Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 91 (2016) 844 – 856

SHC 2015, International Conference on Solar Heating and Cooling for Buildings and Industry

Multi-effect absorption chillers powered by the sun: reality or

reverie
Ali Shirazia,*, Robert A. Taylora, Stephen D. Whiteb, Graham L. Morrisona
a
School of Mechanical and Manufacturing Engineering, The University of New South Wales (UNSW), Kensington, New South Wales 2052,
Australia
b
Commonwealth Scientific and Industrial Research Organization (CSIRO) Energy Centre, Newcastle, New South Wales 2304, Australia

Abstract

The feasibility of solar-powered multi-effect LiBr-H2O absorption chillers is investigated under different climate conditions, and
three configurations of solar absorption chillers are proposed with respect to the type of the chiller. In the first configuration, a
single-effect absorption chiller is coupled with evacuated tube collectors (ETCs), while parabolic trough collectors (PTCs) are
utilized to run double- and triple-effect LiBr-H2O chillers in the second and third configurations, respectively. A simulation
model for each configuration is developed in TRNSYS 17 environment. A parametric study is conducted by varying the fraction
of direct normal irradiance (DNI) under three global horizontal irradiance (GHI) categories (low, medium and high) to determine
which configuration results in the most efficient and cost-effective performance of the plant under a given climate. The
simulation results reveal that there is no advantage in using solar multi-effect absorption chillers over solar single-effect chillers
when the DNI fraction is less than 60% of the total solar global irradiance. However, a significant decrease in the size of the solar
field for both double- and triple-effect chillers can be achieved in climates where the DNI fraction is above 60%. The cost
analysis results indicate that a minimum DNI fraction of about 70% is required in order for solar-driven multi-effect chillers to be
cost-competitive compared to solar single-effect chillers. The sensitivity analysis shows that the capital cost of PTCs needs to be
20% below today’s nominal cost for multi-effect absorption chillers to be more cost-effective than ETC powered single-effect
chillers in climates with relatively high DNI fraction (>~50í60%).
© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2015 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review by the scientific conference committee of SHC 2015 under responsibility of PSE AG.
Peer-review by the scientific conference committee of SHC 2015 under responsibility of PSE AG
Keywords: solar cooling; absorption chiller; multi-effect; air-conditioning; TRNSYS

* Corresponding author
E-mail address: a.shirazi@unsw.edu.au

1876-6102 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review by the scientific conference committee of SHC 2015 under responsibility of PSE AG
doi:10.1016/j.egypro.2016.06.251
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 845

