International Journal of Air-Conditioning and Refrigeration

Vol. 26, No. 4 (2018) 1850030 (19 pages)

© World Scienti¯c Publishing Company
DOI: 10.1142/S201013251850030X

Thermal Zoning Based on Design Cooling Loads:

Methodology and Simulation Case Study
for a DOAS with Local Recirculating Units

Mridul Sarkar
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.

Integrated Environmental Solutions India Pvt. Ltd

Pune-411021, India
mridul.sarkar@iesve.com; mridul.rns@gmail.com

Received 18 May 2018

Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

Accepted 19 July 2018

Published 30 August 2018

A framework for creating thermal zones in a building for e®ective and e±cient heating ventilation
and air conditioning (HVAC) system design is introduced here. This method is based on simple
\sort and eliminate" schemes and requires design cooling loads of conditioned spaces obtained
from load calculation tools as primary input. The developed methodology is applied for creating
thermal zones, determining corresponding supply conditions and ascertaining sizing of a dedicated
outdoor air system (DOAS) with local recirculating units. A simulation study on a prototype-
building model shows that a DOAS coupled with zoned recirculating systems that serve distinct
thermal zones in a building (zoned model) perform comparatively better in controlling both space
temperatures and humidity without signi¯cantly compromising HVAC energy and chiller loads
than un-zoned HVAC systems serving the whole building as a single thermal block (un-zoned
model). The consistency in the performance of zoned HVAC systems is veri¯ed by applying
three di®erent simulation weather ¯les for New Delhi. Better performance along with logical
and computational simplicity makes this design procedure a good alternative to traditional
methodologies.

Keywords: Thermal zone; DOAS; weather; cooling loads; energy.

Nomenclature h : Speci¯c enthalpy (kJ/kg)

hfg : Speci¯c latent heat of vaporization (kJ/kg)
Symbols l0 : Speci¯c latent heat of vaporization at T0 ¼ 273:15 K
ADP : Apparatus dew point temperature (
C) (2501 kJ/kg)
BF : Bypass factor LL : Latent load (kW)
cpa : Speci¯c heat capacity of dry air ( 1:006 kJ/kg-K) m : Slope of RSHF (or SHF) line
cpg : Speci¯c heat capacity of dry water vapor P : Saturation vapor pressure (kPa)
( 1:86 kJ/kg-K) Pt : Ambient pressure (kPa)
cpm : Speci¯c heat capacity of moist air ( 1:023 kJ/kg-K) q : Heat load (kW)

Cplv : Di®erence of speci¯c heats of water in liquid and R : Gas constant of water vapor (0.4618 kJ/kg-K)
vapor states ( 2:326 kJ/kg-K) SL : Sensible Load (kW)
FA : Fresh air °ow rate (m 3 /s) t : Temperature (
C)

1850030-1
 M. Sarkar
T : Absolute Temperature (K)
V : Volume of room (m 3 Þ
V_ : Air °ow rate (m 3 /s)
Greek symbols

: Ratio of molecular mass of water vapor and dry air (0.622)
 : Density of air (1.2 kg/m 3 )
! : Humidity ratio (kg moisture/kg DA)
Subscripts
adp-i : Supply ADP condition for space i
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
adp-TZ : Supply ADP condition for space group/zone
chk-i : Resultant parameter for space i at zone supply
conditions
d : Dew point condition
design : Design supply condition
d-i : Supply dew point condition for space i
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
d-TZ : Supply dew point condition for zone/space group
GSHF : Grand sensible heat factor line
outlet : Condenser outlet of WAHP
reheat : Supply air reheat
rev-i : Revised parameter for space i
s-i : Supply condition required for space i
s-TZ : Supply condition required for thermal zone/space group
set-i : Design set point of space i
set-TZ : Set point of thermal zone/space group
1. Introduction
It is estimated that by 2050 the world population will
increase by more than 30% of the present value and
reach 9.7 billion,1 which is a major concern because this
will create a huge burden on the available energy
resources. As per studies,2 30–40% of world's energy
demand is from the present building stock, which is
responsible for 25–35% of global CO2 emissions and
projected to increase in the coming future. Buildings
are multi-zoned entities that have varied and complex
usage, occupancy and application patterns through-
out. That is why a single de¯nition or title cannot fully
de¯ne a building. As per CBECS database, approxi-
mately 40% of building's energy consumption is due to
heating ventilation and air conditioning (HVAC) sys-
tems.3 Therefore, it is imperative that design and op-
eration of building HVAC systems should be e±cient
to reduce energy wastage and provide comfortable
dwelling environments for occupants.
Modeling multi-zoned buildings is not a new con-
cept. Traditionally, building simulation tools are
implemented for multi-zoned thermal models.4 Since,
room-partitioning layout for a °oor plan is not en-
tirely de¯ned during the schematic design phase,
1850030-2
modeling analyses are more focused on geometrical
aspects. As per ASHRAE std. 90.1, °oor plates with
unknown room partitioning layouts can be divided
into core and perimeter zones for creating early-stage
models.5 In recent years, researchers have introduced
many computational methodologies that can be
incorporated with building simulation tools for cre-
ating multi-zoned building models from basic archi-
tectural drawings6 or complex un-zoned models.7
These methodologies are extended for preliminary
analyses for urban scale models, too.8 Researchers
have reported signi¯cant variations in simulation
energy and loads with actual architectural plans and
ASHRAE 90.1 compliant zoning scheme.9
During detailed design stage, spatial complexity
in building layouts may impose some complications
for comprehensive modeling and simulation. For
simplifying thermal simulations and HVAC con-
trols, multiple spaces are combined into a single
zone based on space application and load pro¯les.10
Design guidelines also suggest combining spaces into
a single thermal zone based on thermal settings11
and exterior exposures.5 Sources state that the
general methodology for delegating spaces into ap-
propriate thermal zones is more subjective in nature
and depends mainly on the judgment and experience
of designers.12 The e®ect of thermal zoning on the
overall energy consumption of HVAC systems is also
explored in a literature source.13
Despite being comprehensive, the methodologies
for thermal zoning neither address the issues related
to HVAC design nor suggest any systematic proce-
dure to link thermal zoning and HVAC system
design right from the initial stages. As per conven-
tional methodology, the design air°ow rate depends
on space sensible load and temperature di®erence
between supply and room set points,14,15 whereas
the space relative humidity (RH) depends on supply
dew point temperature (DPT) and space latent
load. A distinction between these sensible and latent
loads is very important for HVAC system design
and equipment selection.14 However, most of the
conventional software tools do not explicitly explore
or de¯ne the connection between thermal zoning
and required system supply conditions, even though
thermal zoning directly a®ects the performance
of HVAC systems in maintaining indoor thermal
conditions.
The present work aims towards bridging this gap
by proposing a method to link thermal zoning and
HVAC system design in buildings based on spatial

