University of Southern California Dissertations and Theses

Integration of mass dampers and external shading fins: exploring synergy in structural and environmental control systems

(USC Thesis Other)

Integration of mass dampers and external shading fins: exploring synergy in structural and environmental control systems

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

INTEGRATION OF MASS DAMPERS AND EXTERNAL SHADING FINS:
EXPLORING SYNERGY IN STRUCTURAL AND
ENVIRONMENTAL CONTROL SYSTEMS

by

Tat S. Fu

__________________________________________________________

A Thesis Presented to the
FACULTY OF THE SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF BUILDING SCIENCE

December 2007

Copyright 2007 Tat S. Fu
ii

Table of Contents

List of Figures ............................................................................................................ iv

List of Tables ............................................................................................................ vii

Abstract .................................................................................................................... viii

Chapter 1 – Introduction ............................................................................................. 1

Chapter 2 – Importance ............................................................................................... 5
2.1 Structural Control .............................................................................................. 5
2.2 Environmental Control ...................................................................................... 6
2.3 Integrated Control ............................................................................................. 6

Chapter 3 – Background ............................................................................................. 8
3.1 Structural Control .............................................................................................. 8
3.1.1 Base Isolator ............................................................................................... 8
3.1.2 Passive Energy Dissipation ...................................................................... 10
3.1.2.1 Multiple Mass Damper System ......................................................... 10
3.1.3 Active and Semi-active Control ............................................................... 11
3.2 Environmental Control .................................................................................... 12
3.2.1 Passive Systems ....................................................................................... 13
3.2.2 Active Systems ......................................................................................... 14

Chapter 4 – Structural and Environmental Control Synergy (SECS) ....................... 18
4.1 Synergy ........................................................................................................... 18
4.2 Compatibility between SC and EC ................................................................. 19
4.3 Benefits and Costs ........................................................................................... 21
4.4 Synergy Diagram ............................................................................................ 22
4.5 Case Studies .................................................................................................... 23
4.5.1 Crystal Tower in Osaka (1990) ................................................................ 23
4.5.2 Sendagaya INTES Building in Tokyo (1991) .......................................... 24

Chapter 5 – Shading Fin Mass Damper System........................................................ 26
5.1 Compatibilities between Mass Damper and Shading Fin ............................... 28

Chapter 6 – Mass Damper ......................................................................................... 30
6.1 Distributed Mass Damper ............................................................................... 30
6.2 Formulation / Simulation Model ..................................................................... 32
6.3 Simple case: Equally Distributed Dampers..................................................... 34
6.3.1 Fine Tuning the Parameters ..................................................................... 37
6.4 Optimization of damper parameter ................................................................. 42
iii

6.4.1 Pattern Search Optimization .................................................................... 43
6.4.2 Optimization Results ................................................................................ 45
6.5 Discussion on Performance and Practicality ............................................................... 50
6.5.1 Sub-optimization of Damper Stiffness and Damping Only ..................... 51

Chapter 7 – Shading Fin ............................................................................................ 54
7.1 Movable Shading Fin ...................................................................................... 54
7.1.1 Simulation Model ..................................................................................... 55
7.1.2 Protracting / Retracting Shading Fin ........................................................ 56
7.1.3 Rotating Shading Fin ............................................................................... 58
7.2 Active Shading Fin .......................................................................................... 61
7.2.1 Shading Schedule In EQuest .................................................................... 62
7.2.2 Comparing Simulations with and without Shading Schedule .................. 65
7.2.3 Actively Rotating Fins using Shading Schedule ..................................... 67

Chapter 8 – Conclusion and Future Research ........................................................... 71
8.1 Active + semi-active mass damper system ..................................................... 73
8.2 Active SFMD for Structural Health Monitoring (SHM) ................................ 74

References ................................................................................................................. 75

iv

List of Figures

Figure 3.1: Base isolator diagram ............................................................................... 9

Figure 3.2: Single TMD, MTMD, and MTMD distributed along the building
height.. ....................................................................................................................... 11

Figure 4.1: Sendagaya INTES Building ................................................................... 24

Figure 4.2: Sendagaya INTES building with hybrid mass dampers ......................... 24

Figure 4.3: Crystal Tower ......................................................................................... 25

Figure 5.1: Plan View of the Shading Fin Mass Damper System ............................. 26

Figure 5.2: Overview of the Shading Fin Mass Damper System .............................. 27

Figure 5.3: Shading Fin Mass Damper System: front and section ............................ 27

Figure 5.4: Shading Fin Mass Damper System: details and outside details ............. 28

Figure 6.1: DMD, TMD, MTMD diagrams .............................................................. 31

Figure 6.2: Design earthquakes ................................................................................. 33

Figure 6.3: Frequency content of design earthquakes and Kanai-Tajimi filter ......... 33

Figure 6.4: Objective function over damper parameters: equally distributed mass
damper and single mass damper located at the top floor .......................................... 35

Figure 6.5: Comparison between the equally distributed and single damper
systems ...................................................................................................................... 36

Figure 6.6: Performance of the EDMD and single TMD on different historical
earthquake records .................................................................................................... 37

Figure 6.7: Performance and parameters for single TMDs on different floors ......... 38

Figure 6.8: Comparsion between the DMD systems of all floors and top half of
floors only ................................................................................................................. 40

Figure 6.9: Performance of the weighed DMD system on the four desgined
earthqaukes ................................................................................................................ 41

v

Figure 6.10: Performance of the weighed EDMD on the other easrthquake
records and random excitation .................................................................................. 42

Figure 6.11: Pattern search optimization with randomly chosen initial
parameters (1) ........................................................................................................... 46

Figure 6.12: Pattern search optimization with randomly chosen initial
parameters (2) ........................................................................................................... 46

Figure 6.13: Pattern search optimization with the EDMD system as the initial
guess .......................................................................................................................... 48

Figure 6.14: Pattern search optimization with the weighed EDMD system as the
initial guess ............................................................................................................... 48

Figure 6.15: Performance of the optimized DMD system (from Figure 6.14)
subject to the four designed earthquake records. ...................................................... 49

Figure 6.16: Performance of the optimized DMD system (from Figure 6.14)
subject to the other earthquake records and random excitation ................................ 49

Figure 6.17: Pattern search sub-optimization with the EDMD as the initial
guess .......................................................................................................................... 52

Figure 6.18: Pattern search sub-optimization with the weighed EDMD as the
initial guess. .............................................................................................................. 52

Figure 6.19: Pattern search sub-optimization with the stepped EDMD as the
initial guess. .............................................................................................................. 53

Figure 7.1: Movements of shading fins (plan view) ................................................. 55

Figure 7.2: Sun paths for summer and winter ........................................................... 55

Figure 7.3: Hall of Records, Los Angeles (1962) by Richard Neutra....................... 55

Figure 7.4: Caltrans District 7 Headquarter, Los Angeles (2004) by Thom
Mayne ........................................................................................................................ 55

Figure 7.5:eQuest building model ............................................................................. 56

Figure 7.6: Effect of 5ft shading fins ........................................................................ 56

Figure 7.7: Sunlight effect for short and long vertical shading fins in plan view ..... 57
vi

Figure 7.8: Effect of protracting shading fins ........................................................... 58

Figure 7.9: Sunlight effect for rotated vertical shading fins in plain view ............... 59

Figure 7.10: Effect of rotating shading fins .............................................................. 60

Figure 7.11:eQuest building model for active rotating shading fins ......................... 61

Figure 7.12: Comparison of shading fins using shading schedule ............................ 65

Figure 7.13: Comparison of rotating fins using shading schedule ............................ 67

Figure 7.14: Comparison of active rotating and fixed fins using shading
schedules.. ................................................................................................................. 68

Figure 8.1: Active + semi-active mass damper ......................................................... 73

vii

List of Tables

Table 6.1: Performance and parameters for single TMDs on different floors .......... 38

Table 6.2: Performance of the weighed EDMD and optimized DMD system on
different excitations ................................................................................................... 50

Table 7.1: Shading schedule arrangment .................................................................. 62

Table 7.2: Daylight calculation effects ..................................................................... 63

Table 7.3: Shading schedule for the actively rotating fins throughout a year ........... 70

viii

Abstract

Synergy of integrating structural and environmental control systems is
explored with a proposed Shading Fin Mass Damper (SFMD) system. External
shading fins control sunlight entering the building, adjusting the internal temperature
and lighting conditions. In the SFMD system, the fins also serve as tuned mass
dampers (TMDs) that move and dissipate energy during strong structural motion.
The additional mobility allows the fins to change positions for greater control to
minimize energy consumption. Using eQuest, a building energy simulation program,
the movable shading fins substantially decrease energy consumption. Since the
shading fins are placed along the height of the building, the dampers are distributed
rather than concentrated in a few locations as in typical TMD systems. The
distributed mass damper (DMD) system is formulated, simulated and optimized to
reduce structural vibration significantly. Sub-optimal DMD systems are also studied
for simpler design and constructability without compromising structural benefits.

1

Chapter 1 – Introduction

As buildings become larger and taller, building technologies have been
evolving and have become more complex. Buildings today have to be maintained
and controlled for safety and comfort of the occupants and efficiency of building
operations. For example, tall buildings adopt computer algorithms to control the
movements of elevators for optimizing people flow. Large and complex buildings
require layers of control systems to work together for smooth operations. This thesis
focuses on two vital building control systems — structural and environmental
controls — and explores synergy between them to improve efficiency.
Structural control (SC) is an emerging technology that started in the 1970s.
The technology aims to prevent structural damage for buildings by reacting to the
forces (from earthquake, wind, etc.) the buildings are experiencing. Dampers and
base isolators are two major examples of SC. Although there has been increasing
interest in the technology of building designs, SC systems exist in a small percentage
of the buildings constructed. Soong and Spencer’s review paper (2002, p.243) states
that, as of that time, there were only 41 full-scale implementations of “active” SC
systems with 40 of them in Asia (mostly Japan) and one in America (active SC
systems excludes more traditional passive SC systems such as the John Hancock
Tower in Boston — the first building with tuned mass dampers). There are three key
reasons behind the lack of buildings with SC — familiarity, perceived cost and legal
reasons (especially in the U.S). Most structural engineers are aware of such
technology but few are familiar enough with the technology to incorporate it into
2

their designs; and even fewer developers and architects are familiar with SC to invest
in the technology. The unfamiliarity leaves a costly perception of SC. This creates a
cycle: while the unfamiliarity and perceived cost of the technology discourage
developers, the scarcity of its applications results in a lack of research and
development, which further slows down its popularization. The lack of
popularization then results in lack of attention in building code to include such
“new” technology. This cycle of unfamiliarity, cost and legal reasons is one of the
reasons why many structural technologies take decades to become mature and
practical, with the exception of the spikes of developments after catastrophic events
such as the Northridge Earthquake in 1994, which exposed many structural
problems.
Environmental Control (EC), in comparison, is a more mature field. All large
buildings require elaborate systems to control lighting and human comfort while
conserving energy cost. Cost is also a major concern for EC, originating in two
sources: initial (installation) cost and annual energy cost. Initial cost is based on the
system design and is a part of the construction cost. Annual energy cost is directly
related to the performance of the designed system and is a part of building operation
cost. Since construction cost is much higher than the annual building operation cost,
a less effective and less costly EC system is often chosen to cut construction cost.
However, this practice is shortsighted because the long term savings of a better
performing system can easily surpass the extra initial cost after a number of years.
Although this logic is simple, the budgeting politics and the fact that different groups
3

may handle the construction cost and building operation cost make it difficult to
build a more expensive EC system with better performance.
One solution to both SC and EC problems that this thesis suggests is creating
synergy by combining both control systems in buildings. By attaching the less
familiar technology of SC to the familiar technology of EC, builders (developers,
architects, engineers, etc.) will not dismiss the combined technology as easily.
Additionally, the combined structural and environmental benefits can attract interests
from both researchers and builders, while the cost can be reduced by sharing the
synergies between the two control systems. The initial cost is now owed to both
structural and environmental aspects of the building. And since structural safety
directly affects occupants’ safety, builders cannot easily sacrifice the performance of
the combined control system for cutting construction cost. Nonetheless, this thesis
does not focus on cost analysis of the combined SC and EC system. Instead, it
explores possible synergies of the combined systems to make such integration
worthwhile.
A new system that integrates SC and EC is also proposed in this thesis — the
Shading Fin Mass Damper system. The proposed system combines both SC and EC
by using shading fins as mass dampers. The fins block sunlight and reduce excessive
heat gain while they dissipate energy from large building vibration. More
explanation of the synergy system will be discussed in Chapter 4.
The following chapters discuss about the importance of structural,
environmental and integrated controls (Chapter 2) and background of the control
4

systems (Chapter 3). Chapter 4 talks about synergy of SC and EC and some case
studies. Chapter 5 introduces and gives details about the proposed synergy system of
Shading Fin Mass Damper system. The mass damper and shading effects of the
synergy system are further examined in Chapter 6 and 7, respectively. Lastly,
Chapter 8 summarizes the findings and impact of this thesis and suggests future
work.

