Mohamed Salim

A review on the flow structure and pollutant dispersion in urban street canyons for urban planning strategies

As a result of rapid urbanization in numerous cities around the world, the demand for transportation has increased rapidly, resulting in emission of high levels of exhaust pollutants into the atmosphere. This is a major cause of deterioration in the local air quality, with consequent escalating risk of adverse health conditions amongst urban inhabitants. Understanding dispersion of pollutants in street canyons, local urban configurations, meteorological processes, and other physical factors are essential for predicting and assessing air quality. This article presents a comprehensive review of the state-of-the-art research works relevant to the investigation of flow structures and pollutant dispersion phenomena in urban street canyons. Various factors, including building geometries, local atmospheric conditions, static and dynamic obstructions, as well as chemical reactions of exhaust pollutants, are critically discussed by taking into account field measurements, wind tunnel experiments, operational modeling techniques, and computational fluid dynamics (CFD). The most critical pollutant levels in street canyons under several physical circumstances are identified. Elements leading to discrepancies and resulting in inconsistencies of different research methods are briefly addressed and suggestions for future research are offered.

Wall y+ approach for dealing with turbulent flows over a wall mounted cube

An approach to dealing with turbulent flows over a wall mounted cube using the wall y+ (non-dimensional distance from the wall) as guidance in selecting the appropriate grid configuration and corresponding turbulence models is investigated using FLUENT 6.3. The study is divided into two sections – Case I and Case II, dealing with low and high Reynolds numbers, respectively. Five turbulence models – the standard k-e, standard k-ω, Reynolds Stress Model (RSM), Spalart-Allmaras (SA) and renormalisation group (RNG) k-e are used to solve the closure problem. Their behaviours, together with the accompanying near-wall treatments, are investigated for wall y+ covering the viscous sub-layer, the buffer layer and the log-law region.

Numerical Simulation of Wind Flow Structures and Pollutant Dispersion within Street Canyon under Thermally Unstable Atmospheric Conditions

Numerical studies were conducted to envisage the wind flow structures in street canyon for differential heated wall and Richardson numbers. Two-equation turbulence models, namely the Standard k-ε, Renormalization Group (RNG) k-ε and Realizable k-ε were applied to investigate the effect of flow structure on pollutant dispersion in a square street canyon. The obtained results demonstrate that the differentially heated wall/floor gives significant effects on the wind flow field compared with those under isothermal conditions. At low Richardson number, vortex intensification was observed for the case of ground or leeward heated wall. Meanwhile, for the case of windward heated wall, the ventilation was much reduced due to the buoyancy force that produces upward motion near the wall to form the secondary vortex. At higher Froude number (convective flows) case, the buoyancy flow has no discernible effects on the flow structures. Keywords— Street canyon; thermal flow; wind flow structures; pollutant dispersion; turbulence models

Performance of RANS, URANS and LES in the prediction of airflow and pollutant dispersion

The performance of 3 different CFD numerical approaches, namely RANS, URANS and LES are evaluated to determine their suitability in the prediction of airflow and pollutant dispersion in urban street canyons. Numerical results are evaluated against wind tunnel experimental data available from an online database (www.codasc.de). LES was observed to produce more accurate and reliable results compared to both RANS approaches, because LES resolves the inherent fluctuations, thus capturing the turbulent mixing process in the flow field within the canyon. Although URANS also computes for transience, it fails to account for unsteadiness, and hence is not an appropriate replacement for LES.

Numerical prediction of air flow within street canyons based on different two-equation k-ε models

Numerical simulations on airflow within street canyons were performed to investigate the effect of the street aspect ratio and wind speed on velocity profiles inside a street canyon. Three-dimensional Standard, Renormalization Group (RNG) and Realizable k-e turbulence model are employed using the commercial CFD code FLUENT to solve the Reynolds-averaged Navier-Stokes (RANS) equations. A comparison of the results from the presently adopted models with those previously published demonstrated that the k-e model is most reliable when simulating wind flow. The model is then employed to predict the flow structures in a street canyon for a range of aspect ratios (building height to street width ratio) between 0.5 – 2 at Reynolds number of 9000, 19200 and 30700 corresponding to the ambient wind speeds of 0.68m/s, 1.46m/s and 2.32m/s respectively. It is observed that the flow structure in the street canyon is influenced by the buildings aspect ratios and prevailing wind speeds. As the street aspect ratio increases, the air ventilation within the canyon reduces.

Numerical Simulation of the Aerodynamic Loads on Trees During Storms

Fluid-structure interactions for a single tree and a pair of trees with varying spacing subjected to gentle breeze and storm wind conditions were evaluated using Computational Fluid Dynamics (CFD). The generated velocity and pressure fields are then analysed using Finite Element Analysis (FEA) to determine the likelihood of tree damage due to the aerodynamic loads induced by the two wind conditions. It is observed that the pressure difference between the windward and leeward sides of the trees are much larger during the storm condition resulting in greater mechanical stresses and deformation magnitudes experienced by the tree trunks. Increasing the spacing between neighbouring trees resulted in larger aerodynamic loads on the sheltered trees downstream.

Numerical Validation for Wind Flow Field in a Street Canyon under Unstable Atmospheric Condition

This paper reports on the validation study of a commercial computational fluid dynamics program, ANSYS FLUENT v14. The purpose of the study was to determine the best turbulent model that can reproduce appropriately the wind flow field of turbulent flow in street canyon under different thermal atmospheric condition. Standard k-ε (SKE) and Large Eddy Simulation (LES) techniques were chosen as the turbulent model for the validation study. It was found that LES could well calculate a complex simulation involving isothermal flow and thermal flow with different thermal intensity and different heat location in street canyon.

Model sensitivity test of large eddy simulation for wind flow and pollutant dispersion in a street canyon

This paper reports on the model sensitivity analysis of a commercial computational fluid dynamics program, ANSYS FLUENT v14. The purpose of the analysis was to determine the appropriate modeling settings for numerical model of the case study. A full scale of a simplified urban street canyon was modelled and the turbulent flow was calculated using Large Eddy Simulation (LES) techniques. The model sensitivity tests involved are mesh sensitivity, statistically steady state and sampling. Adequate numbers of cells, period time to achieve statistically steady state (SST) and sampling time to simulate wind flow and pollutant dispersion in street canyon were determined through systematic tests.

Large Eddy Simulation's Parallel Processing Performance for Urban Air Pollution Studies Using FLUENT

The paper shares the experience of executing enormous computations by Large Eddy Simulation (LES) for environmental studies, using the commercial Computational Fluid Dynamics (CFD) code FLUENT 6.3® on an Intel Xeon® quad-core workstation and Intel Core® dual-core PC. The computational performancein terms of iteration resource and time expenditure for both single and parallel processingare assessed, with the later proving more efficient for large computational domainswhich arenecessary for simulations where both external and internal fluctuations dominate such in the case of flow and pollutant dispersion within urban street canyons.