1. Introduction

Air-conditioning demand in many countries accounts for about half of the energy use in buildings – which is
primarily supplied by fossil fuels [1, 2]. The use of renewable energy technologies in air-conditioning can reduce
fossil fuel consumption and its associated environmental impacts [3]. Solar thermal cooling technology offers a
clean alternative where supply is generally well-correlated with demand [4-6]. Absorption chillers are reliable,
energy efficient and commercially developed, as compared to the other types of thermal cooling processes [7]. There
are three types of absorption chillers commercially available on the market í single-, double-, and triple-effect
chillers. The driving heat source temperature for single-effect chillers is about 80-100 °C, while their COP is limited
to around 0.7 [8]. Double- and triple-effect chillers, on the other hand, require driving temperatures of around 180-
210 °C, and can reach COPs of up to 1.4 and 1.8, respectively [8].
The majority of solar absorption chillers installed around the world today are based on single-effect chillers
coupled with flat plate or evacuated tube collectors (ETCs), which deliver relatively low-temperatures [9, 10].
Recently, the combination of high-temperature concentrating solar thermal collectors – e.g. parabolic trough
collectors (PTCs) – with multi-effect chillers is becoming more attractive due to their higher COP [11, 12]. This
means that the multi-effect chillers require less solar thermal energy (and potentially less collector area) to supply
the same amount of cooling. However, the requirement for higher driving temperatures necessitates concentrating
collectors. Concentrating systems may be penalized by the lower solar gain per unit area because they can utilize
only the direct normal irradiance (DNI) and require tracking, as opposed to flat plate and evacuated tube collectors
which can harness both beam and diffuse radiation and can be installed closer together [13].
While much of the literature is dedicated to solar-assisted single-effect absorption chillers [14-17], a few
researchers have investigated the use of multi-effect absorption chillers fed by solar thermal energy. Ayadi et al. [11]
reviewed the available studies employing concentrating solar collectors for solar cooling and defined the technical
aspects necessary for future development. They found that such systems are still not common but estimated a
steadily growing market for air-conditioning and refrigeration facilities driven by concentrating solar thermal
collectors. Osborne et al. [18] studied a unique tri-generation system integrating parabolic trough collectors, flat
plate collectors and photovoltaic panels to provide chilled water, hot water, and electricity to the Faculty of
Engineering and Information Technology building at the University of Technology, Sydney (UTS), Australia. The
heat from the parabolic trough collectors was used to run a small-scale Organic Rankine Cycle (ORC) unit as well as
a small-scale ammonia/water absorption chiller to provide either electricity or chilled water, respectively. The heat
rejected from the condenser of both units was fed into the domestic hot water supply of the building along with the
heat from the flat plate collectors. The authors found that using commercially available components to design such
systems with a limited budget was challenging and care was required with system design and integration. Ming et al.
[19] conducted a numerical and experimental analysis of a small-scale high-temperature solar absorption chiller
prototype installed at Carnegie Mellon University, USA. The plant consisted of 52 m2 of parabolic trough collectors,
a 16 kW double-effect LiBr-H2O absorption chiller, and a heat recovery heat exchanger. The authors determined the
optimal collector orientation, collector receiver area, storage tank volume, and pipe lengths and diameters through a
detailed parametric study of the system. Li et al. [20] studied the effect of evaporator, condenser and absorber
temperatures of a double-effect absorption chiller on the solar collector outlet temperature in order to improve the
overall performance of the plant from thermodynamic point of view.
Triple-effect absorption chillers can operate more efficiently than single- and double-effect chillers when coupled
with concentrating solar thermal collectors [21]. Very few researchers have studied the potential of triple-effect
absorption systems fed by solar thermal energy. Agrawal et al. [22], for example, proposed a solar-powered triple-
effect absorption chiller integrated with an ejector and cascaded cycles for refrigeration applications at temperatures
ranging from í80°C to í50°C. The authors investigated the effect of key design parameters on the energetic and
exergetic efficiencies of the plant and concluded that solar-driven triple-effect absorption refrigeration systems are
suitable for this application from thermodynamic viewpoint.
Rather than focus on a particular technology, this study seeks to answer the question: Which type of chiller
(single-, double- or triple-effect) is the most efficient/economical design when coupled with solar thermal collectors
in a given climate? To the authors' knowledge, no comprehensive performance comparison amongst solar-assisted
multi-effect absorption chillers has been done for air-conditioning applications. As such, the present work evaluates
846 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

the energetic and economic feasibility of such systems under different climate conditions for air-conditioning
applications. Evacuated tube collectors are used to feed a typical single-effect chiller, while parabolic trough
collectors are employed to supply high-temperature heat to double- and triple-effect chillers. The proposed systems
have been modeled in TRNSYS 17 software [23] in detail. A parametric study is performed in order to evaluate the
influence of DNI fraction (under low, medium, and high GHI levels) on the size of the solar field for multi-effect
chillers as compared to the typical single-effect chiller. Annual cooling output from solar energy (per unit collector
area) as well as the solar field capital cost ratio has been considered as the energetic and economic indicators,
respectively. Finally, a sensitivity analysis is conducted to investigate the effect of variation in the capital cost of
PTCs on the economic performance of the proposed configurations.
Nomenclature

Aa aperture area (m2)

Ar receiver area (m2)
c1 first-order heat loss coefficient (W/m2 K)
c2 second-order heat loss coefficient (W/m2 K2)
cp specific heat at constant pressure (kJ/kg K)
CRSC the collector concentration ratio (-)
Fc  WD n collector zero loss efficiency at normal incidence (-)
Gt global irradiance on the tilted collector (W/m2)
K(ș) incidence angle modifier (-)
m mass flow rate (kg/s)

Q heat transfer rate (kW)
T temperature (°C)
R2 coefficient of determination
U overall heat loss coefficient (W/m2 K)
¨¨Tƍ characteristic temperature difference (°C)
Ș thermal efficiency (-)
ș solar incidence angle on the collector (°)
Subscripts
a air, ambient
AC absorber-condenser
ACH absorption chiller
AH auxiliary heater
avg average
b beam
C cooling
CHW chilled water
CT cooling tower
CW cooling water
d diffuse
DV diverting valve
E evaporator
ETC evacuated tube collector
G generator
HW hot water
MV mixing valve
P pump
PRV pressure relief valve
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 847

SC solar collector
SCW solar collector water
ST storage tank
u useful
w water

2. Solar absorption system configuration

The main components of the plant are shown in Fig. 1, the solar field (either ETCs or PTCs), storage tank (ST),
auxiliary gas burner (AH), absorption chiller (ACH), and the control system.
ACH
MV1 MV2
PRV
P2
P3