5

Chapter 2 – Importance

The foci of this thesis are structural and environmental controls. This chapter
discusses why these two subjects should be studied.

2.1 Structure Control

A structure is the backbone that physically supports an entire building. There
are two key design considerations — vertical and lateral loads. Vertical loads are
mostly caused by gravity and therefore remain generally constant during the life
cycle of buildings. In contrast, lateral loads can be caused by winds, ground
motions, etc. These events are constantly changing and sometimes unpredictable.
Ground motions of different characteristics can have wildly different effects on the
same structure. Moreover, winds and ground motions often are better buffered by
different structural characteristics. Structural control offers a good solution to this
problem, controlling the behavior of structures in response to the external forces.
From the Northridge Earthquake in 1994 to the collapse of the World Trade
Center in 2001, the effect of structural failure can be widespread and deadly.
Although natural and man-made disasters happen rarely, their consequences cannot
be ignored. The cost of repairing/rebuilding and the immeasurable cost of human
lives are two very valid reasons that structure design must account for these
disasters.

6

2.2 Environmental Control

If structures are comparable to backbones for buildings, environmental
control (EC) systems are analogous to the vital organs that regulate homeostasis of
the buildings. The productivity of the buildings’ occupancy is directly related to the
ability of the EC systems to maintain a comfort zone for the occupancy. For
example, the EC systems must maintain comfortable temperature and humidity while
providing sufficient lighting, clean/hot water, and fresh air. Sufficiency is the basic
requirement for EC systems and performance of the system is mostly measured by
building integration, energy efficiency and ease of maintenance. If the EC system
can be fit seamlessly with the rest of the building, then compromises or changes can
be kept minimal in the building designs. An energy efficient EC system not only
saves building operation cost, it is also sustainable to nature. All systems wear out
and break eventually, and providing simple maintenance requirement can reduce the
cost of the system by maximizing the lifecycle of the system and retaining the
designed energy efficiency. This thesis focuses more on the building integration and
energy efficiency aspects of EC systems.

2.3 Integrated Control

With larger and more complex buildings and ever advancing building
technologies, an effective integration of building systems becomes more crucial in
designing and constructing efficient buildings. Cost, constructability, energy
consumption, maintenance, durability, daily operations and other building criteria are
7

all subjected to both how well each building system works and how well they work
together. Though this thesis will not cover the integration of all building systems, it
explores the synergy and integration of two major building systems: SC and EC
systems.

8

Chapter 3 – Background of Structural and Environmental
Control

This chapter gives a brief background of Structural Control (SC) and
Environmental Control (EC). SC is a component of smart structures while EC is an
essential architectural system. For SC, there are base isolation systems, passive
energy dissipation systems and active/semi-active systems. EC systems are
categorized into passive and active systems.

3.1 Structural Control (SC)
SC prevents structural damage by reducing building vibration induced by
natural and man-made hazards. There are three key SC systems — base isolation
systems, passive energy dissipation systems and active/semi-active systems. The
following sections explain briefly and offer some examples of these systems. For
more details regarding SC systems, the reader is referred to the review paper by
Soong and Spencer (2002).

3.1.1 Base Isolation System
A base isolation system is considered to be the most mature system of the
three SC types. There are more buildings constructed with base isolators than other
SC systems. The system isolates the building from its foundation during strong
motions such that most of the relative motion is in the base, instead of the
9

superstructure (Figure 3.1). The superstructure of the building experiences less
motion and therefore less a decreased

Figure 3.1: Base isolator diagram (Takenaka Corp. 2001)

likelihood of damage. The challenge of the base isolation system is the special
connection between the foundation and the structure above. The connection must
allow lateral movements while transferring the weight vertically from the structure to
the foundation. Moreover, the lateral movements must be restricted and damped out
in such a way that the building is not “sliding off” the foundation. Additionally,
while connections across the isolation layer (e.g., utility lines, plumbing, etc.) use
flexible components, the isolator motion is limited by these connections. There are
several types of connection designed for base isolators such as elastomeric bearings,
lead rubber bearings, and sliding friction pendulum.

10

3.1.2 Passive Energy Dissipation
In general, passive energy dissipation (PED) systems use dampers to
dissipate energy from excited structures to reduce vibrations. These dampers mostly
operate on the dissipating nature of friction, metal yielding, phase transformation in
metals, viscoelastic (VE) solids or fluid, and fluid orificing. The following are some
of the well known examples:
• Metallic dampers
• Friction dampers
• VE dampers
• Viscous fluid dampers
• Tuned mass dampers
• Tuned liquid dampers

Of all the PED systems mentioned, mass dampers are most utilized, with the first
application in the John Hancock Tower (1976) at Boston. The greatest challenge of
PED systems is that their passive nature limits their applicability to different types of
external forces that excite the structures. PED systems might not be effective at all
for excitations for which they were not designed. Since earthquakes can produce
excitations with different frequencies, PEDs are not always useful for seismic design.

3.1.2.1 Multiple Mass Damper System
Since the proposed Shading Fin Mass Damper (SFMD) system employs a
type of a multiple tuned mass damper (MTMD) system, this section briefly
introduces the MTMD. MTMD was first proposed by Igusa and Xu (1994) in the
early 1990s to compensate the sensitivity of a single TMD to the uncertain natural
frequencies of the building system. The MTMD was later extensively studied by
11

Yamaguchi and Harnpornchai (1993), Abe and Fujino (1994), and Kareem and Kline
(1995). However, the MTMD in these 1990s studies concentrates the multiple
dampers in one floor in contrast with the SFMD herein which has dampers on every
floor. The shading fin function would require the dampers to be distributed along the
height of the building (dampers on all floors). Figure 3.2 illustrates the difference
between the single TMD and the two types of MTMDs. Recently, Chen and Wu
(2001, 2003) studied a MTMD system with TMDs placed in multiple floors on a 6
story simulation model and a 3 story, 1/4 scaled experimental building model. They
showed that MTMDs can effectively reduce seismic responses.

Figure 3.2: Single TMD, MTMD, and MTMD distributed along the building height

3.1.3 Active and Semi-Active Control
Unlike passive systems, active and semi-active systems are designed to adapt
to various kinds of excitations. Active control uses actuators to apply forces on the
12

structure to counteract external forces. Examples include active bracing systems and
active mass drivers. Theoretically, active control can counter all excitation and keep
the structural vibration to a minimum. However, to achieve such an ideal result, the
system often would require energy too great to be practical, and perfect actuators and
noiseless sensors throughout the structure. Semi-active control addresses such
impracticalities by uniting active and passive control systems. Semi-active systems
are essentially PED systems with controllable parameters, such as stiffness and
damping. By controlling PED system parameters according to structural conditions,
semi-active systems can adapt to various kinds of excitations. Additionally, since
semi-active control does not directly use energy to restrict structural motion, it works
with limited energy requirements. Some examples of semi-active systems are
variable stiffness or damping systems, and magnetorheological (MR) dampers.

3.2 Environmental Control System
Besides the structure, the environmental control (EC) system is also an important
component in a building. EC maintains the productivity of the building occupants by
providing sufficient human comfort in areas such as lighting, humidity and
temperature. Well-designed EC systems also aim to provide services under minimal
energy cost. The following sections briefly describe some examples of EC systems
by separating them into passive and active systems.

13

3.2.1 Passive Systems
Passive EC systems require little or no input energy from man-made sources.
Some examples include:
• Natural Ventilation – aided by natural air flow, ventilation cools and brings
fresh air into spaces. Openings such as windows, doors and vents allow
air movement between exterior and interior spaces. There are two types
of natural circulation techniques — wind-induced cross ventilation and
gravity or convection ventilation. Cross ventilation places openings
carefully to exploit local wind pattern while gravity ventilation draws
cool air from lower inlets by letting the warm air out on higher outlets
when the outside air is cooler than the upper vent insider air.
• Thermal mass – large masses such as masonry walls that trap heat during
day-time and release heat slowly throughout the rest of the day. The
advantages of a thermal mass are that the spaces can be heated for a
prolonged period of time (i.e., after sunset). Also, the temperature
increase is less intense with thermal masses, preventing overheating from
direct sunlight. The main disadvantage is the uncontrollability of the heat
stored. It takes a long time to heat up spaces and it is difficult to stop heat
gain even when the spaces are warm enough. In addition to temperature
increase, thermal masses can also be used for cooling when the masses
are colder than the surrounding temperature.
14

• Sunspace – attached space that is heated directly by the sun and transfers the
heat to connecting rooms. Unlike the thermal mass, sunspace can quickly
increase the temperature in the attached space but it also cause large
temperature fluctuations. Another disadvantage is that sunspace does not
store heat to prolong temperature increase.
• Shading devices – overhangs and fins that block portions of direct sunlight.
Not only can solar heat gain be blocked, glazing can also be controlled
with shading devices. Nonetheless, a compromise must be made when
only one of solar heat gain or daylighting is needed and the other is
undesired. Site condition and orientation also play major roles in
designing shading devices. Overhangs are typically placed in the south
façade while vertical fins are typically placed in the east and west façades
to deal with sunlight coming in from different angles throughout the day.
Shading fins are discussed further in Chapter 7 for the proposed Shading
Fin Mass Damper system.
• Insulation – separation between interior and exterior of the building that
prevents quick heat gain and loss. A well insulated space can decrease
the amount of heat gain in hot weather and heat loss in cold weather.

3.2.2 Active systems
Active EC systems, on the other hand, usually require constant energy input
such as electricity and natural gas. Some examples include:
15

• Refrigeration – a vapor-compression cycle that transfers heat between
locations. Many air-conditioning units use refrigeration to cool or heat
air for buildings. The process involves compressing a refrigerant (gas or
liquid depending on the temperature and pressure) such that it becomes
warmer than outside air. The compressed/heated refrigerant then loses
heat to the cooler outside air though a heat exchanger. After being
cooled, the refrigerant is “decompressed” through an expansion valve,
causing typically a change of phase from liquid to vapor due to the
pressure drop which results in dramatic drop in temperature. The
resulting refrigerant is much colder than the refrigerant before
compression and can be used to cool other mediums (air, water, etc.)
before going through the compression refrigeration cycle again.
• Chiller – water chiller that cools water to supply other cooling units
throughout the building. It uses a large amount of electricity and a
compression refrigeration cycle to chill water. Water chillers are more
suitable for large buildings because of the large capacity. They can be
categorized as reciprocating, centrifugal, rotary and absorption chillers
ranging from 60 to 400 tons, where a “ton” is 12,000 Btuh.
• Movable shading devices – overhangs and fins that can adjust to block
targeted portions of direct sunlight. This increases the effectiveness of
the shading devices by balancing the need for solar heat gain and lighting
according to the weather/sun orientation. More about movable shading
16

fins is discussed in Chapter 7 for the proposed Shading Fin Mass Damper
system.

