The development of high strength concrete, higher grade steel, new construction techniques

and advanced computational technique has resulted in the emergence of a new generation of
tall structures that are flexible, low in damping, slender and light in weight. These tall structures
are sensitive to dynamic wind loads. Due to high slenderness, low natural frequencies, low
inherent damping levels and high wind speeds at upper level, super-tall buildings are
susceptible to wind excitations, particularly to vortex-induced oscillations. From design point of
view, not only the wind loads, the wind-induced building motions are also within the scope of
design to ensure building’s serviceability. It is well known that the behavior of wind response is
largely determined by building shapes.
Considerations regarding aerodynamic optimization of building shapes in early architectural
design stage is proved to be the most efficient way to achieve in windresistant design.
Wind-resistant design and aerodynamic optimization are the modern topics in building design
community. However, its practice and successful example can be traced back a long time ago.
In ancient China, tall buildings appear to be those of traditional pagodas. Some of them even
meet the modern definition of slenderness for super-talls Fig. 1.1 shows the renowned ancient
pagodas in China.

These three pagodas located in ChongSheng Temple, Dali, Yunnan Province, were built 1180
years ago (824–859AD). The tallest one is 69.13 m in height with a square base of 9.9m in
width, the slenderness (height/width ratio) being 7. The two identical shorter pagodas have a
height of 42.19m. After the completion of these pagodas, a monastery was built. Over the long
period of extreme climates and natural disasters, the original monastery was completely
destroyed by natural forces but the pagodas have miraculously survived. In addition to
extremely strong earthquakes in 1514 and 1925 the pagodas also experienced strong winds in
history.These surviving ancient structures at least reveal two important facts which are helpful
even for modern design practice.

Aerodynamic Optimization

AEORODYNAMIC FORCES ON BUILDINGS

A structure which is immersed in a given flow field is subjected to aerodynamic forces. For
typical tall buildings, aerodynamic forces includes are drag (along-wind) forces, lift (across-wind)
forces and torsional moments. The alongwind forces act in the direction of the mean flow. The
alongwind motion primarily result from pressure fluctuations on windward and leeward faces
and generally follows fluctuations in the approaching flow. The crosswind forces act
perpendicular to the direction of mean wind flow. The common source of crosswind motion is
associated with ‘vortex shedding’.

The torsional motion is developed due to imbalance in the instantaneous pressure distribution
on each face of the building. In other words, if the distance between elastic center of the
structure and aerodynamic center is large, the structure is subjected to torsional moments that
may significantly affect the structural design. It has been recognized that for many high-rise
buildings, the crosswind and torsional responses may exceed the along wind response in terms
of both limit state and serviceability designs.

ALONG-WIND AND ACROSS-WIND RESPONSES

For wind-resistant design of buildings, it is important to identify the type of wind response that
governs the design. For most super-tall buildings, it is often found that acrosswind dynamic
response dominates the design wind loads and some-times causes excessive motions in terms
of building’s serviceability criterion. Fig. 2.1 presents a typical azimuth plot of overall
wind-induced shear force in y-direction for a building. The plot indicates that for along-winds
(i.e., 25degree and 205degree), the mean loads are 3.5*10^7 N and -3.3 *10^7 N, and the peak
absolute loads are 6.7*10^7 N and 6.6*10^7 N, respectively. For across-wind directions (i.e.,
115degree and 295degree), the mean loads are almost zero but the peak absolute loads reach
to 1.4*10^8 N and 1.0*10^8 N. The main reason that the across-wind loads can dominate the
design of super-tall buildings is explained in Fig. 2.2. Fig. 2.2 shows a typical across-wind force
spectrum in comparison with an along-wind force spectrum.

While the along-wind force spectrum mainly reflects the approaching wind turbulence
properties, the across-wind force spectrum is largely determined by flow separation and vortex
formation, so called “signature turbulence”. Compared with alongwind response, across-wind
response is more sensitive to wind speed. At lower wind speeds, the along-wind loads normally
dominate but with increase of wind speed the acrosswind loads take over. Due to relatively
lower natural frequencies of super-tall buildings (or longer natural periods),in addition to higher
wind speeds at upper levels of the boundary layer, the reduced frequency of a super-tall building
at design wind speed can be very close to the reduced frequency where the peak of the
across-wind force spectrum occurs.