Recirculation
CW

HW
CT
SC

ST AH
SCW

DV1 DV2
P1

CHW Load

P4

Fig. 1. Schematic diagram of a solar absorption chiller system

The incident solar radiation absorbed by the solar collectors increases the storage tank water temperature through
the solar loop water circulated by a variable speed pump (P1). A feedback controller is used to adjust the pump flow
rate to achieve a fixed set-point temperature at the collector outlet. A pressure relief valve was used to limit the fluid
outlet temperature to the maximum allowable value specified by the user as an input to the model. The single-effect
chiller was coupled with evacuated tube solar collectors operating up to 100 °C, while high-temperature parabolic
trough collectors were employed to supply hot water with temperatures up to about 185 °C and 215 °C for double-
and triple-effect absorption chillers respectively. When the temperature at the bottom of the storage tank is above the
chiller’s required hot water temperature, only the solar-heat source is used to drive the chiller. Once the temperature
at the top of the tank drops below the required value, the chiller is switched to gas mode where the auxiliary gas
burner supplies the full load required by the chiller. This approach ensures a longer operating period through solar
heat and avoids possible frequent switching between the tank and the auxiliary gas burner [24].

3. System simulation

The solar absorption system described in the previous section was modeled in TRNSYS 17 [23], which is
commonly used for simulation of transient thermodynamic systems. The majority of component models were taken
from the TRNSYS library, while a number of new components (user-defined types) were developed by the authors
in C++ and were imported into TRNSYS environment (e.g. the parabolic trough collector, solar-gas heat source
controller, and the absorption chillers). In the following, the mathematical models of the main system components
are briefly described.

3.1. Evacuated tube solar collector (ETC)

The thermal efficiency of the collector can be determined using Hottel-Whillier-Bliss equation [25]:
848 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

KSC Fc  WD n K  T   c1
T avg  Ta   c Tavg  Ta  (1)
2
Gt Gt

where Fc  WD n is the collector zero loss efficiency at normal incidence, c1 and c2 are the first and second order heat
loss coefficients, and Tavg and Ta are the average temperature of the collector working fluid and ambient
temperature, respectively. Gt is the global irradiance on the tilted collector and K(ș) is the incidence angle modifier
(IAM). The evacuated tube collectors have been modeled using the ‘Type 71’ model of the TRNSYS standard
library. The parameters used for the model in the TRNSYS simulations are representative of the Apricus 30-tube
evacuated tube collector [26]. The collector manufacturer’s specific parameters are listed in Table 1. The test data
for the IAM is given in the collector test report [26].

Table 1. Collector performance parameters according to the test report provided by manufacturer

Parameter Unit Apricus 30-tube ETC [26] NEP PolyTrough 1800 [27]
Aperture area m2 2.83 18.45
Max operating pressure bar 8 40
Test flow rate L/h m2 61 75.9
Collector azimuth ° 0 0
Concentration ratio - NA 54
Fc  WD n - 0.687 0.689
c1 W/m2 K 1.505 0.36
c2 W/m2 K2 0.011 0.0011
Collector set-point °C 100 185/215
Pump 1 non-zero output % 5 5
b0 - NA 0.122

3.2. Parabolic trough solar collector (PTC)

Parabolic trough solar collectors are used to provide pressurized high-temperature hot water into the generator of
the multi-effect absorption chillers. The useful heat collected from the collector ( Q 
u ,SC ) can be determined as

follows [25]:
ª FcU L º

Q u ,SC
 w c p,w  TSC,out  TSC,in 
m A a « Fc  WD n K Tb  T  G b 
CR SC

Tavg  Ta »  (2)
¬ ¼
where

FcU L c1  c 2 Tavg  Ta  (3)
Aa
CR SC (4)
Ar

Kșb (ș) and Gb are the incidence angle modifier for beam radiation and the solar beam irradiance on the collector
surface. F c and UL are the collector efficiency factor and overall heat loss coefficient, and Aa and Ar are the collector
aperture area and receiver area. CRSC is the collector concentration ratio. The IAM coefficient for beam radiation
(Kșb) is calculated by [25]:

§ 1 ·
KTb  T  1  b0 ¨  1¸ (5)
© cos T ¹
where b0 is a coefficient, and ș is the beam incidence angle. The design parameters and IAM coefficient used for this
type are representative of NEP Polytrough 1800 [27]. The specifications of the collector according to its
performance test report are presented in Table 1.
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 849

3.3. Hot water storage tank

A TRNSYS Type534 TESS model [28] was used to simulate the performance of a stratified hot water storage
tank in the solar cooling plant. The tank is assumed to have an aspect ratio of 3.5 [29], with a heat loss coefficient of
0.833W/m2K. The tank model used 10 thermal stratification nodes. More details about the mathematical model of
the tank can be found in TRNSYS TESS Library documentation [28].

