Simulating wind-driven rain on building facades using Eulerian multiphase with rain phase turbulence model
Artikel i vetenskaplig tidskrift, 2016
Wind-driven rain (WDR) is responsible for many types of damage to building façades and contributes to storm water management problems in urban environments. Consequently there is significant interest in accurately predicting WDR using computational fluid dynamics (CFD) simulations. In this paper an Eulerian multiphase (EM) method is proposed which includes applying a turbulence model for solving the rain phase equation closure rather than a response coefficient to approximate turbulent behaviour. The simulations are conducted using AVL FIRE™ with standard k−ε model and are validated against experimental data for two cases. The results produce an error ranging from 5% to 39% in the first case and from 9% to 72% in the second case. The discrepancies are attributed largely to the unsuitability of the standard k−ε in predicting the flow accurately as well as the limited number of raindrop phases. While the results are not as accurate as other research has shown, the method described in this paper allows for greater flexibility when working with transient rainfall conditions and gives an alternative to the previously proposed response coefficient for modelling turbulent dispersion. The method is viable for calculating general WDR distribution patterns on a façade and can be improved to offer more accurate results.
Standard k−ε model
Wind-driven rain (WDR)
Eulerian multiphase (EM)
Computational fluid dynamics (CFD)