Download Simulation of intermittent transient cooling load characteristic in an academic

2011 International Conference on Environment Science and Engineering
IPCBEE vol.8 (2011) © (2011) IACSIT Press, Singapore
Simulation of intermittent transient cooling load characteristic in an academic
building with centralized HVAC system
Petrus Tri Bhaskoro
Syed Ihtsham Ul Haq Gilani
Mechanical Engineering Department
University Teknologi PETRONAS
Bandar Seri Iskandar, Malaysia
[email protected]
Mechanical Engineering Department
University Teknologi PETRONAS
Bandar Seri Iskandar, Malaysia
[email protected]
Mohd Shiraz Bin Aris
Mechanical Engineering Department
University Teknologi PETRONAS
Bandar Seri Iskandar, Malaysia
[email protected]
different cooling load characteristics. In office or residential
buildings, the room is usually occupied during cooling period
even though the number of occupants may be changing over
the period [4,5]. In a university, there are occupied and
unoccupied hours during cooling period [6]. Due to this, the
cooling type is likely to be continuous load for office or
residential buildings and intermittent load for a university.
The types of rooms are also found to be more various in a
university than that in office or residential building.
Simulation and experimental studies showed that external
window glazing could have significant effect on indoor
thermal comfort and energy consumption. Due to the glazing
constructions and properties, more shortwave solar radiation
will be transmitted into the zone while long wave solar
radiation from internal heat sources will be trapped inside the
zone. On the other hand, it will reduce heat gain from artificial
lighting during daytime and release the heat buildup during
daytime easily during nighttime [7,8,9,]. These results
indicate that there is a trade-off between thermal comfort,
lighting need and energy consumption.
Simulations and experimental studies to study cooling
load characteristic have been done in the past by taking into
account internal (occupant, lighting, etc) and external (wall
heat gain, window heat gain and solar radiation heat gain)
cooling load. The building thermal model was first
constructed based on physical and operation characteristic of
the building [4,9,10,11]. The model was then validated by
comparing simulations results with experimental results
before it is used for further analysis. From the simulations
results, it was found that for office and residential buildings,
heat gained from building envelope and lighting are the major
cooling load [4,9]. The previous studies of cooling load
characteristics, however, were mostly conducted for office
and residential building.
Therefore, the objective of this work is to do simulation
study of intermittent transient cooling load characteristic of an
academic building with large glazing area using transient
system simulation software TRNSYS. A building thermal and
Abstract— This paper aims to estimate intermittent transient
cooling load characteristics in an academic building with large
glazing area using TRNSYS. Simulation and experimental
studies have been done in the past to estimate cooling load
characteristic and to find strategies to reduce the energy used
by the centralized HVAC system in various ways. However,
these studies were mostly conducted for a building with
simultaneous occupancy pattern. Since occupancy pattern in a
university is intermittent, the cooling load characteristic would
be different with that with simultaneous occupancy pattern.
The building thermal and simulation model was first built and
validated by comparing the results with experimental results.
The validated model was then used to estimate the cooling load
characteristic on a year basis. In addition, the effect of the
occupancy pattern were also analyzed and discussed. The
results showed that the ratio of cooling loads at minimum and
maximum were 0.52, 0.54, and 0.68 for the workshop,
classroom and office due to the occupancy pattern. The results
also implied that the HVAC system should be equipped with
additional equipment to track the occupancy and make it work
based on the occupancy pattern. This would reduce the energy
wasted during unoccupied period.
Index Terms— Keywords: Building simulation, intermittent
transient cooling load characteristic, centralized HVAC system,
academic glazed building
I. INTRODUCTION
Recently, efforts to reduce energy usage are doubled due
to global warming and energy source shortage. With the fact
that HVAC system is the major energy consumption in office
buildings in Malaysia (up to 64% of the total energy
consumed) [1] and a trend of higher cooling energy usage due
to climate change [2], strategies to reduce the cooling load is
highly needed .
Davis III and Nutter found that occupancy pattern in a
university differ from that in many office and residential
buildings [3]. Since cooling load in those building is mostly
driven by the occupancy pattern, the difference will result in
291
(Monday-Friday), working hours (08.00-17.00) and Malaysia
national holidays.
simulation model was first constructed and validated by
comparing the results with experimental results. The validated
model was then used to simulate the hourly intermittent
transient cooling load characteristic for a year. In addition, the
effect of usage factor and the occupancy pattern on the energy
consumption and the HVAC system performance were
analyzed and discussed.