3.4. Auxiliary gas burner

The model of the auxiliary heater takes into account the overall efficiency of the burner and the heat losses
between the heater and the environment. The outlet temperature of the heater can be determined by [30]:
 Q
Q 
TAH,out TAH,in  AH loss,AH (6)
m w,AH cp,w

3.5. Absorption chiller

In this study, three types of absorption chillers (i.e. single-, double-, and triple-effect) have been considered for
solar cooling purposes. The chiller model is based on the characteristic temperature function, ¨¨Tƍ, which can be
expressed as follows [31]:
''Tc TG  a u TAC  e u TE (7)
The characteristic equations for the evaporator and generator heat transfer rates can be expressed as a function of
¨¨Tƍ:
Q s E ''T c  rE (8)
E

QG s G ''T c  rG (9)
The terms a, e, sE, rE, sG, and rG in Eqs. (7)-(9) are the characteristic coefficients which are determined using
multiple linear regression algorithms applied to a set of performance data points given by the chiller manufacturer.
The heat removal rate from the absorber and condenser can be determined by:

Q  Q
Q  Q
 (10)
AC E G aux

where Q
aux is the energy consumption of the absorption chiller pump. The COP of the chiller is calculated as
follows:
Q
E
COP (11)
 Q
Q 
G aux
Based on the general mathematical model presented above, a new TRNSYS type was developed in C++ to
simulate the performance of each chiller. The model determines one of the following parameters: (i) the hot water
temperature required to satisfy a given cooling load assuming that the chilled water supply temperature and the
cooling water inlet temperature are known or (ii) the chilled water supply temperature based on the maximum
allowable hot water temperature and a known cooling water inlet temperature for part-load conditions, ensuring that
the hot water temperature never exceeds the maximum allowable limit. The model normally works in the first mode
unless the chiller cannot satisfy the whole cooling demand at the specified chilled water set-point temperature. The
procedure allows the chiller to work at either part or full load condition while the mass flow rates in the external
circuits remain constant. For the purpose of this study, a single- and double-effect chiller were selected from Broad
Inc. [32], while the triple-effect chiller was selected from Thermax Ltd [33]. The technical specifications of these
chillers (at rated conditions) are listed in Table 2.
850 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

Table 2. Technical specifications of the absorption chillers used in this study

Parameter Unit Single-effect chiller [32] Double-effect chiller [32] Triple-effect chiller [33]
Chilled water temperature
°C 14/7 14/7 12/7
(inlet/outlet)
Cooling water temperature
°C 30/37 30/37 30/37
(inlet/outlet)
Hot water temperature
°C 98/88 180/165 210/195
(inlet/outlet)
3
Chilled water mass flow rate m /hr 125 143 200
Cooling water mass flow rate m3/hr 291 244 222
Hot water mass flow rate m3/hr 116 51 37
Cooling capacity kW 1023 1163 1163
Pump electricity consumption kW 5.7 6.8 6.55
COP - 0.76 1.41 1.81

3.6. Cooling tower

In this study, an open circuit cooling tower was used to reject heat from the chiller to the ambient. TRNSYS
‘Type51b’ was used to simulate the performance of the cooling tower with a sump volume of 0.2 m3. This model
requires manufacturer specific parameters c and n for correlating the number of transfer units (NTU) of the cooling
tower as follows:
1 n
§m ·
NTU c¨ w ¸ (12)
m
© a ¹
These parameters were obtained by curve fitting data provided by Evapco Australia [34], with c 1.62 and
n 0.62 .

4. Results and discussion

A parametric study was carried out by varying the DNI fraction under three global irradiance categories (low,
medium and high in GHI) in order to compare the annual performance of the proposed configurations. A scaling
factor was used in the TRNSYS model to vary the DNI fraction from 0% to 100% at a given GHI value which is
taken from the corresponding weather data file. In order to clearly distinguish between the effect of COP of the
single-effect and the multi-effect chillers on the size of the solar field, it was assumed that the chillers operate at a
high, fixed cooling capacity (1000 kWc) from 7am to 7 pm every day. This extreme approach ensures that each
chiller runs at its highest COP (0.76 vs 1.41 vs 1.81) as opposed to partial load conditions, thereby requiring a
minimum heat input to supply the imposed constant cooling load. This approach also enables us to solely investigate
the influence of variation of GHI and DNI fraction on the size of the solar field irrespective of cooling demand
pattern. Therefore, a set of cut-off points can be determined, showing that under what climate conditions the
combination of multi-effect chillers and concentrating collectors can out-perform the typical solar single-effect
chiller. The performance of each configuration is evaluated by two key performance indicators: a) annual solar-
derived cooling output per square meter of collector area and b) the solar field capital cost ratio. The latter is defined
as the cost of the solar field per kWh cooling from solar thermal ($/kWhc) for the solar-assisted double/triple effect
chiller divided by that of the solar single-effect chiller. Since the economics of solar absorption systems is dominated
by the cost of solar field [35], only the capital cost of solar collectors has been taken into account for the cost
analysis of each configuration. Therefore, it was assumed that the difference between the capital costs of the chillers
is marginal relative to the cost associated with solar collectors. The capital costs of the ETC and PTC units were
assumed to be 470 and 520 AUD per unit aperture area of the collector [26, 27]. It should be mentioned that the
storage tank capacity for the proposed solar single- and multi-effect chillers were assumed to be 70 and 50 L/m2 (of
the collector aperture area), respectively. These values were obtained by a separate parametric study where the
variation of solar fraction with respect to the tank volume was analyzed.
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 851

