase of Rayleigh number, the convective flow of the plume is enhanced. The plume is dominated by conduction and there is no distinct plume structure for the Rayleigh number below 102. The plume may finally approach a steady or a unsteady state in the fully developed stage. The plume is also dependent on the aspect ratio; that is, the plume is steady for large aspect ratios but unsteady for small aspect ratios for the Rayleigh number is 103. Additionally, the dependence of the plume on the Prandtl number is complex. The dimensionless governing parameters also determine heat and mass transfer of the plume. Numerical results show that heat transfer of the plume may approach the maximum for large Rayleigh number and there is an approximation of Nu-Ra0.2275. Further, the dependence of the flow rate on the Rayleigh number may be quantified by Q～Ra-0.25. The numerical results are consistent with the scaling predictions.Although flow patterns and heat transfer of the plume on the roof under the different dimensionless parameters had been observed and discussed in this study, flow features and heat transfer efficiency dependent on main governing parameters have been quantified. Thus, it is still necessary to perform further experiments and numerical simulations for further understanding of the mechanisms.