• Heating units – use electricity, gas or other fuels to heat the building by
warming a medium (air, water, etc.). The medium is then supplied to
various parts of the building. A furnace is a typical type of heating unit
that warms air using an electric or combustion heating chamber. Boilers
are closed vessels that produce hot waters or even steam. Heat pumps are
very efficient heating units for mild climates since they, instead of
producing heat, transfer heat by compression refrigeration cycles. They
are also easily scalable, ideal for buildings unsuitable for central systems.
Heating coils are one of easiest ways to heat air. Transferred through air
ducts, air is heated before it is supplied to building areas.

This is also not true. Heat pumps are used where are variety of different
temperatures are required within a building.
• Ice storage – cools or freezes water (or other media) at off-peak hours
(nights) to be used to cool the building throughout the day. By cooling
water at night, energy cost is decreased while the colder temperature can
help cooling more efficiently. A storage system is needed and its size is
proportional to the building size.
• Air system – air-handling system that transports air throughout the building
to compliment existing heating/cooling, humidifying/dehumidifying, and
filtering units. Fresh air is drawn into the building by inlets and after
17

conditioning, the air is supplied to the building. Returned air from the
building is then either reconditioned or exhausted through outlets, usually
in some percentage ratio. Fans and ducts are key components of air
systems.
For more details and other EC systems, readers are referred to the many text
books published on this subject, such as Bradshaw’s (2006)“The Building
Environment: Active and Passive Control Systems.”

18

Chapter 4 – Structural and Environmental Control Synergy

As mentioned in Chapter 1, this thesis proposes joining functionalities
between SC and EC to solve some of their inherent problems. The synergy of SC
and EC is only possible with their compatibilities. The following sections define
structural and environmental control synergy and discuss the compatibilities that
allow this to happen.

4.1 Synergy
Synergy is defined as a joint venture that is mutually advantageous to the
involved partners. To clarify mutual advantages, one can consider the following
equation:
Mutual Measure = M = (B
j
– ∑B
i
) – (C
j
– ∑C
i
) (4.1)
where B
j
is the benefits of the joint venture, B
i
is the benefits of the individual
partners in separate settings, and C
j
and C
i
are the costs of the joint venture and
individual partners respectively. Clearly, the joint venture is only profitable or
meaningful when M>0 and only then the joint venture can be called synergy.
Most of the time, M is positively related to the compatibility of the different
partners. If the partners are highly compatible, they are more likely to increase their
joint benefits while decreasing their joint costs. Therefore, partner compatibility is a
major consideration when selecting partners to form synergy. A compatible example
of synergy is gas and electricity in hybrid automobiles. In conventional automobiles,
gas is used to accelerate the vehicles and electricity is used to power the electronics
19

of the vehicles. Hybrid technology draws energy from both gas and electricity for
acceleration. Since more electricity is needed, a battery of larger capacity stores
more electricity converted from the energy created during braking. Clearly, the joint
venture of gas and electricity in automobiles is a synergy since they yield better
mileage compared to conventional automobiles. The success of hybrid automobiles
lies in their compatibilities in the following ways:
• They have compatible objectives. In fact, their objective is the same – to
increase mileage efficiency.
• They are physically compatible; both gas and electricity are used in the
engine to accelerate the vehicles.
As shown in the example, objective and physical compatibilities play important roles
in creating synergy.

4.2 Compatibilities between SC and EC
Although SC and EC have generally different objectives, their objectives are
indeed compatible. The goal of SC is to prevent structural damage while it can also
be used when building movements during high winds and small earthquakes are too
small for damage but large enough for discomfort. The goal of EC is to provide
human comfort for the occupants of the building under a reasonable cost. The
compatibility lies in the scheduling and urgency of the objectives.
EC is essentially continuously operating for large buildings since there are
always occupants within the building. The number of occupants may differ from
20

time to time (for example, there are more people during office hours for office
buildings while there are still maintenance and security crews during off-peak
hours), but EC still provides service regardless of occupancy level. Moreover, EC of
large building often has different duties during off-peak hours such as ice storage
that chills water at nights when electricity is cheaper.
SC, on the other hand, is not continuously operating. Since buildings
undergo violent vibrations only during wind storms, earthquakes or other rare events,
SC, may not be needed daily. Even when these events occur, they typically last for a
short time and common building operations can resume quickly. Strong wind may
occur more frequently and last longer in windy area such as Chicago, but their
severity and occurrence are somewhat predictable with weather forecasts. Such
information can be used in sharing service times between SC and EC in a structural
and environmental control synergy system.
Since the objectives of SC and EC do not overlap much in scheduling, they
seldom get in each other’s way. However, under rare circumstances when their
objectives are needed concurrently, there is another fundamental factor of the
objectives that makes SC and EC compatible — the urgency of the objectives. When
different objectives need to be addressed concurrently and they cannot be addressed
separately with full resources, then a compromise must be made between them based
of the importance of the objectives under the current situation. It is difficult to argue
whether SC or EC is more important since their importance is measured in different
time frames. Although the failure of SC may be more devastating than the failure of
21

EC (structural safety compared with human comfort), the frequency of a building
using SC is relative small compared to EC. Over time, EC may even have greater
effects on a building because it is utilized more frequently. Nonetheless, structural
safety is definitely a more urgent concern than human comfort when both concerns
are present. Moreover, controlling building vibration typically takes a short period
of time and EC can be resumed quickly afterward. Thus, the objectives of SC and
EC are still compatible when both objectives are to be addressed.
Other than scheduling, the physical compatibility between SC and EC is
another key factor for creating structural and environmental control synergy.
However, Chapter 3 shows that there are many different types of SC and EC and
each type can have its distinct physical characteristic. Therefore, it is impractical to
discuss the physical compatibility in a general sense. Chapter 5 will propose a
synergy system and discuss its physical compatibility in details.

4.3 Benefits and Costs
In order to make the joint venture of SC and EC a synergy, their M value
must be positive in (4.1). This generally happens if joint benefits increase and/or
joint costs decrease. One of the main costs of control systems is the computational
resource. Based on the current state of the system, control force must be calculated
that stabilize the system. For a complex system, a large amount of computer power
maybe needed to estimate the current state and calculate the control force in real time
before the system’s state changes. Since SC and EC are compatible, they could
share their computational resource to reduce cost. Moreover, the joint and more
22

powerful computational resource maybe able to analyze the control systems
faster/better and therefore give better control input to increase the overall
efficiency/benefits of the structural and environmental control synergy system. More
benefits and costs analysis are discussed in the next chapter for the proposed synergy
system.

4.4 Synergy Diagram

A list of SC systems and a list of EC systems are created. Structural and
environmental control synergy can be created by picking one system from each list
and check if the selected systems can compliment each other.

Structural Control

Base isolation system
• Elastomeric bearings
• Lead rubber bearings
• Sliding friction pendulum

Passive Energy dissipation
• Metallic dampers
• Friction dampers
• VE dampers
• Viscous fluid dampers
• Tuned mass dampers
• Tuned liquid dampers

Semi-active and active control
• Active bracing systems
• Active mass dampers
• Variable stiffness or
damping systems
• Smart materials
• Magnetorheological (MR)
dampers are new semi-
active control devices that
use MR fluids to form a
controllable damper

Environmental Control

Passive Systems
• Natural ventilation
• Thermal mass
• Sunspace
• Shading devices (overhangs, fins)
• Ventilation cooling
• Radiant cooling and heating
• Evaporation cooling
• Insulation
• Daylighting

Semi-active and active systems
• Roof radiation trap
• Movable shading devices (overhangs, fins)
• Convective cooling
• Heating units
• Cooling tower
• A/C units
• Air supply (fans)
• Artificial lighting

23

The list can be used to identify the following structural and environmental
control synergy cases. The Shading Fin Mass Damper (SFMD) example is identified
by realizing the dual use of the shading fins as mass dampers. The combined control
system is designed and analyzed from both structural and environmental aspects in
the next chapters. Improvement of structural strength and energy saving is also
mentioned.

4.5 Case studies

There are not many built examples of structural and environmental control
synergy. In fact, the following examples were built by the same firm around the
same time. Nonetheless, the combined systems illustrate synergy in the integrated
SC and EC systems.

4.5.1 Sendagaya INTES Building in Tokyo (1991)

An example application of the synergy system is the Sendagaya INTES
Building (Figure 4.1) in Tokyo designed by Takenaka Corporation in 1991. The
building is 11 stories tall and contains 10,602m
2
in floor area. On the top floor, there
are two ice thermal storage tanks serving as Hybrid Mass Dampers (HMDs) to
control transverse and torsional motions of the structure (Figure 4.2). Hydraulic
actuators are also placed to provide the active control capabilities. Using the ice
thermal storage tanks as mass dampers avoids introducing extra weight to the
structure, as typically required for mass dampers.

24

Figure 4.1: Sendagaya INTES Building (Sendagaya 2003)

Figure 4.2: Sendagaya INTES building with hybrid mass dampers (Higashino and Aizawa 1993)

4.5.2 Crystal Tower in Osaka (1990 by Nagase & Hisatoku)
The Takenaka Corporation also designed the Crystal Tower (Figure 4.3) in
Osaka in 1990 (Nagase and Hisatoku 1992). It is a 157m tall building weighing
44,000 metric tons. Instead of using the ice thermal storage tanks as HMDs, the tanks
were used as pendulum weights. The six 90-ton pendulum dampers hang from the
roof girders with lengths of 4m and 3m. Oil dampers are connected to the pendulums
to dissipate energy caused by building movements. Under wind excitation, the
25

pendulum mass swings in sync with the sway of the building, reducing the
movement of the building.

Figure 4.3: Crystal Tower (Crystal Tower, 2003)
26

Chapter 5 – Shading Fin Mass Damper System

This thesis introduces a shading-fin-mass-damper (SFMD) system as a
synergy example between SC and an architectural system. Figure 5.1 shows a plan
view of the system; Figure 5.2 shows an over-view; and Figure 5.3 shows the front
and section views of the SFMD system. The movable shading fins can adjust
positions to allow or block sunlight into the building. Moreover, the fins also serve
as mass dampers (see Figure 5.4) which move and damp energy out of the structure
to reduce vibration. The shading fins may also affect the aerodynamics of the
building but this is a subject beyond the scope of the thesis.

Figure 5.1: Plan view of the Shading Fin Mass Damper system
27

Figure 5.2: Overview of the Shading Fin Mass Damper system

Figure 5.3: Shading Fin Mass Damper system: (left) front; (right) section
28

Figure 5.4: Shading Fin Mass Damper system: (left) details; (right) outside details

5.1 Compatibilities between Mass Damper and Shading Fin

Although tuned mass dampers (TMDs) and active shading fins (ASFs) have
generally different objectives, they are indeed compatible. As discussed in Chapter
4, structural control (SC) and environmental control (EC) systems are generally
compatible with each other but further examinations of the physical compatibility are
needed for individual cases. In the SFMD system, the movements of mass dampers
require shading fins to be movable whereas typical exterior shading fins are fixed.
Movable fins can adjust the amount of direct sunlight coming into the building
according to the lighting condition and temperature inside the building. On the other
hand, by using movable shading fins, the cost of the shading-fin-mass-damper
(SFMD) system can be more easily justified since the mass dampers are not often
used (only during strong building vibration). Moreover, the increased number of
29

mass dampers makes the SFMD system more flexible to a wider range of excitations
than conventional single TMD systems. Therefore, although the physically
compatibility of TMD and shading fins may not be obvious, special and careful
design considerations can effectively integrate the SC and EC systems.
30

Chapter 6 – Mass Damper System

This chapter discusses the structural control aspect of the Shading Fin Mass
Damper system. Due to the integration with the shading fins, the mass damper
system adopts a special configuration — called the Distributed Mass Damper (DMD)
system — that will be studied in this chapter. The performance of the DMD system
is examined and compared with the traditional tuned mass damper system. Although
some characteristics of the DMD system can be tuned to optimize performance, they
may also be difficult to implement from the point of views of building design and
construction. A simpler DMD system configuration is proposed that be more
practical for the Shading Fin Mass Damper system while sacrificing only a small
degree of performance.