TABLE 1 BLOCK 16 BUILDING MATERIAL SPECIFICATION [13]
Building
Construction
External
Wall
II. BUILDING THERMAL AND SIMULATION MODEL
A transient building thermal and simulation model for the
academic blok was developed using energy building
simulation software TRNSYS. The building thermal and
simulation model was developed based on the following:
• Building construction detail and occupancy schedules
• List of indoor lights, equipments and machines
• Centralized HVAC system in the building
• Cooling load types in the building
Partition
Wall
Flooring
Window
Roofing
qc,s
qr,s
hconv,s,
σ
εs,
Ta,s
Tfsky
qcomb
RH
Steel (5mm-54 kJ/hmK), Air gap (0.047 hm2K/kJ),
Steel (11mm-54 kJ/hmK)
Plasterboard (25mm-0.576 kJ/hmK), Air Gap
(92mm-0.047 hm2K/kJ), Plasterboard (25mm-0.576
kJ/hmK)
Concrete slab (100mm-4.07 kJ/hmK), Common
concrete (550mm-7.56 kJ/hmK)
Optiwhite glass (8mm-3.24 kJ/hmK)
Alumunium (1mm-846 kJ/hmK), Rockwool
(25mm-0.162 kJ/hmK), Alumunium foil (1-846
kJ/hmK), Common concrete (10mm-7.56 kJ/hmK)
Activity schedule for each workshops in block 16 are
based on practical schedule of each subject. Numbers of
equipment used and usage factors of the equipments in each
type of rooms were based on observation data and discussion
with the technician/staff. Activities in the workshop with no
specific schedule were not considered in this simulation.
Nomenclature
Ts
Tstar
As
Requiv
Details ( thickness-thermal conductivity)
Surface temperature, oC
Artificial temperature node, oC
Surface area, m2
Equivalent resistant between the walls with a
node, h m2 K kJ-1
Convection heat flux an the surface of the wall,
kJ h-1
Long wave radiation heat flux on the surface of
the wall, kJ h-1
Convective heat transfer coefficient on the
surface of the wall, kJ h-1 m-2 K-1
Stephan-Boltzmann constant
Long-wave emissivity of the surface
Ambient temperature, oC
Fictive sky temperature, oC
combined convective and radiation heat
transfer, kJ h-1
Relative humidity, %
TABLE 2 U-VALUES OF BLOCK 16 WALLS
Type
Floor
Roof
Partition Wall
Window
External Wall
U-values (conduction)
(W/m2K)
6.74
6.425
7.474
31.25
21.143
U-value (Overall)
(W/m2K)
6.433
4.928
5.522
20.448
10.58
TABLE 3 STUDENTS ACADEMIC SCHEDULE 2009
Month
Week
I
II
III
IV
V
Month
Week
I
II
III
IV
V
Superscripts
i
inside
o
outside
A. Building thermal details and occupancy schedules
An academic block located at Universiti Teknologi
PETRONAS (UTP) was chosen for the study. The blok is
facing east with large window glazing area (nearly 100% of
glazing area) and on 32m above sea level. It has three levels
with 4m height on each floor. Floor lay out for the block was
detailed on construction drawing [12].
As an academic block, it has offices, classrooms and
workshop to support practical/research works by the students
and staffs. Material used for walls on the workshop,
conductivity and U-values of each material were described in
Table 1 and 2. Student academic schedule was described in
Table 3 while staff schedule was based on UTP working days
B
E
S
Jan
Feb
Mar
Apr
May
Jun
B
B
S
S
-
S
S
S
S
-
S
S
B
S
S
S
S
S
S
S
E
E
E
E
-
B
B
B
B
B
Jul
Aug
Sep
Oct
Nov
Des
S
S
S
S
S
E
E
E
E
-
B
B
B
B
B
B
S
S
B
S
S
S
S
S
S
S
B
S
: Mid semester or semester break
: Examination
: Study week
B. Centralized HVAC system in the building
Centralized AC system with absorption chiller was used to
cover the cooling load from all the academic blocks. Each
block has 3 levels and each level have two wings of group
zones (left and right wing). The buildings are fully sealed and
fresh air is supplied by the air handling unit (AHU) on each
floor. There are two air handling unit (AHU) on each level.