cooling loads and exploring its implications on
resulting zone conditions. The proposed zoning
methodology is applied for designing a dedicated
outdoor air system (DOAS) with local recirculating
systems. The latter part of this paper will be dedi-
cated to a simulation case study for a prototype-
building model located in New Delhi, India. The
primary objective of this study is to perform a
comparative assessment of HVAC systems with and
without thermal zoning considerations with three
di®erent simulation weather ¯les of New Delhi.
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
2. Air Conditioning of Thermal Zones
Primary goal of any air conditioning system is to
maintain indoor conditions that are comfortable and
thermally acceptable for occupants. For this, it is
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
important to identify the key thermodynamic
factors a®ecting the conditions inside a space.
Available literature source presents the general
methodology for conditioning a single space based
on its sensible and latent loads.15 However, in
practical applications, a system is utilized for con-
ditioning a whole zone comprising various spaces
that may have di®erent applications.
A thermal zone is a single space or a group of
spaces that have similar thermal load character-
istics, such that a HVAC system may control and
maintain approximate levels of prescribed condi-
tions inside. In a building, there are di®erent types
of spaces that have varied load variation and oper-
ational patterns. Identifying these spaces and
grouping them into clusters based on similar cooling
load characteristics will allow meeting space-condi-
tioning requirements with limited number of ther-
mal equipment.
In the present work, two terms namely \sample
space group" and \thermal zone" are used fre-
quently. A sample space group is de¯ned as a set of
spaces that are sorted out after satisfying certain
conditions and is under assessment for further divi-
sion into multiple clusters called thermal zones.
Hence, all the equations shown in the later sections
are equally applicable for any sample space group or
thermal zone.
2.1. Methodology
In this section, the general design methodology for
conditioning a single space is extended to a thermal
1850030-3
Thermal Zoning Based on Design Cooling Loads
zone. In a space, the main thermal loads are divided
into sensible and latent parts. To o®set sensible
load, temperature of process air needs to be reduced,
whereas moisture condensation is required to o®set
latent load. The ratio of the sensible to total load of
a space is called the room sensible heat factor
(RSHF),15 mathematically written by Eq. (1):
SLi
RSHFi ¼ : ð1Þ
SLi þ LLi
An air conditioning process can be plotted on a
psychrometric chart by joining the supply and space
conditions by a straight line. This line is the RSHF
line and its slope is given by


cpm 1  RSHFi !  !s-i
mi ¼ ¼ set-i : ð2Þ
hfg RSHFi tset-i  ts-i
If either the humidity or supply temperature is
a±xed, the above equation allows calculation of
other supply parameters. The minimum limit on the
apparatus condition depends on the type and char-
acteristics of the heat exchanger (coil), refrigerant
supply temperature from main cooling plant and
thermal properties of the supplied refrigerant °uid.
Hence, it is generally assumed as such for HVAC
system design. The required supply °ow rate for
conditioning a space is determined by Eq. (3) as
SLi
V_i ¼ : ð3Þ
  cpm  ðtset-i  ts-i Þ
Equations (4a) and (4b) give the corresponding
speci¯c enthalpy of the supply air and return air for
each space, as
hset-i ¼ cpa  tset-i þ !set-i  ð2501 þ cpg  tset-i Þ ;
ð4aÞ
hs-i ¼ cpa  ts-i þ !s-i  ð2501 þ cpg  ts-i Þ : ð4bÞ
When individual spaces form a thermal zone, the
overall sensible heat factor (SHF) is given by the
weighted average of constituting space RSHF as
X
k
V_i  ðhset-i  hs-i Þ  RSHFi
i¼1
SHFTZ ¼ ; kn
X
k
V_i  ðhset-i  hs-i Þ
i¼1
ð5Þ
where `k' is the number of spaces considered in a
zone (or sample space group) and `n' is the total
 M. Sarkar

number of conditioned spaces in the building, whose deviation (SDTZ Þ17 as

corresponding systems are served by a common vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
" #
u X
cooling plant and same supplied refrigerant °uid. u k V_rev-i  ðhseti  hs-TZ Þ
u
The overall zone set point is given by volume- u
u k i¼1  ðRSHFi  SHFTZ Þ 2
weighted average of space set points as SDTZ ¼ u uk  1 
t X k
Xk
V_rev-i  ðhset-i  hs-TZ Þ
Vi  tset-i
i¼1
i¼1
tset-TZ ¼ ; ð6Þ ð12Þ
Xk
Vi
and relative standard deviation (RSD) in % is
i¼1
given by
X
k
Vi  !set-i SDTZ
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.

i¼1 RSD ¼  100% : ð13Þ

!set-TZ ¼ : ð7Þ SHFTZ
X
k
Vi The nature of load variation in a zone is expressed
i¼1
mathematically by the skewness coe±cient (SC).
The absolute humidity set point of a space is also
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

It is given by Eq. (14) as

written in terms of its RH set point as " #

 RHset-i  Pset-i Xk V_rev-i  ðhset-i  hs-TZ Þ
!set-i ¼ : ð8Þ
Pt  RHset-i  Pset-i  ðRSHFi  SHFTZ Þ 3
i¼1
The saturation vapor pressure at room set point SC ¼ : ð14Þ
Xk
temperature (Pset-i Þ is given by Eq. (9) as16 ðSDTZ Þ 3  ½V_rev-i  ðhset-i  hs-TZ Þ
"
 
Cplv  # i¼1
Tset-i  R
Pset-i ¼ PT0  In statistics, skewness is the measure of asymmetry
T0
in data distribution of a sample group. In the pres-

  
l0 þ
Cplv  T0 1 1 ent context, the concerned data is space-cooling
 exp   : load. Positive skewness implies that the overall
R T0 Tset-i
group SHF lies near the minimum RSHF value.
ð9Þ
Contrary to this, negative skewness implies that the
Here, T0 is the reference temperature of 273.15 K SHF lies in closer proximity to the maximum RSHF
and saturation vapor pressure at T0 (PT 0 Þ is value in the group. Zero skewness implies a sym-
0.611 kPa. Similar to Eq. (1), the slope of zone SHF metric distribution of cooling loads across the con-
line is given by stituent spaces.