4.1. Model validation

The thermal efficiency curves of the ETC and PTC modules obtained from simulation results and those reported
in the collector test reports [26, 27] are shown in Fig. 2. In this comparison, the relative difference is less than 2%,
which is considered an acceptable value.
1
ETC (Simulation)
ETC(Test Report)
0,8

Thermal Efficiency (-)

PTC (Simulation)
PTC (Test Report)
0,6

0,4
Gb= 850 W/m2
0,2 Gd= 150 W/m2
Ta=25 °C
0
0 50 100 150 200 250
Tavg-Ta (°C)

Fig. 2. Thermal efficiency curves of the solar collectors obtained from simulation results against experimental data reported by the collectors’
manufacturer [26, 27]

1250 1500
Performance Data Cooling Capacity
Performance Data Cooling Capacity

(a) Single-effect chiller (b) Double-effect chiller

1150 1400

1050 1300

950 1200
(kW)
(kW)

850 1100

750 1000

650 ±10 Deviation 900 ±10 Deviation

±5 Deviation ±5 Deviation
550 800
550 650 750 850 950 1050 1150 1250 800 900 1000 1100 1200 1300 1400 1500
Fitted Cooling Capacity (kW) Fitted Cooling Capacity (kW)
1800
Performance Data Cooling Capacity

(c) Triple-effect chiller

1600

1400
(kW)

1200

1000

800 ±10 Deviation

±5 Deviation
600
600 800 1000 1200 1400 1600 1800
Fitted Cooling Capacity (kW)

Fig. 3. Cooling capacity versus fitted cooling capacity for: (a) the Broad 1023 kW single-effect chiller [32]; (b) the Broad 1163 kW double-effect
chiller [32]; (c) the Thermax 1163 kW triple-effect chiller [33]
852 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

The cooling capacity of the single-, double-, and triple-effect absorption chillers from the manufacturer’s chart
[32, 33] are compared in Fig. 3 with the cooling capacity obtained from the simulation models of the chillers. As
shown in this figure, the data points from the simulation results are mostly within a ±5% deviation range, verifying
sufficient accuracy of the adapted characteristic equation method to predict the performance of these absorption
chiller units. The characteristic coefficients of the chillers’ model obtained from a multiple linear regression analysis
are presented in Table 3.

Table 3. Characteristic coefficients of the commercial absorption chillers characterized by the characteristic temperature function ( ''Tc )
method

Characteristic coefficient
Chiller type
a e sE rE sG rG RE2 R G2
Single-effect -1.45 1.75 68.12 -3281.1 88.05 -4216.1 0.93 0.93
Double-effect -2.46 4.38 18.1 -1350.5 12.54 -917.3 0.94 0.93
Triple-effect -2.14 3.29 14.6 -1171.7 7.05 -504.2 0.98 0.95

