Return to search

Numerical study on flow and pollutant dispersion inside street canyons

This thesis analyzes the characteristics of flow pattern and vehicle-emitted pollutant dispersion in roughness surface layer. In an urban environment, wind flow and transported-pollutant source interfere strongly with buildings and other roughness elements on the surface ground, which results in complex characteristics of flow pattern and pollutant dispersion in 3D circumstances. The present study intends to simplify the research domain and investigate the fundamental modeling problems that exist in the field. The current physical research topic is restricted to 2D street canyon in equilibrium conditions. The study is motivated by the fact that characteristics of flow pattern and pollutant distribution inside street canyons are important for public health. The research has applied the computational fluid dynamics (CFD) methodology. To date, insights have typically focused on idealized street canyons without strictly limited boundary conditions and turbulence models. Those approaches face challenges related to their applicability to real urban scenarios or the reliability of prediction results. The thesis examines the influence of grid density, turbulence models and turbulent Schmidt number on pollutant distribution at windward and leeward surfaces of street canyon. Since numerical results usually are validated with wind-tunnel measurement data, the results between full-size model and wind-tunnel model are compared in order to test the Reynolds number effect. The lack of measurement data means that the morphometric method is used to generate upcoming wind profile, including the mean vertical velocity and turbulence parameters. The thesis also analyzes the potential errors brought by the method (Scenario A). Based on the evaluated numerical model, the thesis continues to study the impacts of surrounding buildings and geometry of street canyon on flow and pollutant distribution inside street canyons. The effect of wind on pollutant distribution inside street canyons was also investigated (Scenario A). Furthermore, the influence of roof shape and configuration of street canyon on characteristics of flow and pollutant distribution is also systematically studied, with the results shown in scenario B. The main conclusions of the thesis are that the uncertainty of numerical results derives from different aspects. Wind profile in the inlet profile generated by morphometric method brings major error to the simulation results. Current turbulence models cannot compromise the simulation results between flow field and pollutant distribution field. Ignored small-scale obstacles also need to be handled carefully. Numerical results revealed that flow and pollutant distribution inside street canyons are mainly dominated by the geometry of the street canyon itself. Medium-spaced surrounding buildings are also better able to transport pollutant out of the street canyon. Through systematic analysis, roof shape is proven to have a significant effect on flow and pollutant distribution inside a street canyon. The major impact is altered turbulence intensity depth and strength of shear layer inside the street canyon, which is important for pollutant removal process out of the street canyon. In the future, advanced turbulence models accompanied by small-obstacle effect models need to be developed in order to reliably simulate flow and pollutant dispersion simultaneously. Based on the advanced turbulence model, simulation of flow and pollutant dispersion in a complex 3D environment is essential in the next steps for the purpose of engineering application. Accurate vertical wind profile provided for inlet profile is another interesting direction for further development. Keywords: Flow; Pollutant dispersion; CFD; Street canyon; Reliability / <p>QC 20130215</p>

Identiferoai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-118327
Date January 2013
CreatorsYunkai, Yang
PublisherKTH, Installationsteknik, Stockholm
Source SetsDiVA Archive at Upsalla University
LanguageEnglish
Detected LanguageEnglish
TypeLicentiate thesis, monograph, info:eu-repo/semantics/masterThesis, text
Formatapplication/pdf
Rightsinfo:eu-repo/semantics/openAccess

Page generated in 0.0022 seconds