 In this paper, maximum permissible variations of
cpm 1  SHFTZ !  !s-TZ
mTZ ¼ ¼ set-TZ : ð10Þ  0:5
C and  5% in design temperature and RH
hfg SHFTZ tset-TZ  ts-TZ set points, respectively, are assumed for the spaces
The revised supply °ow rate of a space in terms of in a valid thermal zone. For the prescribed throt-
zone supply temperature for a constituting space of tling limits, a conservative RSD limit of 5% is con-
the zone is given by sidered here.
SLi
V_rev-i ¼ ; i 2 TZ : ð11Þ
  cpm  ðtset-i  ts-TZ Þ 2.2. Sample space group creation
(`sort' scheme)
It should be noted that supplying air at the deter-
mined zone supply conditions might not o®set all of An air conditioning system serving a particular
the thermal loads inside every space in the zone. thermal zone is sized based on the peak zone cooling
However, with proper thermal zoning, these varia- load or the block load, which is sum of all the cooling
tions can be kept within acceptable limits. The net loads of constituent spaces at a speci¯c time. Since,
variation between zone SHF and RSHF of consti- the variations of thermal loads are dynamic in na-
tuting spaces is given by the weighted standard ture, occurrence of the peak zone load and peak

1850030-4

space loads may not coincide with each other.
Hence, creation of sample space groups requires a
scrutiny of constituting space application and load
characteristics.
To be sorted into a sample group, respective
spaces in a building should satisfy certain assess-
ment criteria. Technical spaces and other areas with
special air conditioning requirements (e.g., kitchens,
natatoriums, etc.) or that are required to be zoned
separately from other comfort-conditioned spaces
should be exempted from application of these cri-
teria. Four di®erent assessment levels are identi¯ed
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
here as listed below:
. All spaces whose design thermostat and RH set
points do not di®er more than prescribed maxi-
mum limits can be sorted into a group. Recom-
mendation of 4
C as the maximum di®erence in
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
thermostat set points11 and a maximum di®erence
of 10% in RH set points are acceptable for spaces
in a sample group. It should be noted that in this
work, satisfactory space RH range is shown to be
achievable entirely by thermal zoning and no
separate RH controlling intervention or controls
are assumed to be present in the primary air
conditioning unit.
. As per ASHRAE std. 90.1 Appendix G5, spaces
whose weekly operation schedules do not di®er by
more than 40 equivalent full-load hours (EFLH)
can be grouped together in a single HVAC zone.
Fig. 1. Criteria for sorting scheme to create sample space groups.
1850030-5
Thermal Zoning Based on Design Cooling Loads
This criterion can be applied as the second level of
scrutiny.
. Those spaces whose peak cooling load occurs in
the same season (months in which cooling is
dominant) can be included into a group. This is
the third level of assessment. As per ASHRAE std.
169,18 summer or cooling season extends from
April to September and October to March in
northern and southern hemisphere, respectively.
. The ¯nal criteria is to check the time of the peak
load occurrence of individual spaces. Spaces whose
peak load occurs within a maximum spread of 3 h
with respect to each other can be included to-
gether in a group. This criterion allows sorting
only spaces that do not have considerable varia-
tion in space load characteristics.
These criteria are the main features of the `sorting'
scheme. Figure 1 summarizes the sorting criteria
applied for creating sample space groups.
2.3. Sample space group to thermal zone
translation (`eliminate' scheme)
The three parameters required for creating thermal
zones from sample space groups are given by
Eqs. (12)–(14). The weighted standard deviation
provides an indication whether a single HVAC unit
can maintain thermal conditions inside the constit-
uent spaces of a group. If relative standard deviation
 M. Sarkar
exceeds the permissible limit, it implies that the
design load characteristics between spaces in the
considered group lack su±cient coherence and cor-
responding skewness coe±cient is checked. If skew-
ness > 0, the contribution from the space with
highest RSHF value is least towards the overall
group SHF and converse is true when skewness < 0.
Consequently, the current group is modi¯ed by
eliminating a single space with extreme RSHF value
based on the skewness coe±cient and equations are
reapplied. This procedure is iterated until the re-
quired level of RSD is achieved. The whole proce-
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
dure is repeated for the new space group composed
of the eliminated spaces, until every space in the
sample group is accounted and placed into a thermal
zone or forms a thermal zone by itself. If thermal
load in a particular space varies considerably from
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
the remaining spaces in the group, it forms a ther-
mal zone by itself after iterations. Figure 2 shows the
°owchart depicting the eliminate scheme.
Fig. 2. Flowchart depicting eliminate scheme for creating thermal zones from a sample space groups.
1850030-6
3. Air System Consideration
A DOAS splits ventilation loads from primary air
conditioning unit. Separate treatment of outdoor air
by DOAS unit not only reduces the size of main
supply duct and supply unit, but also improves in-
door air quality (IAQ) by decoupling dehumidi¯ca-
tion and sensible loads of outdoor air. Ideally, a
DOAS should be designed to handle entire space
latent load along with some portion of sensible
load.19–21 However, limitations on allowable outdoor
air°ow rates and minimum supply DPTs act as de-
terring factors in achieving this objective entirely.
Treated fresh air through a DOAS unit is either
supplied directly to the space or to main air-recir-
culating unit, the former being more e±cient.22,23
Treatment of outdoor air in a DOAS unit can be
done by integrating many ancillary components like
heat recovery devices, active desiccant wheels, heat
pipe coupled dehumidi¯er systems, etc. In the
 Thermal Zoning Based on Design Cooling Loads

present work, a wrap-around heat pipe (WAHP) l0 þ
Cplv T0

based dehumidi¯er system is considered as the main
outdoor tempering unit supplying ventilation air
directly to the spaces. Many researchers have
investigated the application of WAHP in HVAC
systems and studied its implications on the overall
system performance.24,25
In a WAHP enhanced dehumidi¯er system, the
supply dry bulb and DPT of air are controlled by
modulating condenser and dehumidi¯er coil outlet
temperatures, respectively. A WAHP precools in-
coming air through its evaporator, thereby reducing
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
¼ :
R
The sensible and latent loads met by DOAS unit
depend on the ventilation rate (FA) for each of the
spaces and given by
SLiDOAS ¼   FAi  cpm ðtset-i  ts-iDOAS Þ ; ð18Þ
LLiDOAS ¼   FAi  hfg ð!set-i  !s-iDOAS Þ : ð19Þ
If the loads mitigated by DOAS unit are greater
than the space loads, the design supply temperature
and humidity ratio of the DOAS unit need to be

its dew point depression (DPD) before passing it increased. The remaining loads to be met by local
through the dehumidi¯er coil. Due to this, air exits recirculating units are given by
the coil at almost saturated condition. Hence,
without loss of generality, the supply DPT is SLiAHU ¼ SLi  SLiDOAS ; ð20Þ
approximated by the coil exit temperature. The LLiAHU ¼ LLi  LLiDOAS : ð21Þ
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

condenser outlet temperature is maintained via

electrically operated solenoid valves that can control The RSHF of any space due the loads o®set by a
the °ow of refrigerant through heat pipe tubes based local recirculating unit is given by
on the required condenser outlet temperature.26 SLiAHU
RSHFiAHU ¼ : ð22Þ
SLiAHU þ LLiAHU
4. Thermal Zoning for DOAS with Local and slope of this line:
Recirculating Systems