4.2. Parametric study

The annual cooling output from solar energy per unit collector area and the solar field capital cost ratio are shown
in Fig. 4 against DNI fraction for the three solar absorption system configurations proposed in this study under low,
medium and high GHI levels, respectively. As seen in this figure, regardless of GHI level, the solar cooling achieved
by the combination of single-effect chiller and ETC modules is not sensitive to the DNI fraction. In other words, the
ratio of the DNI component has little effect on the size of solar field in such systems, for a given global solar
irradiance. On the contrary, variation of DNI fraction significantly influences the annual solar cooling yield per unit
of collector produced by double- and triple-effect absorption chillers coupled with the PTC modules. This reflects
the fact that PTCs can only harness solar beam radiation, while non-concentrating collectors like ETCs are capable
of using diffuse radiation in addition to beam radiation.
The results also show that the evacuated tube powered single-effect chiller outperforms the concentrating
collector powered double- and triple-effect chillers at low DNI values due to the poor optical performance of PTC
modules when DNI fraction is low, resulting in much more collector areas required to supply a given amount of
cooling as compared to the solar single-effect chiller. Therefore, there is no advantage in using concentrating
collector powered multi-effect absorption chillers when the DNI fraction is less than 60% (at any GHI level) of the
total solar global irradiance. On the other hand, increasing DNI fraction beyond 60% leads to a significant decrease
in the size of solar field for both double- and triple-effect chillers compared to the single-effect chiller. This means
that multi-effect chillers are promising only in clear climates with relatively high levels of solar beam radiation. The
results also show that as the DNI fraction increases, the performance of the solar triple-effect chiller system becomes
more energy-efficient compared to the solar double-effect chiller. In fact, at medium and relatively high DNI values,
the higher COP of the triple-effect chiller relative to the double-effect one leads to less collector area required per
kW of cooling.
The results of the cost analysis, presented in Fig. 4, demonstrate that a minimum DNI fraction of about 70%,
which is relatively rare in many regions in the world, is required in order that the solar multi-effect chillers can be
cost competitive compared to solar single-effect chillers. Although the higher COP of multi-effect chiller systems in
high DNI regions (e.g. ~60% <DNI< ~70%) results in less collector area per kW of cooling, the high capital cost of
the PTC modules makes the total cost of the solar field higher than that of evacuated tube collectors used in the
single effect chiller. As PTC modules are simply evacuated tubes equipped with a reflector and tracking system, the
challenge for solar industry is to reduce the price of high temperature concentrating collectors to pave the way for
high-COP absorption chillers, especially in regions with relatively high DNI fractions.
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 853

800 12
1-e Chiller (a) Low GHI
700 2-e Chiller (~1000 kWh/m2)
3-e Chiller 10
600

Solar Field Cost Ratio (-)

Cost Ratio 2-e/1-e Chiller

QC,Solar/ASC (kWh/m2)
Cost Ratio 3-e/1-e Chiller 8
500 Cost Ratio 3-e/2-e Chiller

400 6

300
DNI § 62% 4
200
2
100 DNI § 70%

0 0
30 40 50 60 70 80 90 100
DNI Fraction (%)
1400 14
1-e Chiller (b) Medium GHI
1200 2-e Chiller (~1600 kWh/m2) 12
3-e Chiller

Solar Field Cost Ratio (-)

Cost Ratio 2-e/1-e Chiller
QC,Solar/ASC (kWh/m2)

1000 10
Cost Ratio 3-e/1-e Chiller
Cost Ratio 3-e/2-e Chiller
800 8

600 6

400 4

DNI § 61%
200 DNI § 70% 2

0 0
30 40 50 60 70 80 90 100
DNI Fraction (%)
2000 10
1-e Chiller
2-e Chiller
3-e Chiller
1600 Cost Ratio 2-e/1-e Chiller 8
Solar Field Cost Ratio (-)

Cost Ratio 3-e/1-e Chiller

QC,Solar/ASC (kWh/m2)

Cost Ratio 3-e/2-e Chiller

1200 6
(c) High GHI
(~2300 kWh/m2)

800 4

DNI § 60-65%
400 2
DNI § 70%

0 0
30 40 50 60 70 80 90 100
DNI Fraction (%)

Fig. 4. Annual cooling output from solar energy per collector area and the solar field capital cost ratio against DNI fraction for the solar-powered
single-, double-, and triple-effect absorption chillers under (a) low GHI level (~1000 kWh/m2); (b) medium GHI level (~1600 kWh/m2); (c) high
GHI level (~2300 kWh/m2)
854 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

4.3. Sensitivity analysis

A sensitivity analysis was conducted to determine the influence of the capital cost of PTCs on the economic
potential of solar powered double- and triple-effect absorption chiller systems particularly in climates where DNI
fraction is relatively high. Fig. 5 shows the sensitivity of the ratio of PTC to ETC solar field costs with variation of
the capital cost of PTC modules at low, medium, and high GHI levels, respectively. The results suggest that the
capital cost of the PTC modules needs to be decreased by 20% for solar-powered multi-effect absorption chillers to
be economically feasible in climates with DNI fractions greater than 60% as compared to the solar-assisted single-
effect chiller.
4 4
2-e chiller: 10% cost (b) Medium GHI 2-e chiller: 10% cost
reduction of PTC reduction of PTC
Solar Field Cost Ratio (-)

(~1600 kWh/m2)

Solar Field Cost Ratio (-)