6.1 Distributed Mass Damper

Since shading fins are placed along the height of the building, the proposed
SFMD system behaves differently than traditional TMD or multiple tuned mass
damper (MTMD) systems (Figure 6.1). The mass dampers are distributed
throughout the height of the building, instead of attaching only the top floors for
tuned mass damper(s).
31

Figure 6.1: DMD, TMD, MTMD diagrams.

The equations of motions for the n-story structure with a DMD can be expressed as
f M1 Kx x C x M + − = + +
g
x & & & & & (6.1)

where
⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎣
⎡
−
−
−
=
d d
d
2
d
2
2
d
1
d
1
1
0 0 0 0
0
0 0 0
0 0 0 0
0 0 0
0 0 0 0
n n
n
m m
m
m m
m
m m
m
L
M O M
L
M ,

⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎥
⎦
⎤
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎢
⎣
⎡
− −
− − + −
− − +
=
−
d
d
1
d
2
3
d
2 3 2 2
d
1
2
d
1 2 1
0 0 0 0 0
0 0 0
0 0 0 0 0
0 0
0 0 0 0 0
0 0 0
n
n n n
k
k k k
k
k k k k k
k
k k k k
K
M O M
L
K ,
32

C takes a similar form as K, [ ]
T
d d
2 2
d
1 1 n n
x x x x x x L = x , x
g
is the ground
displacement, f is the external force vector of the system (e.g., wind) and 1 is a
column vector of ones. Here, m
i
and
d
i
m are the masses of the i
th
floor and of the
damper attached to the i
th
floor, respectively, k
i
and
d
i
k are the stiffness coefficients
of the i
th
floor and between the i
th
floor and the i
th
damper, respectively, and x
i
and
d
i
x are the i
th
floor displacements relative to the ground and the damper displacement
relative to the i
th
floor respectively. The following basic structural terms are useful
in describing dynamic structural systems:
• i
th
floor frequency (for standalone floor):
i i i
m k w =
• i
th
floor damping ratio (for standalone floor): ( )
i i i i
m w c 2 = ς
• i
th
floor damper frequency (for rigid structure):
d d d
i i i
m k w =
• i
th
floor damper damping ratio (for rigid structure): ( )
d d d d
2
i i i i
m w c = ς

6.2 Formulation / Simulation Model

The state space representation of (6.1) is
⎥
⎦
⎤
⎢
⎣
⎡
⎥
⎦
⎤
⎢
⎣
⎡
−
+
⎥
⎦
⎤
⎢
⎣
⎡
⎥
⎦
⎤
⎢
⎣
⎡
− −
=
⎥
⎦
⎤
⎢
⎣
⎡
− − −
f M 1 x
x
C M K M
I 0
x
x
1 1 1
g
x & &
& & &
& 0 0
(6.2)
or Bu Az z + = & , where u is the input of the system. For earthquake loads, f is
assumed to be zeros and u only depends on the ground acceleration. Four
earthquake records are used for this analysis (Figure 6.2) (Ramallo et al. 2002):
33

• El Centro — north-south component of the 1940 Imperial Valley,
California earthquake (magnitude 7.1) recorded at Imperial Valley
Irrigation District substation in El Centro, CA;
• Hachinohe — north-south component of the 1968 Takochi-oki (Hachinohe)
earthquake (magnitude 7.9) signal recorded at Hachinohe City, Japan;
• Kobe — north-south component of the 1995 Hyogo-ken Nanbu (Kobe)
earthquake (magnitude 7.2) recorded at the Kobe Japanese
Meteorological Agency (JMA), Kobe, Japan;
• Northridge — north-south component of the 1994 Northridge earthquake
(magnitude 6.8) recorded at the Sylmar County Hospital parking lot in
Sylmar, CA.
A Kanai-Tajimi shaping filter (Soong and Grigoriu 1993) is then fit (Ramallo et al.
2002) to on the design earthquakes to simplify simulation (Figure 6.3).
0 5 10 15 20 25 30
−5
0
5
Northridge

0 5 10 15 20 25 30
−5
0
5
Hachinhe

0 5 10 15 20 25 30
−5
0
5
time (sec)
Kobe

0 5 10 15 20 25 30
−5
0
5
El Centro
Absolute Acceleration (m/s
2
)

Figure 6.2: Design earthquakes Figure 6.3: Frequency content of design
earthquakes and Kanai-Tajimi
filter (Ramallo et al. 2002)

34

6.3 Simple Case: Equally Distributed Dampers

To understand the behaviors of the DMD, consider an equally-distributed
mass damper (EDMD) system that has equal damper parameters (
d
i
m ,
d
i
k and
d
i
ς ) for
each floor. For a given structure with known m
i
, k
i
, c
i
and a fixed mass ratio,
∑ ∑
= =
=
n
i
i
n
i
i
m m m
1 1
d
, an optimal solution of the two damper parameters (
d
i
k and
d
i
ς )
can be found for the following objective function:
() [ ]
∑
=
−
− =
n
i
i i
x x E J
1
2
1
(6.3)
where x
0
= 0 because x
i
’s are relative to the ground, and () [ ]
2
1 −
−
i i
x x E is the
expected value of the squared floor-to-floor drift. Figure 6.4a shows how (6.3)
changes for different values of damper parameters (
d
i
k and
d
i
ς ) of the EDMD system
for a 20-story building and indicates a minimum value of J in the region. The
damper parameters (
i i
k k
d
=0.00025 and
d
i
ς =11%, i=1,…,20) associated with the
J
min
are the optimal parameters illustrated in Figure 6.5a. Figure 6.5 compares
performance of the EDMD system and a single TMD system with the damper at the
top floor on the 20-story building. The parameters (
n n
k k
d
=0.00025 and
d
n
ς =15%,
n=20) for the single TMD in Figure 6.5b are obtained by finding the parameters
associated with the J
min
in Figure 6.4b.
35

0
0.1
0.2
0
0.5
1
x 10
−3
3
3.5
4
4.5
5
5.5
6
6.5
7
x 10
−3
ζ
d
i
(a) Equally Distrib. Dampers
k
d
i
/ k
i
J = sum of E[drifts
2
]
0
1
2
0
0.005
0.01
2.5
3
3.5
4
4.5
5
5.5
6
x 10
−3
ζ
d
i
(b) Single Damper @ Top floor
k
d
i
/ k
i
J = sum of E[drifts
2
]

Figure 6.4: Objective function over damper parameters: (a) equally distributed mass damper
and (b) single mass damper located at the top floor

36

0 5 10 15 20
0
0.5
1
(a)Equally distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (40.5 % better)
no damper
0 5 10 15 20
0
0.5
1
(b)Single TMD system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (48.6 % better)
no damper

Figure 6.5: Comparison between the (a) equally distributed and (b) single damper systems.

Figure 6.5a shows that the optimal floor stiffness ratio,
i
d
i
k k / , is similar to the given
floor mass ratio,
i
d
i
m m / , while the damping terms of the EDMD are a little different
from the damping term of the single TMD. The single damper system in Figure 6.5b
reduces the objective function, J, by 48.5% compared to the same building without
dampers while the EDMD system reduces J by 35.1%. Although the single damper
system performs better in reducing drifts, the damper at the top floor weighs as much
as the floor itself, affecting significantly the structural and architectural design at the
floor. On the other hand, the EDMD weigh as little as 5% of each floor and
therefore allow more flexible integration with the building. However, the DMD
37

damper system affects the design of all floors while the single damper system affects
only one floor. Figure 6.6 demonstrates the performance of the two damper systems
on the four design earthquake records compared to the building without dampers.
The single damper system marginally outperforms the EDMD system for all four
earthquake records.
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Kobe

no damper
single TMD (69.3 % better)
distrib. damper (61.3 % better)
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Northride

no damper
single TMD (49.2 % better)
distrib. damper (34.0 % better)
0 10 20 30 40 50
0
0.002
0.004
0.006
0.008
0.01
0.012
time(s)
sum(drifts
2
) (ft
2
)
El Centro

no damper
single TMD (44.6 % better)
distrib. damper (36.1 % better)
0 10 20 30 40 50
0
2
4
6
8
x 10
−3
time(s)
sum(drifts
2
) (ft
2
)
Hachinhe

no damper
single TMD (26.8 % better)
distrib. damper (25.0 % better)

Figure 6.6: Performance of the EDMD and single TMD on different historical earthquake
records.

6.3.1: Fine Tuning the Parameters
To improve the performance of the DMD system, the parameters can be fine
tuned according to the single TMD and EDMD systems. To understand the
importance of dampers on different floors, Table 6.1 presents the performance of
38

single TMDs (5% of structural mass) on different floors and their corresponding
damper parameters chosen using similar methods as Figure 6.4a. The upper floors
yield the best performances in reduction of J, with the top floors reduces J by 48.5%.
The lower floors reduce significantly less drifts than other floors.
Floor, i 1 2 3 4 5 6 7 8 9 10
∑ i
d
i
m m / (%)
5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00
i
d
i
k k / (%)
0.51 0.53 0.53 0.53 0.53 0.52 0.51 0.51 0.50 0.49
d
i
ζ (%)
5.00 5.00 5.00 5.00 5.00 10.0 10.0 10.0 10.0 10.0
Reduce in J (%) 3.33 8.89 15.0 20.3 24.4 27.9 31.7 34.7 37.1 39.0

Floor, i 11 12 13 14 15 16 17 18 19 20
∑ i
d
i
m m / (%)
5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00 5.00
i
d
i
k k / (%)
0.49 0.47 0.47 0.46 0.46 0.46 0.46 0.45 0.45 0.46
d
i
ζ (%)
10.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0
Reduce in J (%) 40.5 41.9 43.3 44.6 45.6 46.6 47.4 48.0 48.4 48.5
Table 6.1: Performances and parameters for single TMDs on different floors.