cpm 1  RSHFiAHU
4.1. Optimum supply conditions miAHU ¼ : ð23Þ
hfg RSHFiAHU
at design loads
On a psychrometric chart, the line connecting coil
The supply temperature from DOAS unit is as-
entering and leaving conditions is called as grand
sumed the same as condenser outlet temperature of
sensible heat factor (GSHF)15 line. Since, DOAS
WAHP system. Hence:
unit separately treats and supplies ventilation air
ts-iDOAS ¼ toutlet : ð15Þ directly to the spaces, entering conditions of air to
the recirculating AHU coil is the same as return air
The supply DPT of DOAS unit is an assumed design
condition (or space set point). The slope of this line
value. Hence:
is given by
td-iDOAS ¼ DPTdesign : ð16Þ !set-i  !s-iGSHF
miGSHF ¼ : ð24Þ
Supply humidity ratio corresponding to the supply tset-i  ts-iGSHF
DPT27 of the DOAS unit can be determined by At minimum design ADP of recirculating coil, the
applying Eq. (17): supply temperature of recirculating unit for each

x 
!s-iDOAS ¼ !s-TZDOAS ¼
; ð17Þ space is given in terms of coil bypass factor (BF) as:
1x
ts-iGSHF ¼ ð1  BFÞ  ADPmin-design þ BF  tset-i :
where

 
 ð25Þ
P T0 1 1
x¼ exp  and supply humidity ratio:
Pt T0 td-iDOAS þ T0

!s-iGSHF ¼ ð1  BFÞ  !ADPmin-design þ BF  !set-i :
 f þ A  ðtd-iDOAS þ 2  T0 Þg ;
ð26Þ



Cplv Putting Eqs. (25) and (26) into Eq. (24), the mini-
A¼  ;
2R mum slope of design GSHF line for each space is

1850030-7
 M. Sarkar

determined and compared with the slope of space T0 solve for Tadp , successive roots are determined
RSHF line given by Eq. (23). If the slope of design iteratively until convergence is achieved. The suc-
GHSF line is greater than or equal to the slope of cessive root for Eq. (30) is determined by
RSHF line, the required supply condition of recir-
culating unit can be controlled by modulating gðTadpj Þ
Tadpjþ1 ¼ Tadpj  : ð31Þ
refrigerant °ow through the coil. However, if RSHF g 0 ðTadpj Þ
line is steeper than GSHF line, minimum design
ADP of the coil is not su±cient to meet remaining The design DPT of the space can be considered
space latent load and additional provisions like re- as the initial guess (Tadp0 Þ for obtaining successive
heat or bypass control are required to meet the solutions. DPT can be determined in terms of hu-
loads. Based on this comparison, the sample space midity ratio27 by Eq. (32) as
group are sorted into two sub-groups and corre- ð  kT0 Þ
td ¼
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.

sponding thermal zoning methodologies are applied.

2A
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
4.1.1. Procedure a ð  kT0 Þ 2 þ 4AðAT 02 þ T0 Þ
þ  T0 : ð32Þ
In this case: 2A
miAHU  miGSHF : ð27Þ After determining the supply ADP by numerical
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

iteration, required supply temperature and humidi-

The RSHF line intersects the saturation curve at a
ty ratio are obtained by
temperature higher than or equal to the lowest de-
sign ADP of AHU coil. Slope of the RSHF line given ts-iAHU ¼ ð1  BFÞ  tadp-iAHU þ BF  tset-i ; ð33Þ
by Eq. (23) can be written as !s-iAHU ¼ ð1  BFÞ  !adp-iAHU þ BF  !set-i ; ð34Þ


cpm 1  RSHFiAHU
miAHU ¼ where
hfg RSHFiAHU
!set-i  !adp-iAHU tadp-iAHU ¼ Tadp-iAHU  T0 :
¼ : ð28Þ
Tset-i  Tadp-iAHU The required supply °ow rate by the AHU is
In Eq. (28), both the supply ADP and humidity given by
ratio are unknown. The ADP (in Kelvin) can be SLiAHU
V_iAHU ¼ : ð35Þ
expressed in terms of humidity ratio through a   cpm  ðtset-i  ts-iAHU Þ
quadratic equation 27 as per Eq. (29):
The SHF of the sample space group served by AHU
A  ðTadp-iAHU Þ 2 þ ð  kT0 Þ  Tadp-iAHU and corresponding slope of SHF line are obtained by
 ðAT 02 þ T0 Þ ¼ 0 ; ð29Þ applying Eqs. (5) and (10), respectively. Supply
ADP temperature (in Kelvin) of the AHU serving an
where entire zone is determined by numerically solving

 Eq. (36) below:

Cplv
A¼  ;
2R A  ðTadp-TZAHU Þ 2 þ ð  kT0 Þ  Tadp-TZAHU



Pt !adp-iAHU  ðAT 02 þ T0 Þ ¼ 0 ð36Þ
k ¼ ln þ ln ;
P T0
þ !adp-iAHU
where
l0 þ
Cplv T0
¼ : !adp-TZAHU ¼ !set-TZ  mTZAHU
R
 ðTset-TZ  Tadp-TZAHU Þ : ð37Þ
One cannot solve Eq. (29) explicitly by applying
conventional method and hence, it should be solved
Similar to the methodology applied for individual
numerically. To solve this, Newton–Raphson method
space, ADP of the coil supplying to space group is
can be applied. The general form of Eq. (29) is
determined by applying Newton–Raphson method.
gðTadp Þ ¼ A  ðTadp Þ 2 þ ð  kT0 Þ  Tadp The initial guess in this case is the prescribed
DPT of the space group determined by substituting
ðAT 02 þ T0 Þ : ð30Þ corresponding parameters in Eqs. (7) and (32).