AHU 1 is used to provide air supply to the right wing while
AHU 2 is for the left wing. A variable fan speed and a
292
regulating valve were used in the AHU to give variable supply
air and chilled water flow rate.
The HVAC system use variable air volume (VAV) system
to control supply air flow rate enter the room according to the
cooling load. A temperature sensor is placed in each room to
monitor room temperature [14]. The AC system was operated
from 7 a.m till 7 p.m to keep indoor temperature set point
(24oC). Indoor relative humidity was kept floating depend on
the indoor temperature.
For calculation simplification, influence of indoor
furnishing to indoor RH can be neglected since indoor
furnishing will affect on indoor RH level at less than 2% RH,
especially during daytime operating mode condition [2].
Occupant activity level in classroom and office was set at
level 4 with sensible and latent heat values 269kJ/hr. Level 5
of activity level was applied for workshop where practical
work existed. The sensible heat value is 332.33kJ/hr and the
latent heat value is 342.88kJ/hr. These values are based on
ISO 7730 table.
C. Cooling Load in the Building
Heat gain from equipments (QEQUIP), heat gain from
lighting (QLIGHT), heat gain from electronic equipment and
heat gain from building envelope due to solar radiation
(QENV) would contribute to sensible cooling load (QSEN).
Heat gain from occupants (QPERSON), heat gain from
ventilation (QVENT) and heat gain from infiltration (QINF)
would contribute to both latent (QLAT) and sensible cooling
load. These heat gains were considered in the simulation.
The heat balance method is used in TRNSYS as a base for
all calculations. For conductive heat gain at the surface on
each wall, TRNSYS use transfer function method (TFM) as a
simplification of the arduous heat balance method [15]. Heat
gain through radiation and convection within the zone were
calculated using the star network given by Seem:
q comb,s,i = q c,s,i +q r,s,i =
1
Requiv ,i As ,i
(Ts ,i − Tstar )
IV. EXPERIMENTAL DATA MEASUREMENTS
Indoor and outdoor experimental measurements were
done for 2 weeks for validation purposes only. Indoor
temperature, indoor relative humidity, supply air flow rate and
supply air temperature were measured to get indoor
environmental and supply air conditions. Supply air flow rate
and temperature measurements are performed under the
maintenance department supervision. Scheduled calibrations
were performed by the department to maintain the accuracy of
the equipments.
Ambient temperature, ambient relative humidity, global
solar radiation, wind speed and wind directions measurements
were done to get outdoor environmental conditions. The first
three outdoor parameters measurements were done every 5
minutes by Firmanda [18] while wind speed and wind
direction were not measured on this locations due to technical
difficulties. These measurements data were collected from
nearby weather station at Ipoh.
(1)
Heat gain through radiation and convection for external
surface were calculated by:
q comb,s,o = q c,s,o +q r,s,o
(2)
q c,s,o = h conv,s,o (Ta,s − Ts,o )
(3)
(
4
q r,s,o = σε s,o Ts,4o − Tfsky
)
V. COMPARISON OF EXPERIMENTAL AND SIMULATION
RESULTS
The comparison of indoor temperature between
experimental measurement and simulation model is shown in
Fig. 1 and 2. 81.25% of the simulated temperature values were
found to be within +1oC with the measured values. The
difference occurs may be due to climatic difference between
building area and outdoor experiment area.
(4)
Convection heat transfer coefficient for inside and outside
walls were set 11 kJ/hm2K and 64 kJ/hm2K as recommended
by the software [15]. Minimum ventilation rate required in
each room is calculated based on ASHRAE standard [16].
Degree levels of activities were based on the types of
activities in each room and it was inputted to get the portion of
sensible and latent heat from ISO 7730 table. Heat gains from
the equipments were calculated based on rated power, usage
factor, load factor and its efficiency. Convective and
irradiative fraction for these heat gains were 0.7 and 0.3 while
for artificial lights, the values were 0.6 and 0.4 [17].
III. ASSUMPTIONS AND STANDARDS
Building and occupant ventilation components were
considered based on ASHRAE standard [16]. Building
ventilation component is at rate 0.3 L/sm2 and occupant
ventilation rates per person are 2.5 L/s per person. Infiltration
air exchange due to door openings and leakage air was
assumed to be 0.05/h during AC operation and 0.15/h during
non-AC operation.