For example of a 600 m tall square building that has a width of 60 m and the first sway period of
9 s, the vortex-induced resonance can occur at about 56 m/s winds at upper level, or 31 m/s at
a standard 10 m height. The approaches of aerodynamic optimization would be different when
dealing with along-wind or across-wind responses. Although some approaches that can benefit
both, many approaches focus on one type or other. For optimization of along-wind response, a
basic approach is to modifying building’s corners, such as making round or chamfering.

Aerodynamic Optimization

Opening is sometimes within the list of feasible options. The optimization of across-wind
response, two basic approaches can be taken: (1) to reduce the magnitude of vortex excitation
by modifying the building’s cross-section such as corner recession or opening; and (2) to reduce
the synchronization and correlation of fluctuating forces by varying building’s shape with height
such as tapering or twisting.

Aerodynamic Optimization
Since across-wind dynamic response is normally the main source that causes excessive wind
loading and discomforting motion for super-tall buildings, the following discussion will focus on
the aerodynamic optimizations for across-wind dynamic response.

AERODYNAMIC MODIFICATIONS TO BUILDING SHAPE AND CORNER

The various aerodynamic modifications applied to the tall buildings to mitigate the wind
excitations maybe classified in two groups, Minor modifications: Aerodynamic modifications
having almost negligible effects on the structural and architectural concept, for examples corner
modifications like fitting of fins, fitting of vented fins, slotted corners, chamfered corners, corner
recession, roundness of corners and orientation of building in relation to the most frequent
strong wind direction. Major modifications: Aerodynamic modifications having considerable
effects on the structural and architectural concept, for examples setbacks along the height,
tapering effects, opening at top, sculptured building tops, varying the shape of buildings,
setbacks, twisting of building etc.

MODIFICATION OF CROSS-SECTION

A few examples of aerodynamic modifications for building cross-sections are shown in Fig 3.1 to
3.3. The corner of Taipei 101 Tower, shown in Fig3.2, was designed during wind tunnel testing,
which effectively reduces the overall design wind loads by about 25% compared with the original
design of square section.

Aerodynamic Optimization

During the wind tunnel tests of Taipei 101, other type of corner modifications are also
investigated, shown in Fig3.2. With the dimension of modifications being about 0.1B (10% of
building width), the effects of these modifications are similar. The final selection of the corner
modification was basically the choice of architects. Helical strike is a traditional device to
suppress vortex excitation to chimney stacks. However, few of these devices are used for
buildings due to aesthetical concern. However, it has been found that commonly designed
corner balconies can perform similar roles in suppressing vortex shedding.

The concept was used again in a recent tall building in a very strong typhoon area, shown in
Fig3.3, and proved to be able to reduce design wind loads by about 20%. The overhang eaves
shown in ancient pagodas, Fig1.1, are also the examples, Opening is not commonly used in
design practice due to potential impact on useable spaces. However in some cases the corner
slot can not only significantly reduce the across-wind excitation, but also make internal space
design more logical

EFFECT OF FINS AND VENTED FINS

The aerodynamic modifications to basic square cross-sectional shape of buildings by using
small fins or vented fins have significant effects on the alongwind and crosswind response
characteristics. Small fins/vanes fitted to the corners of a prismatic building with a gap between
the vanes and the corner can help to alleviate negative pressures under the separated shear
layers on the side faces. However, the added drag introduced by these vanes increases the
along wind responses. The fitting of fins served only to increase the critical wind speed without
any noticeable disruption to vortex shedding process.