1850030-8

The required supply temperature and humidity
ratio of sample space sub-group are determined by
replacing corresponding parameters in Eqs. (33)
and (34), respectively. The revised space °ow rate
required for each space in the group is obtained by
applying Eq. (11) and required thermal zones are
created by applying the elimination scheme shown
before. Equation (43) below gives the resultant hu-
midity ratio of any space in the zone:
LLiDOAS þ LLiAHU
!i-chk ¼
  hfg  ðFAi þ V_i-rev Þ
AHU
ð!s-TZDOAS  FAi þ !s-TZAHU  V_i-revAHU Þ
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
þ :
ðFAi þ V_i-rev Þ
AHU
4.1.2. Procedure b
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
This procedure is applicable when condition given
by Eq. (27) is not valid and
miAHU > miGSHF :
In this case, the ADP of recirculating AHU serving
individual spaces of an entire thermal zone is con-
sidered equal to the minimum design (min-design)
coil ADP:
tadp-iAHU ¼ tadp-TZAHU ¼ ADPmin-design :
The corresponding humidity ratio at this saturation
temperature is determined by substituting the
minimum design coil ADP into Eq. (17). The supply
humidity ratio for each space is given by Eq. (34).
The required supply temperature for each space will
be higher than the AHU coil outlet temperature and
given in terms of the slope of space RSHF line as
!set-i  !s-iAHU
ts-iAHU ¼ tset-i  ; ð41Þ
miAHU
where slope of RSHF line is given by Eq. (28). The
required supply °ow rate for each space is deter-
mined by applying Eq. (35). The SHF of the sample
space group served by AHU and corresponding slope
of SHF line are obtained by applying Eqs. (5)
and (10), respectively. The supply humidity ratio for
the space group is given as
!s-TZAHU ¼ ð1  BFÞ  !adp-TZAHU
þBF  !set-TZ
and required zone supply temperature:
!set-TZ  !s-TZAHU
ts-TZAHU ¼ tset-TZ  : ð43Þ
mTZAHU
Thermal Zoning Based on Design Cooling Loads
The set points used in Eqs. (42) and (43) are
obtained from Eqs. (6) to (7). Equation (11) is ap-
plied in a similar way as before to determine revised
supply °ow rate for each space in the space group
and elimination scheme is applied to create thermal
zones. Equation (38) applied before for procedure a
gives the resultant humidity ratio of any space in a
thermal zone.
Section 4.1 concludes the general zoning meth-
odology of a DOAS with local recirculating units for
common ventilation and primary supply, respec-
tively. Based on the design space load character-
istics, speci¯c procedures shown above can be aptly
applied and optimum supply conditions and air°ow
rates for each thermal zone can be determined
ð38Þ
accordingly. Subsequently, cooling coil capacities
required for each thermal zone is determined by
considering these supply conditions and design zone
air°ow rates, which is followed by sizing of the
cooling plant. The proceeding section discusses the
ð39Þ basic controlling methodology for the zoned HVAC
systems considered in this work.
4.2. Consideration at part loads
At part load conditions, variations in space cooling
ð40Þ loads require air conditioning systems to respond in
appropriate manner to maintain comfortable con-
ditions. Dehumidi¯cation performance of HVAC
systems vary with the variations in space cooling
loads. Temperature reset strategy can control the
HVAC supply temperatures in response to the space
cooling loads. However, this may lead to reduced
dehumidi¯cation and higher indoor humidity levels.
In this regard, a constant air volume system (CAV)
that allows modulation of supply temperature, while
maintaining constant air°ow rate, introduces
greater challenge in controlling dehumidi¯cation at
part load conditions.28
Literature study suggests that a variable air
volume system (VAV) supplying at a speci¯c design
temperature can maintain both space temperature
and humidity levels within acceptable limits by
varying supply air°ow rates and operating terminal
reheat coils, whenever needed.29 Figure 3 depicts
ð42Þ typical RH contours of a zone served by a VAV
system at part load conditions. This system supplies
air at a constant DPT determined at design sensible
and latent loads and prescribed space set point
(in this case 50% RH and 24
C). Here, space air°ow
rates are modulated to maintain the required
1850030-9
 M. Sarkar
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
Fig. 3. Zone RH variation at part loads for VAV systems supplying air at design conditions.
temperature set point (24
C). The ¯gure shows that
over a wide range of variations in sensible and latent
loads, space RH levels stay between 40% and 60%
during air conditioning periods. This is the optimum
indoor RH range as per literature recom-
mendation for healthy and comfortable dwelling
environments.30
Extending the argument from a single space to a
multi-space zone, the same design strategy ensures
similar levels of space RH, when compounding with
a DOAS for better dehumidi¯cation performance at
part loads. Hence, under the purview of this system
con¯guration and operational strategy, the thermal
zoning scheme corresponding to design cooling loads
as per Sec. 4.1 remains valid even at part load con-
ditions.
This design scheme predicates the controlling
methodology for the zoned HVAC systems consid-
ered here. The air°ow modulation is achieved by the
VAV boxes for maintaining the space temperature
between the minimum and maximum design set
points. Air°ow is supplied to each thermal zone at
corresponding ¯xed supply temperatures through-
out the conditioning period. Despite active RH
controllers in primary air systems not being con-
sidered, appropriate thermal zoning ensures com-
paratively better RH control of thermal zone due to
similar load varying characteristics of constituent
spaces. The simulation case study in the proceeding
section applies this strategy for analyzing e®ect of
1850030-10
thermal zoning on operational characteristics of a
DOAS with local VAV recirculating units. Varia-
tions in space temperature and RH are the two basic
criteria considered for the performance of zoned
HVAC system in the simulation case study. The
acceptable range of thermal conditions inside a zone
is ascertained by the maximum and minimum limits
on the design thermostat and RH set points based
on the space application and comfort criteria for
occupancy.
5. Simulation Case Study
This section presents an energy performance study
of a prototype building located in New Delhi, India.
Figure 4 depicts the basic building geometry and
constituent spaces. Space typology and corre-
sponding thermal data considered for the model are
summarized in Table 1. The basic energy modeling
input parameters are listed in Table 2 and opera-
tional pro¯les of each space type are shown in Fig. 5.
5.1. Load estimation
Both internal and external sources a®ect the cooling
loads of a space. Lighting, equipment, occupants,
conduction, in¯ltration, solar transmission and
delayed release of stored heat through thermal mass
are the main sources of sensible heat gain, whereas
moisture transfer from occupants, equipment and
 Thermal Zoning Based on Design Cooling Loads
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

Fig. 4. Geometry of BEM.

Table 1. Space input summary.

HVAC Occupancy Lighting Receptacle Ventilation

Occupied Occupied
Space type max. (
C) min. (
C) RH (%) Weekly EFLH m 2 /person W/m 2 W/m 2 lps/m 2

Corridor 25  0.5 20  0.5 45  10 126 — 5 0 0.3

Classroom 24  0.5 22  0.5 50  10 78 5 15 10.8 1.1
Dining 24  0.5 22  0.5 50  10 126 5 10 9 1.7
Lobby 24  0.5 22  0.5 50  10 126 3.3 14 2.7 1.1
Meeting room 24  0.5 22  0.5 50  10 49 5 14 5.4 0.8
O±ce 24  0.5 22  0.5 50  10 78 20 12 10.8 0.4
Retail 24  0.5 22  0.5 50  10 126 6.7 18 2.7 1.2
Stairs — — — — — 6 — —
Toilet — — — — — 10 — —

in¯ltration are the main sources of latent heat gain.
Contrary to internal loads that vary based on ap-
plication trends, external loads depend on the
weather conditions. Due to this, space sensible and
latent loads vary both diurnally and seasonally.
IES-VE is the energy simulation and load esti-
mation tool applied in the present work. VE is a
whole-building simulation tool based on ¯rst prin-
ciple for mathematically modeling dynamic heat
transfer and advanced HVAC system control.31
There are two methods for estimating cooling loads
in IES-VE. The ¯rst method is the conventional way
of load calculation by ASHRAE-LOADS module.
For this method, standard design DBT and WBT of
the location from ASHRAE database are set into the
program as main weather input and heat calculation
is done by applying ASHRAE heat balance meth-
od.32 Another way for load calculation is through
dynamic simulation pathway. In this method, con-
ditioned spaces are linked with a virtual HVAC
system (called as main system) and weather data is
accessed from a simulation weather ¯le. Running
dynamic simulations with this setting enables the
program to determine sensible and latent loads via
heat balance method at realistic hourly outdoor
conditions. In the present work, simulation pathway
is applied for determining space-cooling loads that
are required for thermal zoning and system sizing.