1
2
3
4
0
5
10
15
20
25
0
10
20
30
40
50
floor,i
%
k
d
i
/k
i
m
d
i
/m
i
ζ
d
i
J

Figure 6.7: Performance and parameters for single TMDs on different floors (Table 6.1).
39

Table 6.1 and Figure 6.7 clearly show that the TMDs on the upper floors reduce the
more drifts compared to the lower floors. In the DMD system, the damper masses
can be divided according to the “importance” of the floor location from Table 6.1
and Figure 6.7. For example, the top floor should have the heaviest damper since it
would better reduce the drifts while the first floor should have the lightest damper.
Following such logic, the damper masses shown in Figure 6.8 are computed by
( )
()
∑
=
=
n
i
s i
s i
i
d
i
J
J
n
m
m
1
~
~
05 . 0 , for i = 1, 2, …, n. (6.4)
where n is the number of floors, 0.05 is from 5% of mass damper ratios, and ( )
s i
J
~
is
the reduction in J in i
th
floor for the single TMD shown in Table 6.1. From Figure
6.5, the stiffness ratios are seen to be proportional to the damper mass when
comparing the stiffness ratios between the single TMD and EDMD systems.
Therefore, the stiffness parameters in Figure 6.8 are calculated using
n
m m
k
k
k
k
i
d
i
s
i
d
i
i
d
i
05 . 0
⎟
⎠
⎞
⎜
⎝
⎛
= , for i = 1, 2, …, n. (6.5)
where
s
i
d
i
k
k
⎟
⎠
⎞
⎜
⎝
⎛
is the stiffness ratios of the single damper systems from Table 6.1.
Lastly, the damping ratios in Figure 6.8 are set to the values from Table 6.1 since
Figure 6.5 suggests little changes in damping ratios between the single TMD and
EDMD systems.
40

0 5 10 15 20
0
0.05
0.1
(a)distrib. damper system (all floors)
m
d
i
/ m
i
0 5 10 15 20
0
5
x 10
−4
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (42.9 % better)
no damper
0 5 10 15 20
0
0.05
0.1
(b)distrib. damper system (top half floors)
m
d
i
/ m
i
0 5 10 15 20
0
5
x 10
−4
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (46.1 % better)
no damper

Figure 6.8: Comparison between DMD systems of all floor s(a) and top half floors only (b).

Figure 6.8 shows the improvement from the EDMD system using (6.4) and
(6.5). In fact, the DMD system described (Figure 6.8a) only suffers a small drop
performance compared to the single TMD in Figure 6.5b (42.9% compared to 48.5%
respectively). The advantage of the DMD systems shown is that the dampers can be
fairly distributed throughout the floors instead of being concentrated in one floor.
Figure 6.8b also shows a variation where only the top half of the floors is equipped
with dampers. The performance is very much in line with the single TMD in Figure
6.5b with only 2.4% drop off. This shows how little structural performance is
sacrificed to incorporate the DMD system for shading fins. Since the DMD system
with damper in all floors in Figure 6.8a is obtained by weighing the damper masses
41

and stiffness according to (6.4) and (6.5), it will be referred as the “weighed EDMD”
system and Figure 6.9 illustrates the performance of this system subject to the four
design earthquake records. Figure 6.10 checks that the performance of the weighed
EDMD system is also good for excitations other than the design records; the results
show that the weighed EDMD system perform closely to the single TMD system.
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Kobe

no damper
single TMD (69.3 % better)
distrib. damper (62.8 % better)
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Northride

no damper
single TMD (49.2 % better)
distrib. damper (40.5 % better)
0 10 20 30 40 50
0
0.002
0.004
0.006
0.008
0.01
0.012
time(s)
sum(drifts
2
) (ft
2
)
El Centro

no damper
single TMD (44.6 % better)
distrib. damper (38.5 % better)
0 10 20 30 40 50
0
2
4
6
8
x 10
−3
time(s)
sum(drifts
2
) (ft
2
)
Hachinhe

no damper
single TMD (26.8 % better)
distrib. damper (26.0 % better)
Figure 6.9: Performance of the weighed DMD system on the four designed earthquake records.

42

0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
0.25
time(s)
sum(drifts
2
) (ft
2
)
Newhall

no damper
single TMD (74.5 % better)
distrib. damper (67.8 % better)
0 10 20 30 40 50
0
0.02
0.04
0.06
0.08
0.1
0.12
time(s)
sum(drifts
2
) (ft
2
)
Jiji

no damper
single TMD (58.2 % better)
distrib. damper (53.6 % better)
0 10 20 30 40 50
0
0.02
0.04
0.06
0.08
0.1
0.12
time(s)
sum(drifts
2
) (ft
2
)
Erzincan

no damper
single TMD (40.4 % better)
distrib. damper (38.1 % better)
0 10 20 30 40 50
0
0.5
1
1.5
2
2.5
3
x 10
8
time(s)
sum(drifts
2
) (ft
2
)
Random

no damper
single TMD (48.1 % better)
distrib. damper (43.2 % better)

Figure 6.10: Performance of the weighed EDMD system on the other earthquake records and
random excitation.

6.4 Optimization of Damper Parameters

In order to design an effective DMD system, the damper parameters must be
optimized for vibration reduction. The weighed EDMD system from last section
shows that DMD system can significantly reduce building motions, though not as
well compared with the conventional single TMD systems. The section tries to
improve the DMD system by optimization techniques to find an optimal set of
damper parameters that minimizes the damage caused by the input excitation. The
parameters are
•
d
i
m

, i = 1,2, …, n (masses of the n dampers)
43

•
d
i
ζ

, i = 1,2, …, n (damping ratios of the n dampers: ( )
d
i
d
i
d
i
d
i
m k c 2 = ζ )
•
d
i
k , i = 1,2, …, n (stiffness between the n dampers and the corresponding
floors).
The ranges of these parameters are defined as the following:
Mass: 0 ≤
d
i
m ≤ 15% x m
i
(each damper mass must be less than 15% of the
corresponding floor mass)
∑
d
i
m ≤ 5% x ∑ m
i
(sum of the damper masses must be less than 5% of the
total structural mass)
Damping: 0 ≤
d
i
ζ ≤ 200% (each damper ratio must be less than 200%)
Stiffness: 0 ≤
d
i
k ≤ k
i
(each damper stiffness must be less than floor
stiffness).

Difficulty on this optimization problem:

For an n-story building, there are 3n variables to optimize (damper mass,
stiffness and damping of each damper). With tall buildings being most suitable for
the SFMD system, a large number of variables are being considered for optimization.
For example, the 20-story building discussed in the last section has 60 variables to
optimize.

6.4.1 Pattern Search Optimization

Pattern search is originally a particular set of direct search optimization
methods first proposed by Hooke and Jeeves (1961). Recent developments by
44

Dennis and Torzcon (1991) and Torzcon (1989, 1997) have generalized the method.
One major reason that the pattern search method is used for the damper optimization
is that the method can handle many variables without occupying an enormous
amount of computational resources. But like many other optimization methods,
pattern search is subject to getting stuck at local minima and generally produces
different optimization results with different initial values for the damper parameter
problem. MATLAB® includes a number of routines for the pattern search methods
in its Genetic Algorithm Toolbox and the damper parameters are optimized using
these routines. The following briefly explains how pattern search works:
Consider a problem minimizing an object function f : R
n
Æ R. Let x be the
set of n variables to be optimized and x
0
and x
k
be the initial values and the values at
the k
th
iteration respectively. At iteration k, x
+
= x
k
+ ∆
k
d, where ∆
k
is the weight
and d is the directions applied to x. There are two potential directions for each
variable x
i
(i = 1,2, …, n) — positive and negative (increasing and decreasing the
value of x
i
respectively). Thus there are 2n directions of x if each x
i
is looked at
independently; collectively, there are n
2
possible combinations of x
+
. After
exhausting all combination of x
+
(2n, n
2
or any other number of x
+
depending on the
method applied), if ) ( ) ( min
k
f f x x
x
<
+
+
then iteration k is a success and the algorithm
moves to iteration k+1 with x
k+1
= x
+
and an increased ∆
k+1
(typically ∆
k+1
= 2 ∆
k
).
Otherwise, an unsuccessful iteration k would lead to x
k+1
= x
k
and a decreased ∆
k+1

(typically ∆
k+1
= ∆
k
/2) for the next iteration. The process is repeated until a preset of
stopping criteria is met.
45

6.4.2 Optimization Results

The optimization problem is expected to have many local minima, possibly
caused by so many damper parameters to optimize. Therefore, the optimized result
of pattern search is heavily dependent on the initial set of parameters. Figures 6.11
and 6.12 show two results of pattern search from two initial sets of randomly chosen
parameters. The initial DMD systems perform worse (larger drifts) than the base
system (no damper) in terms of drifts, and the optimized DMD systems outperform
the base system by 32.1% and 27.8%, respectively. In both cases, many of the
damper masses of the DMD systems are reduced to zeros and the optimized DMD
systems only have a few dampers (4 dampers and 3 dampers, respectively).
Dampers that are not properly tuned cause increase in drifts by adding mass to the
system. Since pattern search optimization looks for the largest improvement of the
overall system at each iteration, pattern search can select either improvement through
correctly tuning the damper stiffness and damping ratios or improvement through
reducing damper masses. In the two cases from Figure 6.11 and 6.12, pattern search
favors improvement by reducing damper mass except for a few floors. This implies
that, most of time, reduction in damper masses shows greater improvements than
tuning damper stiffness and damping ratios for the initial DMD systems in Figure
6.11 and 6.13.

46

0 5 10 15 20
0
0.1
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.05
0.1
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (11.3 % worse)
no damper
0 5 10 15 20
0
0.1
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.5
1
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (32.1 % better)
no damper

Figure 6.11: Pattern search optimization with randomly chosen initial parameters (1).

0 5 10 15 20
0
0.1
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.05
0.1
k
d
i
/ k
i
0 5 10 15 20
0
0.05
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (9.4 % worse)
no damper
0 5 10 15 20
0
0.1
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.005
0.01
k
d
i
/ k
i
0 5 10 15 20
0
0.05
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (27.8 % better)
no damper

Figure 6.12: Pattern search optimization with randomly chosen initial parameters (2).
47

Figures 6.13 and 6.14 show two initial DMD systems from Section 6.3 — the
EDMD and weighed EDMD. Before optimizing, they already outperform the
structure alone by 40.5% and 42.9%, respectively. The optimized results of the
pattern search shows improvements of 46.3% and 48.8%, respectively, and 48.8% is
slightly better than the 48.5% improvement from the best single TMD system
(damper on the top floor). The two optimized DMD systems place small or no
dampers at the lower floors, agreeing with Table 6.1 that shows smaller effects from
dampers at lower floors. The optimized DMD systems have small damping ratios
for all floor compared to the initial DMD systems, while damper stiffness remains
similar to the ones of the initial systems except some increases for a few floors.
Interestingly, the floors around 5
th
- 7
th
floors shows the large increases in damper
stiffness for both systems in Figure 6.13 and 6.14. The stiffness jumps could be
caused by tuning to the 2
nd
natural frequency, that equals to 14.40 rad/s, of the
uncontrolled structure. In Figure 6.13,
d d
m k
6 6
= 13.42 rad/s and
d d
m k
7 7
=
14.80 rad/s while
d d
m k
5 5
= 13.39 rad/s and
d d
m k
7 7
= 14.60 rad/s in Figure
6.14.
Figure 6.15 and 6.16 show how the optimized DMD system in Figure 6.14
performs on various types of ground excitations. Although Figure 6.14 shows
slightly better performance of the optimized DMD system compared to the single
damper system, the DMD system only outperforms the single damper system for the
Erincan earthquake record.

48

0 5 10 15 20
0
0.05
0.1
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (40.5 % better)
no damper
0 5 10 15 20
0
0.05
0.1
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (46.3 % better)
no damper

Figure 6.13: Pattern search optimization with the EDMD system as the initial guess.