At the high range of reduced velocities, there was an apparent reduction in galloping response
when vented fins were fitted. The fitting of fins or vented fins is acceptable for general usage
only for certain range of reduced wind velocities. The aerodynamic modifications to buildings
like fitting of fins and vented fins causes noticeable increase in the alongwind response due to
an increase in the projected area normal to wind direction. The aerodynamic modifications,
which in general increase the projected area or the effective width of a building, would not be
beneficial

EFFECT OF SLOTTED CORNERS, CHAMFERED CORNERS, CORNER RECESSION

The double step corner recession modifications had been applied to the cross section of 508 m
high, 101 storey, Taipei101 building, Taiwan. A corner modification Investigations have
established that corner modifications such as slotted corners, chamfered corners/corner cut,
corner recession are in general effective in causing significant reductions in both the alongwind
and crosswind responses compared to basic building plan shape. The modification of windward
corners is very effective to reduce the drag and fluctuating lift through changing the
characteristics of the separated shear layers to promote their reattachment and narrow the
width of wake. This type of modifications is also effective to suppress the aero elastic instability.

The effects of slotted corners and chamfered corners were investigated by Kwok and Bailey
through wind tunnel tests on aero elastic square and rectangular models of dimension 60mm
60mm 540mm and 112.5mm 75mm 450mmrespectively with and without slotted and chamfered
corners. The modifications to the building corners ranged from 9% to 16% of building breadth. It
is concluded that, slotted corners and chamfered corners were causing noticeable reductions in
both the Dynamic alongwind and crosswind responses as compared to plain rectangular shape
building. Venting through the slotted corners appears to be effective in reducing the drag force
without undesirable effect of using vented fins. For building heights ranging from 240 to 280m
the recessed corners are more effective than the chamfered corners in reducing both along wind
and crosswind moments due to buffeting and vortex shedding excitations respectively.

In general, for the range of building forms and heights tested, the construction cost was reduced
with the introduction of chamfered and recessed corners even though the building height was
increased to maintain the total usable floor area for the entire building. This type of corner
modifications (corner recession) had been applied to the 150 m high Mitsubishi Heavy
Industries Yokohama Building as shown in Fig 3.5, which was located in a water front area in
the wake of peripheral tall buildings. To reduce the wind-induce responses, all the four corners
were chamfered, which consequently reapplied to the Taipei101 building reduces the base
moment by 25% as compared to building of basic square section.

CONCLUSION
Aerodynamic optimization of building shapes is an important portion of super-tall building
design. Two categories of optimization are discussed in the paper: aerodynamic modifications
that are normally considered as remedial measures; and aerodynamic designs that integrate
architectural design with aerodynamic study in the early design stage. While aerodynamic
modifications mostly involve building corner treatments, aerodynamic designs have more
options in building shapes, including overall elevation optimizations such as tapering, twisting,
stepping, opening, top sculpturing, etc.

A few examples of aerodynamic optimization schemes that have been implemented with
success in building designs are illustrated in this paper. The main challenges in building
aerodynamic optimization are to compromise aerodynamic solutions with other architectural
design aspects and to compromise between benefits and costs. Therefore, it is important to
have a reasonable assessment of effectiveness of various aerodynamic options in the early
design stage so that the potential pros and cons can be evaluated in the decision-making
process. The method proposed in this paper can serve this purpose. With this method, the
assessment of aerodynamic effectiveness for tapering, stepping and twisting can be conducted
with minimum wind tunnel tests at low cost. The results of the case study reveal some important
phenomena:

1. In general, tapering and stepping can reduce across-wind responses. However, for a
low-return period response such as a building’s performance in common winds where the
corre-sponding reduced velocity is low, it is possible that tapering or stepping may actually
increase accelerations affecting occupant comfort.

2. While twisting can considerably decrease the maximum across-wind responses, it can also
lead to equalized response over wind directions. Therefore, a potential mutualeffect between the
two common optimization options, optimization through twisting and optimization through
building orientation, should be considered.

3.Although aerodynamic effectiveness of twisting generally increases with the increase of

twisting level, the increment of effectiveness tends to decrease after a limit.

4.The corner roundness is the most effective to suppress the aeroelastic instability for a square
building. The amplitude of the wind-induced vibration reduces as the extent of the corner
roundness increases.

5.Tapering effect has a more significant effect in acrosswind direction than that in alongwind
direction.
6.The through building opening along the alongwind and crosswind direction, particularly at top
significantly reduces the wind excitation of the building.