1850030-11
 M. Sarkar

Table 2. Input parameters for energy model.

Parameter Details

Location: New Delhi, India

Geometry: G þ 1, Mixed-use type, Total °oor area ¼ 4545 m 2
Envelope details:
External wall Insulated cavity wall sandwiched with plaster boards, U-value ¼ 0:705 W/m 2 -K
Roof Concrete deck with insulation entirely above deck and weatherproof membrane,
U-value ¼ 0:362 W/m 2 -K
Glazing Double pane window with air cavity, U-value ¼ 6:801 W/m 2 -K and SHGC ¼ 0:25.
Roof light Roof light without curb, U-value ¼ 7:72 W/m 2 -K and SHGC ¼ 0:36.
Base In¯ltration rate 0.362 lps/m 2 -facade
System control:
Primary system VAV AHU with terminal reheat coils
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.

Primary system control Flow modulation based on zone temperature. Terminal air dampers and VFD fan for °ow
modulation. Terminal reheat coils for space temperature control when required. Start-up
period for pre-occupancy operation is 2 h. Temperature setback during unoccupied hours.
Cooling plant Chilled-water system. Higher side speci¯cations as per section G3.1.3 Appendix G, ASHRAE std.
90.1-2007 5 .
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

Simulation speci¯cs:
Simulation time period Summer season (April 1st to September 30th).
Simulation time-step 10 min.
Weather ¯les (1) IND New.Delhi.421820 IWEC.epw (IWEC weather data)
(2) IND New.Delhi.421820 ISHRAE.epw (ISHRAE weather data)
(3) New Delhi WA 2016.fwt (Weather Analytics historical weather data for 2016)

Simulation tool IES-VE (2015)

Fig. 5. Operating pro¯les of various space types considered in the model.

1850030-12

5.2. Weather data
In this article, three simulation weather ¯les of
New Delhi from di®erent sources are utilized for
load calculations and dynamic simulations. The ¯rst
two simulation weather ¯les are created, compiled
by International Weather for Energy Calculation
(IWEC) and Indian Society of Heating, Refrigerat-
ing and Air-Conditioning Engineers (ISHRAE), and
available as EnergyPlus weather data source.33 The
third weather ¯le incorporates historical data for
year 2016 from Weather Analytics (WA) database34
and extracted as an .fwt ¯le through cloud-based
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
building analytics service called IES-SCAN.35 The
variation of temperature and humidity data from
the simulation weather ¯les are shown in Fig. 6.
From the data plot, it is observed that the extreme
monthly DBT during summer season is relatively
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
higher in WA-2016 weather ¯le. However, IWEC
and ISHRAE weather ¯les show higher monthly
median of humidity ratio during summer season
than the WA-2016 weather ¯le.
5.3. Thermal zoning
Occurrence of peak cooling loads inside spaces may
vary with the weather data considered. By applying
Fig. 6. Monthly variation of temperature and humidity data from simulation weather.
1850030-13
Thermal Zoning Based on Design Cooling Loads
the sorting criteria from Sec. 2.2, di®erent space
groups are created. From each space group, thermal
zones are generated by applying the procedure
proposed for DOAS with recirculating AHU. Fol-
lowing assumptions are made for the HVAC systems
in this study:
. Refrigerant for cooling dehumidi¯cation process is
chilled water.
. Design supply DPT of 12
C and condenser outlet
temperature of 18
C are considered for the DOAS
sub-system.
. Coil BF of 0.1 is considered for all the recirculat-
ing AHU.
. A minimum limit of 11
C is considered for the
supply ADP of recirculating AHU serving thermal
zones.
5.4. Dynamic simulations
Detailed HVAC networks with corresponding air-
side and plant-side controllers are compiled in
APACHE-HVAC module of IES-VE. The HVAC
control-network and simulation weather ¯le should
be linked together at APACHE-SIM interface for
modeling dynamic interactions between building
HVAC systems and outdoor climate of the location.
 M. Sarkar

Table 3. Simulation test scenarios.

Scenario # Zoning Ventilation Linked weather ¯le Design supply Temp

1 Zoned DOAS (WAHP dehumidi¯er system) IND New.Delhi.421820 ISHRAE. As per zoning method
epw
1a, 1b, 1c, 1d Un-zoned Untreated direct mechanical ventilation IND New.Delhi.421820 ISHRAE. 11, 12, 13 and 14
C
epw
1e, 1f, 1g, 1h Un-zoned DOAS (WAHP dehumidi¯er system) IND New.Delhi.421820 ISHRAE. 11, 12, 13 and 14
C
epw
2 Zoned DOAS (WAHP dehumidi¯er system) IND New.Delhi.421820 IWEC.epw As per zoning method
2a, 2b, 2c, 2d Un-zoned Untreated direct mechanical ventilation IND New.Delhi.421820 IWEC.epw 11, 12, 13 and 14
C
2e, 2f, 2g, 2h Un-zoned DOAS (WAHP dehumidi¯er system) IND New.Delhi.421820 IWEC.epw 11, 12, 13 and 14
C
3 Zoned DOAS (WAHP dehumidi¯er system) New Delhi WA 2016.fwt As per zoning method
3a, 3b, 3c, 3d Un-zoned Untreated direct mechanical ventilation New Delhi WA 2016.fwt 11, 12, 13 and 14
C
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.

3e, 3f, 3g, 3h Un-zoned DOAS (WAHP dehumidi¯er system) New Delhi WA 2016.fwt 11, 12, 13 and 14
C

Table 3 shows the HVAC system variants for this to determine optimum supply conditions and °ow
case study. Three basic types of air systems are rates of zoned HVAC systems. Figure 8(b) shows
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

considered for the comparative study: the hierarchical levels in the model for thermal
zoning process.
. DOAS with multiple VAV recirculating AHU
serving corresponding thermal zones.
. DOAS with a single VAV recirculating AHU
serving all the conditioned spaces by con-
sidering entire building as a single thermal block
(un-zoned systems). Design supply temperatures:
11, 12, 13 and 14
C.
. A conventional VAV recirculating AHU with di-
rect mechanical ventilation serving all the condi-
tioned spaces by considering entire building as a
single thermal block. Design supply temperatures:
11, 12, 13 and 14
C.
Figure 7 shows the schematics of these system
con¯gurations. In a test scenario, an air system (a)
network is linked with one simulation weather ¯le
at a time and simulation is ran for an entire
summer season i.e., from April to September
months. The result parameters considered for as-
sessment are:
. HVAC energy utilization intensity (EUI–
kWh/m2).
. Total chiller load (kWh/m2).
. Temperature unmet hours: Number of occupied (b)
hours when space temperatures are either
> 24:5
C (occupied maximum þ 0:5
C) or
< 21:5
C (occupied minimum – 0.5
C).
. RH unmet hours: Number of occupied hours when
(c)
space RH are either > 60% or < 40%.
Fig. 7. Schematics of HVAC system con¯gurations considered
Figure 8(a) shows the general layout of the simula- for simulation case study (a) DOAS with multiple VAV AHUs
tion work°ow adapted in this case study. The sort (zoned); (b) DOAS with a single VAV AHU (un-zoned) and (c) A
and eliminate schemes for thermal zoning is applied single VAV AHU with untreated direct ventilation (un-zoned).