3-e chiller: 10% cost 3-e chiller: 10% cost
3 3
reduction of PTC reduction of PTC
2-e chiller: 20% cost 2-e chiller: 20% cost
reduction of PTC reduction of PTC
2 3-e chiller: 20% cost 2 3-e chiller: 20% cost
reduction of PTC reduction of PTC
(a) Low GHI
1 1
(~1000 kWh/m2)
DNI § 60% DNI § 60%
0 0
40 50 60 70 80 90 100 40 50 60 70 80 90 100
DNI Fraction (%) DNI Fraction (%)
4
(c) High GHI 2-e chiller: 10% cost
Solar Field Cost Ratio (-)

(~2300 kWh/m2) reduction of PTC

3 3-e chiller: 10% cost
reduction of PTC
2-e chiller: 20% cost
2 reduction of PTC
3-e chiller: 20% cost
reduction of PTC
1

DNI § 60%
0
40 50 60 70 80 90 100
DNI Fraction (%)

Fig. 5. The sensitivity of PTC to ETC solar field cost ratio with variation in the capital cost of PTC modules at (a) low GHI level (~1000
kWh/m2); (b) medium GHI level (~1600 kWh/m2); (c) high GHI level (~2300 kWh/m2)

5. Conclusions

In the present work, the techno-economic feasibility of solar-powered multi-effect absorption chillers using
parabolic trough collectors was investigated under different climate conditions. Three configurations of solar-
assisted absorption chillers were considered in detail. In the first configuration, a single-effect absorption chiller was
coupled with evacuated tube collectors (ETCs), while parabolic trough collectors (PTCs) were employed to drive
double- and triple-effect chillers in the second and third configurations, respectively. The simulation results showed
that the size of the solar field for the multi-effect chiller is much more sensitive to the DNI fraction than that of the
single-effect chiller. The results also suggested that there is no advantage in using solar multi-effect absorption
chillers over the single-effect chiller when the DNI fraction is less than 60% of the total solar global irradiance. In
fact, even with their higher COPs, multi-effect absorption chillers still require much more collector area than single-
effect chillers (at low DNI fractions) to supply a given amount of cooling. On the other hand, if the DNI fraction is
above 60%, the size of the solar field required for both double- and triple-effect chillers is significantly reduced
compared to the single-effect chiller. This means that multi-effect chillers using concentrating collectors can work
only in the climates with high solar beam radiation (>~60%).
 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856 855

The cost analysis of the solar field showed that a minimum DNI fraction of about 70% is required in order for
solar multi-effect chillers to be cost completive compared to the solar single-effect chillers, while such high DNI
fractions are not easy to find in many regions in the world. Although the higher COP of multi-effect chillers in
relatively high DNI regions (e.g. ~60% <DNI< ~70%) results in less collector area per kW of cooling, the high
capital cost of parabolic trough concentrating collectors makes the total cost of solar field higher than that of
evacuated tube collectors for a single effect system. Finally, the results of sensitivity analysis showed that the cost of
parabolic trough collectors needs to be decreased by 20% for solar powered multi-effect chillers to be more cost
effective than single-effect chiller systems in high DNI climates.

Acknowledgements

This research is funded by The CRC for Low Carbon Living Ltd (Project ID (RP1002)) supported by the
Cooperative Research Centers program, an Australian Government initiative.