Figure 1. Comparison of indoor temperature between measured values and
simulation model during a holiday.
293
VI. INTERMITTENT TRANSIENT COOLING LOAD
CHARACTERISTIC
Intermittent transient cooling load for a year was
calculated by the software using hourly climatic weather data
collected from Ipoh weather station. The characteristics for
typical workshop, classroom and office in the building during
peak and off-peak period were shown in Fig. 5 to Fig. 10
Figure 2. Comparison of indoor temperature between measured values and
simulation model during a working day.
Indoor relative humidity comparison between
experimental measurement and simulation model is shown in
Fig. 3 and 4. 72.5% of the simulated relative humidity values
were found to be within +5% compare to measured values. It
may be due to influence of indoor temperature which will
amplify the difference.
Figure 5. Intermittent transient cooling load characteristic of workshop in the
academic building during peak period.
In the workshop, during peak period, the sensible and
latent heat rise due to occupancy and practical activity. It can
be seen that peak cooling load during peak period occur
during occupied hours. However, the occupant exists only for
several hours and become empty after that. The peak cooling
load during peak period and off-peak period were counted up
to 73 MJ and 39 MJ.
In the classroom, sensible and latent heat rise due to
occupancy and more ventilation air needed. Similar to
workshop, the peak cooling load during peak period occur
during occupied hours. However, occupancy in the classroom
was more frequent compare to that in the workshop. The peak
cooling load during peak period and off-peak period were
counted up to 51 MJ and 25 MJ.
Figure 3. Comparison of indoor relative humidity between measured values
and simulation model during a holiday.
Figure 6. Intermittent transient cooling load characteristic of workshop in the
academic building during off-peak period.
Figure 4. Comparison of indoor relative humidity between measured values
and simulation model during a working day.
The simulation model was then used to calculate
intermittent transient cooling load characteristic of the
building for a year. The result was then used to investigate the
effect of usage factor and the HVAC system performance.
294
Figure 7. Interm
mittent transient cooling load charaacteristic of classrroom in
the academic buiilding during peakk period.
Fiigure 10. Intermitttent cooling loadd characteristic off office in the acaddemic
building durinng off-peak periodd.
It can be seeen that usage factor plays important rolees on
daaily total coooling load. U
Usage factorr is the ratiio of
unnoccupied houurs to total H
HVAC operatting hours. Higher
H
ussage factor wiill result in higgher total coo
oling load andd vice
veersa.
VII. CO
ONCLUSION
Simulationss to determiine intermitteent cooling load
chharacteristic fo
or typical workkshop, classroo
om and office in an
accademic buildiing have been presented andd discussed.
The buildinng has been m
modeled and has
h been validdated
(thhrough compparison betweeen experimen
nt and simullation
results). The result showed thhat the modell behave closeely to
t
v
values
thhe real buildingg. 81.25% of tthe simulated temperature
weere found to be
b within +1oC with the measured values. The
difference occuurs may be duee to climatic difference
d
betw
ween
buuilding area and
a outdoor experiment
e
arrea. 72.5% of
o the
sim
mulated relativve humidity values were fouund to be withhin +5%
coompare to meaasured values. High differeence for the inndoor
relative humidiity was due tto fact that reelative humidiity is
innfluenced by thhe temperaturee so that the diifference on inndoor
tem
mperature wo
ould amplify tthe difference of indoor rellative
huumidity.
The coolingg load was fouund to be flucttuated accordiing to
thhe occupancy pattern
p
(occuppant’s driven cooling load). The
results showed that
t the ratio oof cooling load
ds at minimum
m and
m
maximum
were 0.52, 0.54,, and 0.68 for
f the worksshop,
claassroom and office. Due too the occupanncy pattern, seeveral
prroblems occur on the currennt HVAC systeem:
• During unooccupied hourrs, the HVAC
C system couldd not
be turned off
o due to thee centralized ventilation sy
ystem
and fact thaat the system serves more th
han one room
m with
specific occupancy patteern for each ro
oom.
m
suppply air flow raate according to
t the
• There is a minimum
ducting design and sizinng. The cooliing energy wiill be
wasted whhen the supplyy air flow ratte needed is lower
l
than the miinimum supply air flow ratee. It was the reeason
why indooor temperaturee sometimes falls
f
below thhe set
point.