1850030-14
 Thermal Zoning Based on Design Cooling Loads
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com

(a)

(b)
Fig. 8. General outline of simulation case study: (a) IES-VE application work°ow and (b) Hierarchical levels considered for thermal
zoning.

5.5. Results and discussion
Before assessing the performance of the considered
options, quantitative validation of the simulation is
¯rst con¯rmed. As per performance rating method
(PRM) from ASHRAE 90.1 Standard appendix G5,
the temperature unmet hours in a validated building
energy model (BEM) should not be greater than 300
and di®erence in unmet hours between two consid-
ered models (baseline and proposed models) should
not exceed 50. Since, only the summer season is
considered as the simulation period in this study, the
permissible temperature unmet hours and di®erence
in unmet hours between various cases are reduced to
150 and 25, respectively, on a pro-rata basis.
Figure 9(a) shows the temperature unmet hours
for all the considered cases. The plot provides satis-
factory evidence, that the control settings and sys-
tem sizing methodology in of all the simulation cases
are conformal with the basic validation criteria of
ASHRAE standard 90.1. This is valid for the system
types that are sized by considering a speci¯c tem-
perature di®erence between room set points and de-
sign supply temperatures, as well as that are sized by
the thermal zoning methodology presented in Sec. 2.

1850030-15
 M. Sarkar
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
(a)
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
(b)
Fig. 9. (a) Temperature unmet hours for simulation scenarios and (b) simulation energy summary for the considered scenarios.
Figure 9(b) summarizes the energy consumption
from various end sources considered in the model.
Energy consumption due to chillers, reheat coils,
distribution fans, chilled water distribution pump
and heat rejection loop comprises the net energy
consumption due to HVAC. Energy consumed by
lighting and miscellaneous equipment are considered
the same and contributes 28–30% of overall energy
consumption in all the simulation scenarios. As the
supply temperature reduces, both chiller and reheat
energy increases, whereas distribution fan energy
decreases and vice-versa. The ¯gure shows that
DOAS based VAV systems perform relatively better
than conventional VAV systems with direct venti-
lation. Separately treating fresh air through a
DOAS sub-system equipped with a WAHP en-
hanced dehumidi¯er module reduces the coil loads of
the primary recirculating units. Despite this, the
di®erences in total energy and HVAC energy con-
sumptions between the best and worst cases in
1850030-16
corresponding simulation weather sets only vary
between 5–6% and 7–9%, respectively. These per-
centages are subjected to change with variations in
model inputs and outdoor air°ow rates supplied to
the building. Nevertheless, similar types of systems
with di®erent zoning considerations should not only
consider energy consumption but also system capa-
bility to maintain indoor environmental quality as a
primary factor for qualitative assessment.
Figure 10 depicts the scatter plots of total chiller
loads and RH unmet hours against HVAC EUIs for
all the considered scenarios. Each scatter plot depicts
the performance of nine basic system con¯gurations
that are sized and simulated with a speci¯c simula-
tion weather ¯le. The data points are clustered into
three di®erent groups for qualitative characteristic
assessment by applying K-means clustering meth-
odology based on Euclidian distance metric. Similar
to temperature unmet hours, a permissible RH
unmet hour limit of 0–150 is considered here.