0 5 10 15 20
0
0.05
0.1
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

0 5 10 15 20
0
0.05
0.1
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (42.9 % better)
no damper
with damper (48.8 % better)
no damper

Figure 6.14: Pattern search optimization with the weighed EDMD system as the initial guess.
49

0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Kobe

no damper
single TMD (69.3 % better)
distrib. damper (67.0 % better)
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
time(s)
sum(drifts
2
) (ft
2
)
Northride

no damper
single TMD (49.2 % better)
distrib. damper (48.5 % better)
0 10 20 30 40 50
0
0.002
0.004
0.006
0.008
0.01
0.012
time(s)
sum(drifts
2
) (ft
2
)
El Centro

no damper
single TMD (44.6 % better)
distrib. damper (43.4 % better)
0 10 20 30 40 50
0
2
4
6
8
x 10
−3
time(s)
sum(drifts
2
) (ft
2
)
Hachinhe

no damper
single TMD (26.8 % better)
distrib. damper (24.2 % better)

Figure 6.15: Performance of the optimized DMD system (from Figure 6.14) subject to the four
designed earthquake records.
0 10 20 30 40 50
0
0.05
0.1
0.15
0.2
0.25
time(s)
sum(drifts
2
) (ft
2
)
Newhall

no damper
single TMD (74.5 % better)
distrib. damper (70.9 % better)
0 10 20 30 40 50
0
0.02
0.04
0.06
0.08
0.1
0.12
time(s)
sum(drifts
2
) (ft
2
)
Jiji

no damper
single TMD (58.2 % better)
distrib. damper (55.1 % better)
0 10 20 30 40 50
0
0.02
0.04
0.06
0.08
0.1
0.12
time(s)
sum(drifts
2
) (ft
2
)
Erzincan

no damper
single TMD (40.4 % better)
distrib. damper (43.0 % better)
0 10 20 30 40 50
0
0.5
1
1.5
2
2.5
3
x 10
8
time(s)
sum(drifts
2
) (ft
2
)
Random

no damper
single TMD (48.1 % better)
distrib. damper (47.4 % better)

Figure 6.16: Performance of the optimized DMD system (from Figure 6.14) subject to the other
earthquake records and random excitation.

50

6.5 Discussion of Performance and Practicality

From the optimization results, the DMD system can be optimized to perform
on par with the traditional single damper system. However, the optimized DMD
systems may not be ideal for building design and construction with the damper
masses being wildly different from floor to floor. On the other hand, the EDMD and
the weighed EDMD systems in Section 6.3 have rather simple damper masses
distributions (uniform or gradually increasing respectively), and thus are probably
more practical for building designers and builders. The down side of the EDMD
systems is the drop in performance compared the optimized DMD systems (Figure
6.13 and 6.14).
Excitation
Improvement relative to structure with no dampers
Weighed EDMD
system
Optimized
DMD system
Single TMD
system (top floor)
Kobe 63.50% 67.00% 69.30%
Northridge 41.00% 48.50% 49.20%
El Centro 39.10% 43.40% 44.60%
Hachinhe 26.10% 24.20% 26.80%
Newhall 67.80% 70.90% 74.50%
Jiji 53.60% 55.10% 58.20%
Erzincan 38.10% 43.00% 40.40%
Random (KT-filter white noise) 43.20% 47.40% 48.10%
Table 6.2: Performance of the weighed EDMD and optimized DMD system on different
excitations

From Table 6.2, though there is a clear performance gap between the
weighed EDMD and optimized DMD systems, the gap varies between the simulated
excitations with an average difference at about 3%. Since future earthquakes will
behave differently than the historical records, simplicity in the actual implementation
51

of the DMD systems maybe more practical than a slight increase in performance.
Therefore, DMD systems with a simple damper mass distribution (such as the
EDMD system) are more suitable for the proposed Shading Fin Mass Damper
system.

6.5.1 Sub-optimization of Damper Stiffness and Damping Only

Furthermore, the DMD systems can improve performance by sub-optimizing
the damper stiffness and damping ratios while leaving the damper masses unchanged
(thus keeping the whole system more constructible). Figures 6.17, 6.18 and 6.19
show improvements with the sub-optimization with initial guesses as the EDMD,
weighed EDMD and a stepped EDMD systems, respectively. The stepped EDMD
system divided floors into groups (i.e., 1
st
to 5
th
floors as a group and 6
th
to 10
th
as
another group) in the similar fashion where high-rise buildings are designed and built
by grouping together adjacent floors. Figure 6.17 with the EDMD initial guess
reduces the least drifts (42.5% improvement from structure alone) in the three
systems, though it would be the easiest to build due to the uniform damper masses.
Weighed EDMD (Figure 6.18) performs the best (47% improvement) but it is the
hardest to build with different damper masses each floor. The stepping system in
Figure 6.19 performs closely to the weighed EDMD system with an improvement of
46.3%; this is also very close the 48.6% improvement from the best performer — the
single TMD. The fine performance of the stepped EDMD suggests that only a little
sacrifice is needed for greater practicality.
52

0 5 10 15 20
0
0.02
0.04
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−4
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
s
)
floor, i
with damper (40.5 % better)
no damper
0 5 10 15 20
0
0.02
0.04
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
2
4
x 10
−4
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
s
)
floor, i
with damper (42.5 % better)
no damper

Figure 6.17: Pattern search sub-optimization with the EDMD as the initial guess.

0 5 10 15 20
0
0.02
0.04
0.06
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.5
1
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
0.15
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
s
)
floor, i
with damper (42.9 % better)
no damper
0 5 10 15 20
0
0.02
0.04
0.06
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.5
1
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.05
0.1
0.15
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
s
)
floor, i
with damper (47.0 % better)
no damper

Figure 6.18: Pattern search sub-optimization with the weighed EDMD as the initial guess.
53

0 5 10 15 20
0
0.05
Initial distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.5
1
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (42.8 % better)
no damper
0 5 10 15 20
0
0.05
Optimized distrib. damper system
m
d
i
/ m
i
0 5 10 15 20
0
0.5
1
x 10
−3
k
d
i
/ k
i
0 5 10 15 20
0
0.1
ζ
d
i
0 5 10 15 20
0
0.5
1
x 10
−3
E[drift
i
2
] (ft
2
)
floor, i

with damper (46.3 % better)
no damper

Figure 6.19: Pattern search sub-optimization with the stepped EDMD as the initial guess.

Another important aspect of the sub-optimization of the damper stiffness and
damping ratios is the inclusion of a non-passive control system. Passive control
systems have the control parameters tuned and fixed while non-passive systems vary
the parameters according the condition of the system. For the DMD system, stiffness
and damping varying devices can be installed with the dampers to become an active
or semi-active control system. Such a non-passive system can be easy to build while
achieving excellent performance.

54

Chapter 7 – Shading Fin

Similar to a TMD, shading fins are typically passive devices that are fixed on
building façades. However, by allowing shading fins to be movable (through
mechanical means), the system can adjust to the weather and to sun orientation.
Moreover, movable fins can integrate with a mass damper system to achieve synergy
between structural and environmental control systems. This chapter discusses the
effectiveness of movable shading fins from the Shading Fin Mass Damper (SFMD)
system.

7.1 Movable Shading Fin

Due to the motion requirement of the mass dampers, the shading fins must be
movable. There are typically two types of shading fin movements —
protracting/retracting and rotating (Figure 7.1). Movable fins can track sun paths for
different hours in a day and for different seasons in a year (Figure 7.2). More about
the movable fins will be discussed later this chapter. There are several buildings that
adopt movable shading devices. One of the oldest examples is the Hall of Records
Building (Figure 7.3) in Los Angeles designed by Richard Neutra in 1962; a more
recent example is the Caltrans District 7 Headquarters Building (Figure 7.4) in Los
Angeles by Thom Mayne in 2004. The tall vertical shading fins on the south side of
Hall of Records Building rotate throughout the day to block direct sunlight, trying to
decrease heat gain in the warm southern California climate. The “flappable” shading
devices (due to the slapping motion of the devices) in the Caltrans Building are
55

placed in the east and west sides of the building. The shades flap open and close to
control heat gain from direct sunlight in winters and summers, morning and
afternoon, respectively.

Figure 7.1: Movements of shading fins Figure 7.2: Sun paths for summer and winter
(plan view). (Alward and Shapiro 1981).

Figure 7.3: Hall of Records, Los Angeles (1962) Figure 7.4: Caltrans District 7
by Richard Neutra (photo: Martin 1995). Headquarter, Los Angeles (2004) by
Thom Mayne (photo: Halbe 2004).

7.1.1 Simulation Model

A 3-story office building model (Figure 7.5) is used to analyze the effect of
different shading movements using eQuest (version 3.5) — the QUick Energy
Simulation Tool for building. The building is 150ft by 60ft located in Los Angeles
(warm climate region), with the longer dimension in the north-south direction. The
Pro/retracting Rotating
56

windows and fins are located only on the east and west faces of the building. The
HVAC system of the building is automatically chosen by eQuest to satisfy the
energy consumption profile of the building. Figure 7.6 compares the effect of having
5ft fixed fins with no fins at all on the office building. As shown, the building with
fins improves energy efficiency in most of the months. For more discussion on
economy of shading devices, the reader is referred to a book, Solar Control and
Shading Devices, by Olgyay and Olgyay (1957).

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2400
2600
2800
3000
3200
3400
3600
Energy Cost($)
5ft fin
no fin

Figure 7.5: eQUEST building model. Figure 7.6: Effect of 5ft shading fins.

7.1.2 Protracting / Retracting Shading Fin

Protracting and retracting shading fins have simple effects on sunlight
(Figure 7.7). Retracted fins allow more sunlight to enter the building while
protracted fins allow less. Figure 7.8 shows the monthly energy consumption
profiles for shading fins at different lengths. Since cooling is the largest portion of
the electric consumption, longer fins outperform shorter ones in electricity because
longer fins allow less sunlight (solar heat gain) into the building. On the contrary,
buildings with longer fins use more gas for heating due to the reduction of sunlight.
57

In most months, the energy cost is lower for longer fins because electricity currently
costs more than gas (i.e., it is usually cheaper to heat a space using gas than cooling
it using electricity for the same amount of changes in temperature). However, the
energy cost is higher in winter months for longer fins because the building is using
more heating than cooling. This study of long and short fins shows that buildings
should, generally, protract shading fins during summer and retract them during
winter to minimize energy. The daylighting effect is also included in the study, but
the lighting loads do not differ significantly between different fin lengths.

Figure 7.7: Sunlight effect for short and long vertical shading fins in plan view.

Retracted Fins Protracted Fins
N
summer sunrise
winter sunrise
summer sunrise
winter sunrise
58

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
10
20
Energy Consumption
Electricity (MWh)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
20
40
60
Gas (MBtu)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
1
2
3
Cost($x000)

5ft fin
3ft fin
1ft fin
no fin

Figure 7.8: Effect of protracting shading fins.

7.1.3 Rotating Shading Fin

Rotating shading fins have more complicated effects on sunlight (Figure 7.9).
Fins that are rotated toward the north let in summer early morning (and late
afternoon) sun while blocking off all winter sun into the building. On the other hand,
fins that are rotated toward south let in summer late morning sun (and early
afternoon) sun while letting in some winter early morning (and late afternoon) sun.
59

Figure 7.9: Sunlight effect for rotated vertical shading fins in plan view

Figure 7.10 shows the monthly energy consumption profiles for shading fins
at different angles and directions. Fins rotated 45
o
toward the north cause the least
electricity usage because less cooling is needed since the sun is blocked in late
morning to early afternoon when the day is at its hottest temperature. For gas
consumption, north facing fins typically perform in similar levels as 90
o
fins and no
fins because all these cases let in sunlight in the early mornings when the building
needs to heat up for the beginning of office hours. In contrast, south facing fins
cause higher gas demand during the summer months because they block sunlight in
the early mornings. Nevertheless, the south facing fins outperform other
configurations in gas consumption for winter months. The cold weather requires
heating throughout the day and south facing fins let in sunlight during the late
morning and early afternoon when the sun is producing the most solar heat. This
study of fin orientations shows that buildings should rotate shading fins toward the
north for most of the year, but should also rotate them to track the sun in cold
weather to minimize energy cost. Here the energy saving (28.6% per year) is
North Facing Fins South Facing Fins
N
summer sunrise summer sunrise
winter sunrise winter sunrise
60

substantially larger than the pro/retracting fins, though the rotating motion may be
harder to incorporate with the mass damper motion.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
10
20
Energy Consumption
Electricity (MWh)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
20
40
60
Gas (MBtu)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
1
2
3
Cost($x000)
Months

45
o
fin (north)
45
o
fin (south)
90
o
fin
no fin

Figure 7.10: Effect of rotating shading fins

Both studies of pro/retracting and rotating fins demonstrate the benefits of
movable fins. The next phase of research will focus on simulating energy
consumption profiles for shading fins that actively adjust throughout the day and
year.