• Occupancy
y pattern forr each room cannot be fixed
because it is specific annd can be chaanged. Due too the
occupancy pattern and ooccupant’s drriven cooling load,
Figure 8. Interm
mittent transient cooling load charaacteristic of classrroom in
thhe academic buildding during off-peeak period.
mittent cooling loaad characteristic of
o office in the accademic
Figure 9. Interm
building during
d
peak period.
In the offfice room, sennsible and lattent heat rise due to
office activities from the occupant. The cooling load pattern
was not as fluuctuates as in classroom beecause of conttinuous
occupied houurs (working hours). The peak coolinng load
during peak and
a off-peak period were accounted up too 32 MJ
and 24 MJ. The
T peak cooliing load occuurred around 1p.m
1
to
3p.m for bothh periods.
295
additional equipment is needed by the current system to
track the occupant and control the HVAC system based
on the occupancy.
Strategies to solve the problems and handle the
intermittent load by taking into account usage factor and
occupancy pattern for each type of room will be discussed and
presented in the near future.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Energy efficiency. Pertubuhan arkitek Malaysia. CPD seminar. 2004
K. K. W. Wan, D. H. W. Li, D. Liu, and J. C. Lam (2010). Future
trends of building heating and cooling load and energy consumption
on different climates. Building and Environment. Volume 46 (2011),
PP 223-234. Available: http://www.elsevier.com/locate/buildenv
James A. Davis III, and D. W. Nutter (2010). Occupancy diversity
factors for common university building types. Energy and Buildings.
Volume
42
(2010),
PP
1543–1551.
Available:
http://www.elsevier.com/locate/enbuild
Z. Li, W. Chen, S. Deng, and Z. Lin (2005). The characteristic of space
cooling load and indoor humidity control for residences in the
subtropics. Building and environment. Volume 41 (2006), PP
1137-1147. Available: http://www.elsevier.com/locate/enbuild
K.W. Mui, and L.T. Wong (2006). Cooling load calculations in
subtropical climate. Building and Environment. Volume 42 (2007),
PP 2498–2504. Available: http://www.sciencedirect.com
P. T. Bhaskoro, “Transient cooling load simulation of a mechanical
workshop at UTP,” presented at the ICPER , Kuala Lumpur, Malaysia,
June 15-17, 2010, Paper EF 12
A. Stegou-Sagia, K. Antonopoulos, C. Angelopoulou, and G.
Kotsiovelos (2007). The impact of glazing on energy consumption
and comfort. Energy Conversion and Management. Volume 48 (2007),
PP 2844–2852. Available: http://www.sciencedirect.com
N. A. M. Al-Tamimi, S. F. S. Fadzil, and A. Abdullah, “The effect of
orientation and glazed area to the indoor air temperature in
unventilated building in hot-humid climate,” 3rd International
Conference on Built Environtment in Developing Countries, Kuala
Lumpur, 2009
J. C. Lam, and D. H. W. Li (1998). An analysis of day lighting and
solar heat for cooling dominated office buildings. Solar Energy.
Volume 65 (1999), No. 4, PP 251–262
M. Bessoudo, A. Tzempelikos, A.K. Athienitis, and R. Zmeureanu
(2010). Indoor thermal environmental conditions near glazed facades
with shading devices-Part I: experiments and building thermal model.
Building and Environment. Volume 45 (2010), PP 2506 – 2516.
Available: http://www.elsevier.com/locate/buildenv
Z. Lin, and S. Deng (2004). A study on the characteristics of nighttime
bedroom cooling load in tropics and subtropics. Building and
Environment. Volume 39 (2004), PP 1101 – 1114. Available:
http://www.sciencedirect.com
Construction drawing, Universiti Teknologi PETRONAS
Material specification, University Teknologi PETRONAS
Air conditioning and ventilation drawing, University Teknologi
PETRONAS
TRNSYS 16 manual
Ventilation for acceptable indoor air quality, ASHRAE, Inc., 2005
J. D. Spitler, D. E. Fisher, and C. O. Pederson, “The radiant time series
cooling load calculation procedure,” ASHRAE trans. Vol. 103, part 2,
1997
D. Firmanda,”Energy and techno-analysis of solar power system for
residential house lighting,” Symposium paper, unpublished
296