by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
Fig. 10. Scatter plots of total chiller loads and RH unmet hours versus HVAC EUIs for simulation scenarios.
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
The maximum percentage di®erences in HVAC
EUI and total chiller loads between zoned recircu-
lating AHU and the best performing un-zoned recir-
culating AHU are less than 4% and 3%, respectively,
in all the considered simulation scenarios. Since,
DOAS supplies ventilation air at the same condition,
the di®erences in HVAC EUI and total chiller loads
are primarily due to di®erences in supply conditions
and design air°ow rates. Un-zoned recirculating sys-
tem considers the whole building as a single thermal
block and supplies conditioned air at the same tem-
perature regardless of its thermal load character-
istics. On the other hand, zoned recirculating systems
consider load characteristics to optimize design sup-
ply conditions for each thermal zone. In addition, the
moisture controlling capability of zoned recirculating
systems with DOAS is consistently better than other
systems. The zoned recirculating system con¯gura-
tion shows its worst performance with WA-2016
simulation weather ¯le, where approximately 98% of
occupied hours stay within the mandated RH limits
(40–60%). This performance is substantially better
than the other considered system con¯gurations. The
clustering methodology applied here places the per-
formance points of this system con¯guration into
`better performing' clusters across all the simulation
scenario sets. The results clearly shows that a DOAS
with zoned recirculating systems perform better in
maintaining indoor environmental conditions than
un-zoned systems without signi¯cantly compromis-
ing the energy consumption and total equipment
cooling loads.
1850030-17
Thermal Zoning Based on Design Cooling Loads
This simulation case study provides su±cient
evidence that the thermal zoning methodology in-
troduced in the present work is an e®ective design
strategy, which is not only e±cient but provides
good control over the space conditions without any
detrimental e®ects on energy performance.
6. Conclusions
A thermal zoning methodology based on design
space cooling load is presented in this paper. This
method followed \sort and eliminate" schemes for
¯rst creating sample space groups and then forming
multiple clusters of spaces served by respective
HVAC systems. The optimized design supply con-
ditions for the served zones are also determined
through the thermal zoning framework that pri-
marily depends on the type of HVAC system and
weather data considered for load calculations. The
present methodology is applied for sizing and de-
signing a DOAS with local recirculating system.
A simulation case study is done by applying three
di®erent simulation weather ¯les of New Delhi to a
prototype-building model. Results showed that a
DOAS with zoned recirculating systems consistently
keep more than 98% of occupied hours within the
design temperature and RH limits, without in-
creasing the overall HAVC energy and chiller loads
beyond 4% and 3% from the best performing cases,
respectively. This substantiates its superior quali-
tative performance characteristics. The methodolo-
gy presented here for creating thermal zones is
 M. Sarkar
computationally simple and established on basic
mathematical formulations making it more intui-
tive than the conventional way, which is more
subjective and depends on discretion and experi-
ence of designer.
Acknowledgments
The author acknowledges no con°ict of interest.
This research did not receive any speci¯c grant from
funding agencies in the public, commercial, or not-
for-pro¯t sectors. The author acknowledges full
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
support given by IES during the course of this work.
References
1. United Nations Department of Economic and Social
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
A®airs: Population Division, World Population
Prospects — The 2015 Revision. New York, USA,
2015.
2. United Nations Development Programme, Promot-
ing Energy E±ciency in buildings: Lessons Learned
from International Experience, New York, USA,
2010.
3. U.S. Energy Information Administration, Commer-
cial Buildings Energy Consumption Survey: Energy
Usage Summary, Washington DC, USA, 2012.
4. S. Royer, S. Thil, T. Talbert and M. Polit, A pro-
cedure for modeling buildings and their thermal
zones using co-simulation and system identi¯cation,
Energy Build. 78 (2014) 231–237.
5. ANSI/ASHRAE Standard 90.1-2007, Energy stan-
dard for buildings except low-rise residential build-
ings. American Society of Heating, Refrigeration
and Air Conditioning Engineers, Inc., GA, U.S.A.,
2007.
6. N. Jones, C. McCrone, B. Walter and K. Pratt,
Automated translation and thermal zoning of digital
building model for energy analysis, Proc. 13th Int.
Build. Perform. Sim. Assn. Conf., Chambery,
France, 2013, pp. 202–209.
7. T. Dogan and C. Reinhart, Autozoner an algorithm
for automatic thermal zoning of buildings with un-
known interior space de¯nitions, J. Build. Perform.
Sim. 9 (2015) 176–189.
8. T. Dogan, C. Reinhart and P. Michaltaos, Auto-
mated multi-zone building energy model generation
for schematic design and urban massing studies, Int.
Build. Perform. Sim. Ass. (IBPSA) e-Sim Conf.,
2014, Ottawa, Canada, Jan., 2014.
9. T. Dogan, C. Reinhart and E. Saratsis, The optimiza-
tion potential of °oor-plan typologies in early design
energy modeling, Proceedings of BS2015: 14th Con.
1850030-18
Int. Build. Perform. Sim. Assn., Hyderabad, India,
Dec. 7–9, 2015, pp. 1853–1860.
10. BEMBOOK (2017), Thermal Zoning Determina-
tion, http://bembook.ibpsa.us/index.php?title=
Thermal Zoning Determination.
11. ISO 13790-2008, Energy performance of buildings —
calculation of energy use for space heating and
cooling, International Organization for Standardi-
zation, Geneva, Switzerland, 2008.
12. COMNET (2017), Thermal Blocks, HVAC Zones
and Space Functions, https://comnet.org/22-ther-
mal-blocks-hvac-zones-and-space-functions.
13. C. De Souza and S. Alsaadani, Thermal zoning in
speculative o±ce buildings: Discussing the connec-
tion between space layout and inside temperature
control, First Building Simulation and Optimazation
Conference, Loughborough, UK, 10–11 Sept., 2012,
pp. 417–424.
14. Chapter 28 — Nonresidential Cooling and Heating
Load Calculations, ASHRAE Handbook — Funda-
mentals, American Society of Heating, Refrigeration
and Air Conditioning Engineers, Inc., GA, U.S.A.,
1997.
15. Chapter 8 — Applied Psychrometrics, Part 1 Load
Estimating, Carrier System Design Handbook, Car-
rier Corporations, Syracuse, New York, USA, 1960,
pp. 115–151.
16. M. Sarkar, Theoretical comparison of cooling loads
of an air handling unit in blow-through and draw-
through con¯gurations, Energy Build. 64 (2013)
239–248.
17. Mathematics stack exchange, Standard deviation
of weighted-mean, https://math.stackexchange.
com/questions/320441/standard-deviation-of-the-
weighted-mean.
18. ANSI/ASHRAE Standard 169-2013, Climatic Data
for Building Design Standards, American Society of
Heating, Refrigeration and Air Conditioning Engi-
neers, Inc., GA, U.S.A., 2013.
19. S. Mumma, Designing dedicated outdoor air system,
ASHRAE J. 43 (2001) 28–31.
20. M. Larranaga, M. Beruvides, H. Holder, E.
Karunasena and D. Straus, DOAS and humidity
control, ASHRAE J. 50 (2008) 34–40.
21. H. Crowther and Y. Ma, Design considerations for
dedicated OA systems, ASHRAE J. 58 (2016) 30–42.
22. S. Mumma, Terminal equipment with DOAS: Series
vs. parallel, Engin. Syst. 25(5) (2008) 86–89.
23. S. Deng, L. Lau and J. Jeong, Do all DOAS con¯g-
urations provide the same bene¯ts? ASHRAE J. 56
(2014) 52–57.
24. H. Chaudhary, B. Hughes and S. Ghani, A review of
heat pipe systems for heat recovery and renewable
energy applications, Renew. Sustain. Energy Rev. 16
(2012) 2249–2259.

25. L. Zhang and W. Lee, Evaluating the use heat pipe for
dedicated ventilation of o±ce buildings in Hong Kong,
Energy Convers. Manage. 52 (2011) 1983–1989.
26. D. John, Wrap-around heat pipes in humid climates,
ASHRAE J. 58 (2016) 28–38.
27. M. Sarkar, A new theoretical formulation of dew
point temperatures applicable for comfort air-cool-
ing systems, Energy Build. 86 (2015) 243–256.
28. J. Murphy, Dehumidi¯cation performance of HVAC
systems, ASHRAE J. 44 (2002) 23–31.
29. J. Murphy, Using VAV to limit humidity at part
load, ASHRAE J. 52 (2010) 18–22.
30. E. M. Sterling, A. Arundel and T. D. Sterling, Cri-
by WSPC on 09/10/18. Re-use and distribution is strictly not permitted, except for Open Access articles.
teria for human exposure to humidity in occupied
buildings, ASHRAE Trans. 91 (1985) 611–622.
Int. J. Air-Cond. Ref. Downloaded from www.worldscientific.com
1850030-19
Thermal Zoning Based on Design Cooling Loads
31. VE for engineers, https://www.iesve.com/software/
ve-for-engineers.
32. J. Spitler, Load Calculations Application Manual,
American Society of Heating, Refrigeration and Air
Conditioning Engineers, Inc., GA, U.S.A., 2009.
33. Weather Data — New Delhi, https://energyplus.
net/weather-region/asia wmo region 2/IND%20%
20.
34. Weather data from Weather Analytics, https://
www.iesve.com/support/weather¯les/weather-ana-
lytics.
35. IES SCAN: Interrogating metered data to optimize
energy e±ciency in real buildings, https://scan.
iesve.com/unauthenticated/home.