61

7.2 Active Shading Fins

Figure 7.11: eQuest model for active rotating shading fins.

To simulate the effect of active shading fins on building energy consumption,
the eQuest simulation model is used with some modification. Since eQuest treats the
shading elements as physically fixed, shading schedules are applied on the fins to
“move” them throughout the year. Shading schedules in eQuest deal with shading
elements that behave differently throughout the year, such as trees with their leaves
falling off in winters. To mimic the movements of active shading fins, a building
model is equipped with multiple fins at each location and each fin is turned
“transparent” or “solid” by the shading schedule. For example, an active rotating fin
model would have multiple shading fins at various angles (45
o
North, 90
o
and 45
o

South) at each location (Figure 7.11). If sunlight is desirable throughout the day,
such as in a cold summer day, the following fin arrangement would be placed:

62

45
o
North Fins 90
o
Fins 45
o
South Fins
Early morning S T T
Mid morning T S T
Late morning T T S
Noon T T S
Early afternoon T T S
Mid afternoon T S T
Late afternoon S T T
S = Solid T = Transparent
Table 7.1: Shading schedule arrangement

The shading schedule arrangement only has one fin solid at a time and thus mimics
rotating fins that track the sun path, allowing the maximum amount of direct sunlight
into the building.

7.2.1 Shading Schedule in eQuest
The limitation of the shading schedule is that it only adjusts the solar-
transmittance of the fins and leaves the visibility-transmittance constant. In other
words, the shading schedule adjusts heat-gain due to sunlight but not the lighting
effect. This makes daylight calculation unreliable. Daylight calculation assumes
that there are light sensors in the building tracking natural lighting throughout the
day. The amount of electric lighting needed can be decreased in daytime with the
presence of both direct and indirect sunlight (indirect sunlight is more useful for
daylighting because direct sunlight causes glaring problem and therefore is diffused
or reflected for daylighting).
Table 7.2 shows the effects of daylight calculations without shading
schedules, combining space cooling and heating in terms of monetary cost.
63

However, it should be noted that space cooling uses electricity while space heating
burns natural gas as energy sources. Electricity and natural gas differ in more than
their cost, but also in site and source energy consumption. Site energy is the energy
consumed on site, at the building location as an end-user. Source energy is the
energy cost to natural resources. Natural gas uses equal amounts of site and source
energy, meaning that there is no energy loss in converting from source to site energy.
In contrast, electricity uses significantly more source energy than site energy, or
there is a considerable amount of energy exhausted in producing the electricity to
provide to end-users from natural resources. This remains true as long as fossil fuels
are used to generate most electricity.

a) No daylight calculation
No Fin 90
o
Fins 45
o
South Fins 45
o
North Fins
Space Cooling / Heating ($) 22,064 19,176 (6.4%) 15,922 (13.6%) 12,378 (21.5%)
Electric Lighting ($) 14,848 14,579 (0.6%) 14,498 (0.8%) 14,522 (0.7%)
Misc. Equip. ($) 8,227 8,077 (0.3%) 8,032 (0.4%) 8,046 (0.4%)

b) With daylight calculation
No Fin 90
o
Fins 45
o
South Fins 45
o
North Fins
Space Cooling / Heating ($) 21,043 18,649 (6.5%) 14919 (16.7%) 11,466 (26.1%)
Electric Lighting ($) 7,354 6,822 (1.5%) 7,375 (-0.1%) 6,684 (1.8%)
Misc. Equip. ($) 8,294 8,166 (0.4%) 8,066 (0.6%) 8,090 (0.6%)
numbers in ( ) are savings in % of the total energy cost from No-Fin case

c) Daylight difference (%)
No Fin 90
o
Fins 45
o
South Fins 45
o
North Fins
Space Cooling / Heating 4.63% 2.75% 6.30% 7.36%
Electric Lighting 50.47% 53.20% 49.13% 53.97%
Misc. Equip. 0.82% 1.11% 0.43% 0.55%
Table 7.2: Daylight calculation effects

Changes in miscellaneous equipment can be neglected because such changes
are due to the difference in energy rates for peak and off-peak hours; the amount of
64

energy consumption by miscellaneous equipments remains the same for all cases.
As shown in Table 7.2c, daylight calculations clearly have profound effects on
electric lighting. In contrast, the effect of daylighting is smaller and less obvious on
space cooling and heating because the effect occurs indirectly when electric lighting
gives off heat while lightening the space.
Since the shading schedules make daylight calculation unreliable, eQuest can
only calculate accurately the energy effect of active shading fins without the daylight
calculation. In other words, electric lighting will be kept fairly constant throughout
all cases (similar to Table 7.2a) and eQuest is primarily computing the effect on
cooling and heating energy cost. Although the daylighting effect of active shading
fins on electric lighting savings cannot be computed, it should remain moderately
small (within 2% of the total energy cost) as suggested by Table 7.2b. Meanwhile,
savings in cooling and heating can be significantly larger (more than 25% in one
case). Moreover, Table 7.2a shows similar trends of energy saving in cooling and
heating for the different fin cases when compared to Table 7.2b. The difference in
percentage is largely due to the difference in total energy costs between the cases;
the actual cost savings are quite consistent between Table 7.2a and Table 7.2b.
Therefore, despite the lack of daylight calculation, the simulation of the active
shading fins by eQuest using shading schedules can be treated confidently while
being aware of its limitations.

65

7.2.2 Comparing Simulations with and without Shading Schedule
Since shading schedule in eQuest cannot accurately compute the daylighting
effects, the following comparison cases turn off daylight control such that all cases,
despite having different shading fin configuration, use the same amount of energy on
lighting. Thus the difference in energy consumption profiles is caused by the
difference in heat gain affected by the fins with or without the shading schedules.
Figure 7.12 shows that the energy costs of unshaded and shaded buildings with or
without using the shading schedule. For the shaded case using shading schedule, the
transmittance of the fin is set to be 0% such that the fin should be “opaque” like a
regular fin. On the other hand, the unshaded case has the transmittance is set to be
100% for the fin to be “transparent.” The shaded cases (with and without shading
schedule) yield identical electricity and gas costs, suggesting that the shading
schedule is working accurately for these cases. Nonetheless, the unshaded cases do
not match, casting doubts about how accurately the shading schedule calculates
“transparent” fins.
unshaded shaded
0
50
100
150
200
250
300
annual electricity cost(kWh)
unshaded shaded
0
50
100
150
200
250
300
350
400
annual gas cost(Btu)
without shading schedule
with shading schedule

Figure 7.12: Comparison of shading fins using shading schedule
66

Figure 7.13 compares the effects of shading schedules on different fixed
(non-movable) shading fin orientations on east and west windows (45
o
from north,
90
o
and 45
o
from south). For the cases that using shading schedules, all of the three
oriented fins are presented in the building model similar to Figure 7.11. The fins are
then turned “solid” or “transparent” using the shading schedules to mimic the cases
without shading schedules. From Figure 7.13, the cases with shading schedule
follow both the electricity and gas consumption pattern of the cases without
schedules. However, it is clear that the no fin cases are the least accurate, which
agrees with the observation from Figure 7.10. And for the cases with fins, although
the absolute energy cost is off, the errors are more consistent throughout different
orientations. In other words, the finned cases with shading schedules are able to
capture the changes on energy consumption due to different fin orientations. This
suggests that any further comparison study using shading schedules should have a
base case that has some “solid” fins using shading schedules. In fact, the cases with
fixed fins using shading schedules (Figure 7.13) are shown for the following
comparison study of actively rotating fins.

67

no fin 90 45(south) 45(north)
0
100
200
300
annual electricity cost(kWh)
fin orientation (degree)
no fin 90 45(south) 45(north)
0
100
200
300
400
500
annual gas cost(Btu)
fin orientation (degree)
without shading schedule
with shading schedule

Figure 7.13: Comparison of rotating fins using shading schedule

7.2.3 Actively Rotating Fin using Shading Schedules

To simulate the effect of actively rotating fins, a year-long shading schedule
is determined by trying to decrease the energy cost from the case of 45
o
North
oriented fins (the least energy consuming case). More specifically, every month of
the 45
o
North case is substituted with different shading schedules to determine one
that uses the minimum energy cost. A monthly interval is chosen instead of a shorter
(daily) or longer (seasonal) interval because weather changes too little over shorter
intervals to justify computational cost for the increased cases (i.e., 356 cases in daily
intervals instead of 12 in monthly intervals), and weather changes too much over
longer intervals for the determined shading schedule to be fully representative of the
current weather. There is another reason why a longer interval such as a seasonal
interval (3-month period) is not suitable for eQuest. Depending on the annual peak
demand of the energy profile on the simulated building model, eQuest automatically
68

chooses an appropriate HVAC system that satisfies the demand. And longer
intervals can cause too great of an effect in the energy profile of the overall system
that changes the size of the HVAC system in the simulation, which outweighs the
effects of the rotating fins in energy consumption.
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
10
20
Energy Consumption
Electricity (MWh)
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
20
40
60
Gas (MBtu)
electricity($x000) gas($x00) total($x000)
0
20
40
Anual Cost

Actively rotating fin
45
o
fin (north)
45
o
fin (south)
90
o
fin
no fin

Figure 7.14: Comparison of actively rotating and fixed fins using shading schedules

Table 7.3 details the shading schedule chosen for the entire year. In winter
months, since heat-gain is more desirable, the fins are more likely to orient toward
45
o
South where the sun is. In summer, when the sun is avoided to prevent
overheating, the fins are mostly oriented toward South in the morning/evening and
North for mid-days to counter the sun path in summer. In spring and fall, a
69

combination of embracing and avoiding direct sunlight is useful to minimize both
cooling and heating costs. Figure 7.14 compares the effectiveness of the actively
rotating fin to other fixed fin orientations with the shading schedule. The actively
rotating fin case is the most energy efficient system with a very significant advantage
in gas consumption. The advantage of the electricity consumption is less obvious
since it includes more than cooling loads, such as lighting and equipment that are
constant throughout different shading configurations.
The lighting loads are not affected by the active shading fins because the
daylighting effect is not considered in the shading schedule in Table 7.3 and Figure
7.14. As discussed earlier in this chapter, eQuest cannot accurately simulate the
daylighting effect with the solid fin shading schedule. If daylighting could be
included in the analysis, electricity consumption could be further optimized by
exploiting natural light and lessening the lighting loads. The changes would depend
on individual requirements for light, heat gain and the compromises between them.
Table 7.3 and Figure 7.14 illustrate the effect of active shading fins for a
warm climate (Los Angeles). If the study was to be performed for colder climates,
the shading fins are expected to welcome more direct sunlight compared to the
schedule in Table 7.3. Gas consumption would be a larger portion of the overall
energy consumption with increased heating loads.

70

Time
Fin Orientation
Time
Fin Orientation
45
o
N 90
o
45
o
S 45
o
N 90
o
45
o
S
Jan
5 ~ 8am x
Jul
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
Feb
5 ~ 8am x
Aug
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
Mar
5 ~ 8am x
Sept
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
Apr
5 ~ 8am x
Oct
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
May
5 ~ 8am x
Nov
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
Jun
5 ~ 8am x
Dec
5 ~ 8am x
8 ~ 10am x 8 ~ 10am x
10 ~12pm x 10 ~12pm x
12 ~ 2pm x 12 ~ 2pm x
2 ~ 4pm x 2 ~ 4pm x
4 ~ 8pm x 4 ~ 8pm x
Table 7.3: Shading schedule for the actively rotating fins throughout a year (x = oriented
direction).

