THESIS
2016
xxx, 373 pages : illustrations (some color), maps (1 color) ; 30 cm
Abstract
Two main wind-related problems in Hong Kong are originated from high-speed winds of
typhoons and undesirable low-wind speeds at the ground level due to congested urban
morphology. High wind speeds of typhoons damage structures severely while low wind
speeds responsible for a number of environmental issues, including poor air quality, outdoor
thermal discomfort and increase of respiratory diseases among citizens. However, both wind
speeds are substantially modified by the surrounding complex terrains and the congested
urban morphology with numerous closely-spaced high-rise buildings commonly found in
Hong Kong. Therefore, the main focus of this study is to investigate the characteristics of
high and low speed wind climates in Hong Kong and to evaluate the influence of local terra...[
Read more ]
Two main wind-related problems in Hong Kong are originated from high-speed winds of
typhoons and undesirable low-wind speeds at the ground level due to congested urban
morphology. High wind speeds of typhoons damage structures severely while low wind
speeds responsible for a number of environmental issues, including poor air quality, outdoor
thermal discomfort and increase of respiratory diseases among citizens. However, both wind
speeds are substantially modified by the surrounding complex terrains and the congested
urban morphology with numerous closely-spaced high-rise buildings commonly found in
Hong Kong. Therefore, the main focus of this study is to investigate the characteristics of
high and low speed wind climates in Hong Kong and to evaluate the influence of local terrain
on them.
The characteristics of typhoon wind were evaluated by constructing mean wind speed and
turbulence intensity profiles using Doppler-SODAR and Wind Profiler measurements taken
at the Sui Ho Wan (SHW) weather station. It was found that the low-level jet observed by
previous researchers was absence in mean wind speed observations and calculated roughness
lengths are unrealistically large for the relatively long sea fetch. The data fitting process
revealed that the log-law and the power-law models are adequate to estimate the vertical
mean wind speed variation below 300 m height oppose to the use of modified Vickery model.
The vertical profile of turbulence intensity was calculated from the backscattered signals of
Doppler-SODAR by using the EDR method. The comparison of calculated turbulence
intensity with direct turbulent wind measurements concluded that the proposed method can
adequately predict the vertical variation of turbulence intensity in the tropical cyclone
boundary layer adequately.
The topography induced horizontal flow direction deviation or the yaw angle variation was
studied by using five 3-D isolated hill models with aspect ratios of 1/3, 1/2, 1, 2, 3. Measured
yaw angle near the surface was higher on the upwind hill slope than on the downwind hill
slope, which contradicts with the ESDU model assumption of higher yaw angle on the
downwind hill slope. Nevertheless, the ESDU model could predict the vertical yaw angle
variation accurately in despite the large deviations of calculated near ground yaw angle,
particularly for hills with aspect ratio larger than 1. This discrepancy is a result of the use of
hill slope in the lateral direction and the wind speed-up ratio for the ESDU calculations.
Observed yaw angle profiles in previous topographical studies were employed to generate
two twisted wind profiles in a wind tunnel to investigate the influence of twisted wind flows
in the pedestrian-level wind environment around isolated buildings and rows of buildings.
There were noticeable flow feature alternations found in asymmetric wind fields generated by
the twisted wind flows. Twisted wind flows generated smaller high wind speed areas and
larger low wind speed areas compared to CWP. Both flow feature modifications and high and
low wind speeds area vary with the degree of the wind twist as well as with building
dimensions in twisted wind flows. In buildings rows, twisted wind flows generated dissimilar
gap wind speeds, which increased with the gap width and the degree of wind twist.
A novel technique, named as the equivalent wind incident angle method was developed to
replicate wind environments resulted from twisted wind flows by only using the conventional
wind flow. It was found that, regardless the wind incident angle, 6°(±2) and 13°(±2) are the
equivalent wind incident angles for TWP13 and TWP22, respectively. This finding was
applied to the pedestrian-level wind tunnel test of a real urban model of Tsuen Wan area, in
order to improve the current AVA practice. The threshold, and the proportional methods of
corrections, which account for the effect of twisted wind flows were assessed to estimate
their accuracies. The wind tunnel test results revealed that the proposed method can
effectively replicate wind conditions resulted from twisted wind flows while the threshold
method overestimates the wind availability by about 5% and the proportional method
underestimates the value by 3-5%.
The Computational Fluid Dynamics (CFD) simulations were employed to determine the
driven flow mechanisms of flow alternations for twisted wind flows. It was found that the
deviation of the flow impinging location on the windward side of a building is the main
reason for flow alternations including asymmetric corner streams and deviation of the far
wake area from the building centre line. The deviation of the impinging location of the
vertical streamlines passed over the building indicates the effect of the twisted wind flow is
three-dimensional and should be considered in other applications of wind engineering such as
calculating the wind load calculations or estimation of plume dispersion of a stack on a roof.
Post a Comment