References

[1] Pérez-Lombard L, Ortiz J, Pout C. A review on buildings energy consumption information. Energy and Buildings 2008;40(3):394-8.
[2] Hitchin R, Pout C, Butler D. Realisable 10-year reductions in European energy consumption for air conditioning. Energy and Buildings
2015;86:478-91.
[3] Prasartkaew B, Kumar S. Design of a renewable energy based air-conditioning system. Energy and Buildings 2014;68, Part A:156-64.
[4] Yong C, Yiping W, Li Z. Performance analysis on a building-integrated solar heating and cooling panel. Renewable Energy 2015;74:627-32.
[5] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by CPVT and ET solar collectors: A novel transient simulation
model. Applied Energy 2013;103:588-606.
[6] Otanicar T, Taylor RA, Phelan PE. Prospects for solar cooling – An economic and environmental assessment. Solar Energy 2012;86(5):1287-
99.
[7] Zhai XQ, Qu M, Li Y, Wang RZ. A review for research and new design options of solar absorption cooling systems. Renewable and
Sustainable Energy Reviews 2011;15(9):4416-23.
[8] Gomri R. Investigation of the potential of application of single effect and multiple effect absorption cooling systems. Energy Conversion and
Management 2010;51(8):1629-36.
[9] Calise F, Palombo A, Vanoli L. Design and dynamic simulation of a novel polygeneration system fed by vegetable oil and by solar energy.
Energy Conversion and Management 2012;60:204-13.
[10] Tierney MJ. Options for solar-assisted refrigeration—Trough collectors and double-effect chillers. Renewable Energy 2007;32(2):183-99.
[11] Ayadi O, Aprile M, Motta M. Solar Cooling Systems Utilizing Concentrating Solar Collectors - An Overview. Energy Procedia
2012;30:875-83.
[12] Liu YL, Wang RZ. Performance prediction of a solar/gas driving double effect LiBr–H2O absorption system. Renewable Energy
2004;29(10):1677-95.
[13] Calise F. High temperature solar heating and cooling systems for different Mediterranean climates: Dynamic simulation and economic
assessment. Applied Thermal Engineering 2012;32:108-24.
[14] Tsoutsos T, Aloumpi E, Gkouskos Z, Karagiorgas M. Design of a solar absorption cooling system in a Greek hospital. Energy and Buildings
2010;42(2):265-72.
[15] Baniyounes AM, Rasul MG, Khan MMK. Assessment of solar assisted air conditioning in Central Queensland's subtropical climate,
Australia. Renewable Energy 2013;50:334-41.
[16] Arsalis A, Alexandrou AN. Parametric study and cost analysis of a solar-heating-and-cooling system for detached single-family households
in hot climates. Solar Energy 2015;117:59-73.
[17] Eicker U, Pietruschka D, Haag M, Schmitt A. Systematic design and analysis of solar thermal cooling systems in different climates.
Renewable Energy 2015;80:827-36.
[18] Osborne JP, Kohlenbach P, Jakob U, Dreyer J, Kim J. The Design and Installation of a Combined Concentrating Power Station, Solar
Cooling System and Domestic Hot Water System. Energy Procedia 2015;70:486-94.
[19] Qu M, Yin H, Archer DH. A solar thermal cooling and heating system for a building: Experimental and model based performance analysis
and design. Solar Energy 2010;84(2):166-82.
[20] Li Z, Ye X, Liu J. Optimal temperature of collector for solar double effect LiBr/H2O absorption cooling system in subtropical city based on
a year round meteorological data. Applied Thermal Engineering 2014;69(1–2):19-28.
[21] Shirazi A, Taylor RA, White SD, Morrison GL. A systematic parametric study and feasibility assessment of solar-assisted single-effect,
double-effect, and triple-effect absorption chillers for heating and cooling applications, Energy Conversion and Management (2016), DOI:
10.1016/j.enconman.2016.01.070.
[22] Agrawal SK, Kumar R, Khaliq A. First and second law investigations of a new solar-assisted thermodynamic cycle for triple effect
refrigeration. International Journal of Energy Research 2014;38(2):162-73.
856 Ali Shirazi et al. / Energy Procedia 91 (2016) 844 – 856

[23] TRNSYS 17: A Transient System Simulation Program, Solar Energy Laboratory, University of Wisconsin, Madison, USA.
[24] Shirazi A, Pintaldi S, White SD, Morrison GL, Rosengarten G, Taylor RA. Solar-assisted absorption air-conditioning systems in buildings:
Control strategies and operational modes. Applied Thermal Engineering 2016;92:246-60.
[25] Duffie JA, Beckman WA. Solar Engineering of Thermal Processes. 3rd ed. Hoboken, NJ, USA: Wiley, 2006.
[26] Evacuated Tube Solar Hot Water Systems - Apricus Australia, URL: http://www.apricus.com.au; 2014 [accessed December 2014].
[27] NEP Solar Pty Ltd, URL: http://www.nep-solar.com/; 2015 [accessed July 2015].
[28] TRNSYS 17 TESS Library, Component Libraries for the TRNSYS Simulation Environment: Volume 11 Storage Tank Library
Mathematical Reference.
[29] Han YM, Wang RZ, Dai YJ. Thermal stratification within the water tank. Renewable and Sustainable Energy Reviews 2009;13(5):1014-26.
[30] TRNSYS 17, Component Libraries for the TRNSYS Simulation Environment: Volume 4 Mathematical Reference.
[31] Kühn A, Ziegler F. Operational results of a 10 kW absorption chiller and adaptation of the characteristic equation. Proceedings of the
International Conference Solar Air Conditioning, Bad Staffelstein, Germany. 2005;6-7.
[32] BROAD X Non-electric Chiller: Model Selection & Design Manual, URL: http://www.broadusa.com; 2008 [accessed July 2015].
[33] Thermax Ltd, URL: http://www.thermaxglobal.com/; 2015 [accessed July 2015].
[34] EVAPCO Australia Pty Ltd, Counter-flow MSS Series Cooling Tower Performance Data, URL: http://www.evapco.com.au/; 2015 [accessed
July 2015].
[35] Shirazi A, Taylor RA, White SD, Morrison GL. Transient simulation and parametric study of solar-assisted heating and cooling absorption
systems: An energetic, economic and environmental (3E) assessment. Renewable Energy 2016;86:955-71.