71

Chapter 8 – Conclusion and Future Work

The proposed Shading Fin Mass Damper (SFMD) system efficiently
integrates structural and environmental controls. The two different control systems
are shown to be compatible and the structural and environmental control synergy
system can benefit both controls in the following ways. Structural control (SC) can
increase its presence and gain further consideration by building designers and
developers when it is well integrated with the required environmental control (EC).
Additionally, SC’s high cost can be partially deducted by the energy saving from the
EC component of the SESC system, provided that the synergy system is more energy
efficient than the EC system alone. This is a likely scenario because the structural
and environmental control synergy system would increase the importance and
resources for the EC system that couples with structural safety — one of the biggest
concerns in buildings. The EC component of the synergy system is no longer merely
a required system that sometimes is replaced by less competent design to lower
installation cost, neglecting the long term energy cost. Although the synergy system
can be costly initially, the combination of measurable energy saving and
immeasurable safety can convince many builders.
Structural and environmental impacts are the two measures of the
effectiveness of the structural and environmental control synergy system. The
SFMD system employs a distributed mass damper (DMD) system that doubles as
active movable shading fins. Unlike typical tuned mass damper (TMD) systems, the
DMD are placed along the height of the building (for serving as shading fins) rather
72

than concentrated in a few locations. The DMD system shows improved vibration
control over a single TMD system when the parameters — damper mass, stiffness
and damping terms of each floor— are optimized. Several sub-optimizations are
also performed for the DMD systems for constant, gradually increasing and stepping
damper masses for simple design and constructability; the results are on par with the
performance of the single TMD system. The external shading fins are placed in the
east and west sides of the building and are shown, using the building energy
simulation program eQuest, to be more energy efficient compared to the same
building without fins. Furthermore, different lengths and orientations of the shading
fins are simulated to demonstrate possible benefits of actively movable shading fins
that can adjust positions to control direct sunlight and minimize energy cost (mass
dampers requires the shading fins to be movable). Though there are some
restrictions, active shading fins are simulated in eQuest that can turn 45
o
, 90
o
and
135
o
from North in different time periods to control heat gain from sunlight. This
active shading fin system is shown to be more energy efficient than the static
counterpart.
The current result focuses the integration of an active EC system (active
shading fins) and a passive SC system (tuned distributed mass dampers). However,
the active EC system with its movable equipment can also be used to help install an
active SC system and the following sections outline the future work of an
active/semi-active SC system for the current SFMD system. Moreover, another part
73

of smart structure — structural health monitoring — is also suggested to integrate
with the SFMD system.

8.1 Active + Semi-active Mass Damper System

Mass dampers that are equipped with actuators to control the damper
movements are called active mass drivers (AMD). Technically, the active SFMD
system is also an AMD system because of the actuators that adjust the movements of
the SFMDs for both shading and vibration reduction purposes. However, a typically
AMD requires powerful actuators that move the massive dampers effectively. In the
case of the SFMD system, the primary (or most frequent) purpose of the actuators is
to control the fins for shading. The adjustments for shading are not as demanding as
structural control since the sun moves gradually. To compensate for the loss of
power in the small actuators, semi-active dampers could also be connected to the
SFMDs (Figure 8.1). This mass damper system that uses small actuators and semi-
active dampers will be called active + semi-active mass damper (A+SAMD) system.
This system should be the next phase in further studies of this thesis.

Figure 8.1: Active + semi-active mass damper.

74

8.2 Active SFMD for Structural Health Monitoring (SHM)

Another function of the active SFMD system is for Structural Health
Monitoring (SHM). The small actuators can locally excite the structure by vibrating
the fin mass dampers. A local excitation may be beneficial to locate damage in the
structure using vibration based SHM. Furthermore, combinations of the local
excitations are possible by exciting several fin mass dampers simultaneously. This
can help excite certain modes of the structure for SHM purposes.
To test the idea, an n-story model with the distributed mass dampers and the
smaller actuators could be used to detect damage. Damage could be caused by
decreasing the stiffness in various stories. Detection could be based on the global
excitation by multiple actuator and local excitation caused by the small actuators.
For large damage (large stiffness loss), a global vibration SHM approach would be
sufficient. However, for smaller damage, the global vibration approach may not be
enough and a complimenting local vibration approach may be desired. Furthermore,
as the number of stories increases (taller structure), it could be more difficult to
localize damage to a particular story using the global vibration SHM approach alone;
the local vibration SHM approach may be helpful.
75

References

Abe M., Fujino Y., 1994. Dynamic characterization of multiple tuned mass dampers
and some design formulas. Earthq Engng Struct Dynam, 23, 813–35.

Alward, R., and Shapiro, A., 1981. Low-Cost Passive Solar Greenhouses, National
Center for Appropriate Technology, Butte, MT. 173.

Bradshaw, V., 2006. The Building Environment: Active and Passive Control
Systems, John Wiley & Sons, Hoboken, N.J.

Chen, G., and Wu, J. 2001. Optimal Placement of Multiple Tune Mass Dampers for
Seismic Structures. J. Structural Engineering, 127, 1054-62.

Chen, G., and Wu, J., 2003. Experimental study on multiple tuned mass dampers to
reduce seismic responses of a three-storey building structure. Earthquake Engng
Struct. Dyn, 32, 793-810.

Dennis, J. E. Jr. and Torczon, V., 1991. Direct search methods on parallel machines.
SIAM J. Optim., 1, 448-474.

Halbe, R, 2004. caltrans.jpg Retrieved April 1, 2007 from
http://architettura.supereva.com/architetture /20050325/index.htm

Higashino, M., and Aizawa, S., 1993. Application of active mass damper system in
actual buildings. G.W. Housner and S.F. Masri (eds), Proc. Int. Workshop on Struct.
Control, 194-205, Los Angeles, CA.

Hooke, R. and Jeeves, T. A., 1961. “Direct search" solution of numerical and
statistical problems. J. Assoc. Comput., 212-229.

Igusa, T., and Xu, K., 1994. Vibration control using multiple tined mass dampers. J.
Sound and Vibration, 4, 491-503.

Kareem, A., and Kline, S., 1995. Performance of multiple mass dampers under
random loading. J Struct Engng, 121, 348–61.

Marin, 2005. hall_of_records.jpg. Retrieved April 6, 2006 from http://you-are-
here.com/skyscraper/ 1962_hall_of_records.html

Olgyay and Olgyay, 1957. Solar Control and Shading Devices, Princeton University
Press, Princeton, NJ. 73-74.

76

Ramallo, J. C., Johnson, E. A., and Spencer, B. F., Jr., 2002. “Smart” Base Isolation
Systems. Journal of Engineering Mechanics, 128(10), 1088-1099.

Sendagaya INTES building, 2003. Sendagaya_intes.gif. Retrieved November 5,
2005 from http://www.japan-photo.de/e-mo-j17.htm

Soong, T., and Grigoriu, M., 1993. Random Vibration of Mechanical and Structural
System, Prentice Hall, Englewood Cliffs, NJ.

Soong, T. T., and Spencer, B. F. Jr., 2002. Supplemental energy dissipation: state-of-
the-art and state-of-the practice. Engineering Structures, 12: 243-259.

Takenaka Corporation, 2001. tatemono.gif. Retrieved July 18, 2006 from
http://www.takenaka.co.jp/ takenaka_e/ quake_e/menshin/menshin.htm

Torczon, V., 1989. Multi-Directional Search: A Direct Search Algorithm for
Parallel Machines. Thesis (Ph.D), Department of Mathematical Sciences, Rice
University, Houston, TX.

Torczon, V., 1997. On the convergence of pattern search algorithms. SIAM J.
Optim., 7, 1–25.

Nagase, T. and Hisatoku, T., 1992. Tuned pendulum mass damper installed in
Crystal Tower. The Structural Design of Tall Building, 1, 35 −56.

Crystal Tower, 2003. Crystal01.jpg and Crystal02.jpg. Retrieved November 5, 2005
from http://www.takenaka.co.jp/ takenaka_e/t-file_e/d_synthesis/crystal/

Yamaguchi H., and Harnpornchai, N., 1993. Fundamental characteristics of multiple
tuned mass dampers for suppressing harmonically forced oscillations. Earthq Engng
Struct Dynam, 22, 51–62.

Abstract (if available)

Abstract Synergy of integrating structural and environmental control systems is explored with a proposed Shading Fin Mass Damper (SFMD) system. External shading fins control sunlight entering the building, adjusting the internal temperature and lighting conditions. In the SFMD system, the fins also serve as tuned mass dampers (TMDs) that move and dissipate energy during strong structural motion. The additional mobility allows the fins to change positions for greater control to minimize energy consumption. Using eQuest, a building energy simulation program, the movable shading fins substantially decrease energy consumption. Since the shading fins are placed along the height of the building, the dampers are distributed rather than concentrated in a few locations as in typical TMD systems. The distributed mass damper (DMD) system is formulated, simulated and optimized to reduce structural vibration significantly. Sub-optimal DMD systems are also studied for simpler design and constructability without compromising structural benefits.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Smart buildings: synergy in structural control, structural health monitoring and environmental systems

PDF

Developing environmental controls using a data-driven approach for enhancing environmental comfort and energy performance

PDF

Performative shading design: parametric based measurement of shading system configuration effectiveness and trends

PDF

Harvesting energy with solar panels and adaptive shading for building skins: A case study of an office building in Saudi Arabia

PDF

Economic model predictive control for building energy systems

PDF

Simulation-based electric lighting control algorithm: integrating daylight simulation with daylight-linked lighting control

PDF

Lateral design with mass timber: examination of structural wood in high-rise timber construction

PDF

Computationally efficient design of optimal strategies for passive and semiactive damping devices in smart structures

PDF

Design proposal and analysis of curved brace system to reduce drift in moment frame structures

PDF

Kinetic facades as environmental control systems: using kinetic facades to increase energy efficiency and building performance in office buildings

PDF

The intelligent control strategy of kinetic façades for daylight and energy performance: evaluating the daylight effect of adaptive systems based on parametric workflow

PDF

Economizer performance and verification: effect of human behavior on economizer efficacy and thermal comfort in southern California

PDF

Energy performance of different building forms: HEED simulations of equivalent massing models in diverse building surface aspect ratios and locations in the US

PDF

Energy savings by using dynamic environmental controls in the cavity of double skin facades

PDF

Adaptive façade controls: A methodology based on occupant visual comfort preferences and cluster analysis

PDF

Exploring participatory sensing and the Internet of things to evaluate temperature setpoint policy and potential of overheating/overcooling of spaces on the USC campus

PDF

The effect of microhydration on ionization energy and proton transfer in nucleobases: analysis and method development

Asset Metadata

Creator Fu, Tat Shing (author)

Core Title Integration of mass dampers and external shading fins: exploring synergy in structural and environmental control systems

School School of Architecture

Degree Master of Building Science

Degree Program Building Science

Degree Conferral Date 2007-12

Publication Date 12/03/2007

Defense Date 04/01/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag environmental control,mass damper,OAI-PMH Harvest,shading fin,structural control,synergies

Language English

Advisor Schierle, G. Goetz (committee chair), Schiler, Marc E. (committee chair), Johnson, Erik A. (committee member), Spiegelhalter, Thomas (committee member)

Creator Email tsf@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m953

Unique identifier UC1172804

Identifier etd-Fu-20071203 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-594487 (legacy record id),usctheses-m953 (legacy record id)

Legacy Identifier etd-Fu-20071203.pdf

Dmrecord 594487

Document Type Thesis

Rights Fu, Tat S.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu