Вы находитесь на странице: 1из 23

Energy Conversion and Management 171 (2018) 59–81

Contents lists available at ScienceDirect

Energy Conversion and Management


journal homepage: www.elsevier.com/locate/enconman

Review

Solar-powered absorption chillers: A comprehensive and critical review T


a,b,⁎ a a c
Ali Shirazi , Robert A. Taylor , Graham L. Morrison , Stephen D. White
a
School of Mechanical and Manufacturing Engineering, The University of New South Wales (UNSW), Kensington, New South Wales 2052, Australia
b
Research School of Engineering, The Australian National University, Canberra, Australian Capital Territory 2601, Australia
c
Commonwealth Scientific and Industrial Research Organization (CSIRO) Energy Centre, Newcastle, New South Wales 2304, Australia

A R T I C LE I N FO A B S T R A C T

Keywords: Solar heating and cooling (SHC) systems are currently under rapid development and deployment due to their
Solar cooling potential to reduce fossil fuel use and to alleviate greenhouse gas emissions in the building sector – a sector
Solar collectors which is responsible for ∼40% of the world energy use. The available technologies on the market for thermally
Absorption chiller driven cooling systems are absorption and adsorption chillers, solid and liquid desiccant cooling systems, and
Air-conditioning
ejector refrigeration cycles. Of these, absorption chillers are considered as the most desirable method for har-
Refrigeration
nessing solar thermal energy due to their relative maturity, reliability, and higher efficiency. In addition, ab-
sorption chillers can take advantage of economies of scale in large buildings to obtain a relatively good levelized
cost of cooling as compared to other thermally-driven air-conditioning systems. In this paper, the background
theory on solar-powered absorption chillers is presented followed by a comprehensive literature review of the
recent existing theoretical and experimental investigations on this technology is conducted. The review shows
that the majority of solar absorption chillers installed and much of the research around the world is based on
single-effect chillers and low-temperature solar thermal collectors, while less emphasis has been placed on the
combination of high-temperature solar thermal collectors and multi-effect absorption chillers, especially triple-
effect chillers. Research studies indicate the use of gas-fired backup systems for single-effect chillers is inefficient
due to its very low primary energy savings. It was also found that the storage tank and piping can be major
sources of heat losses in solar absorption cooling systems. Thus, special care should be taken to ensure sufficient
and appropriate insulation for all heat loss components. In regions with low direct normal incidence solar re-
sources (e.g. most of Europe), solar multi-effect chillers are relatively inefficient, so single-effect chiller-based
solar cooling systems are the best techno-economic choice in such regions. Conversely, multi-effect absorption
chillers with high-temperature collectors are indeed promising in regions with high solar resources. However,
the review shows that using currently available technology, SHC absorption chillers are not able to economically
compete with conventional cooling without government subsidies and incentives. Therefore, improving the
economic performance of these systems is essential. While there is clearly more that can be done on chiller and
solar collector components themselves, we believe some R&D emphasis going forward should also be dedicated
to the balance of the system, including optimization of the system configuration, minimizing parasitic losses,
improved design and integration of thermal storage and auxiliary system, and numerous controls and opera-
tional aspects. To date, many of these topics have been largely overlooked in favor of chiller performance
studies.

1. Introduction typically harvested in collectors that operate at temperatures well


below 100 °C. The average global annual solar energy resource poten-
Solar energy − a vast, renewable and relatively untapped re- tial is around 1.6 MWh m−2, greatly exceeding the total average energy
source − is freely and continually delivered to the rooftops of our demand per unit area [1]. With increased environmental concern over
global building stock. The total global solar irradiance striking the fossil fuel consumption as well as government policies encouraging the
Earth's surface consists of two components: (i) direct beam and (ii) use of solar energy, the global solar industry is rapidly accelerating [2].
diffuse radiation. The beam component can be concentrated and con- The rapid fall in the cost of PV modules (in $/W) has dramatically in-
verted to heat at high temperature whereas the diffuse radiation is creased the use of PV systems for electricity generation over the past


Corresponding author at: School of Mechanical and Manufacturing Engineering, The University of New South Wales (UNSW), Kensington, New South Wales 2052, Australia.
E-mail address: a.shirazi@unsw.edu.au (A. Shirazi).

https://doi.org/10.1016/j.enconman.2018.05.091
Received 29 January 2018; Received in revised form 3 May 2018; Accepted 25 May 2018
0196-8904/ © 2018 Elsevier Ltd. All rights reserved.
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

sunny hours [8]. Without batteries, a separate storage system (i.e. hot/
cold water storage) may be required to cover the mismatch between
available solar electricity gains and building loads. It should be noted
that for a high COP system, 4–6 times as much thermal storage capacity
is required to meet the capacity of an electrical storage system (e.g. for
a COPe ≈ 4–6) [9]. Investment costs for solar-driven vapor compression
chillers could be high due to the need for an additional electrical heat
pump to cover the building heating demand in the winter. According to
recent studies, the total cooling cost for solar PV cooling systems is not
competitive without a feed-in-tariff for the PV-derived electricity sur-
plus [10]. Currently, the average conversion efficiency of most com-
mercial PV modules is relatively low (10–15%) [11]. This significantly
reduces the overall sun-to-cooling efficiency of PV-driven systems (e.g.
Fig. 1. Estimation of the number of solar cooling systems installed across
10–15% × 4–6 = 40–90%). Although this technology may indeed be-
Europe and the world [6].
come viable in the future, conventional compression chillers driven by
solar PV panels is well-established [12–14], and will not be considered
few years. However, low efficiency levels and the high price of battery in the present dissertation.
storage are the main barriers to stabilizing of this technology [3]. Solar As shown in Fig. 2 solar thermal cooling technologies span a wider
thermal technologies continue to grow (albeit at more modest rates) for range of potential options and, consequently, are less well-understood.
applications such as power generation, industrial processes, domestic Solar thermal collectors convert solar energy into thermal heat which
hot water, and space heating and cooling. Although solar PV has gar- can be used to run a thermally-activated cooling device, thereby gen-
nered more public recognition, solar thermal technology still represents erating chilled water or conditioned air for use in buildings. Thermally-
a significant part of the global installed capacity of solar energy [4]. driven cooling systems, compared to electrical vapor compression
chillers, have a lower COP (0.6–1.8) but higher collector efficiency
2. Solar cooling overview (35–70%). Since the chiller’s COP and the collector efficiency are in-
versely related, total system efficiencies fall between 35 and 80%. This
The aim of a solar cooling system is to utilize the solar energy falls roughly within the same range as PV-driven systems. Thermally
landing on a building for useful space-conditioning for the occupants driven cooling systems can also be integrated with thermal storage (at a
within. This is widely considered to be a sustainable and en- lower cost than batteries) and can be designed for excellent annual
vironmentally-friendly alternative to conventional air-conditioning performance through dual design to meet the building’s heating needs.
systems [5], and, as such, interest in solar air-conditioning has grown Thermal storage, in particular, is seen as a big benefit of these systems
steadily over the last 10 years. A relatively recent survey, as shown in since (relative to conventional HVAC) it alleviates peak cooling and
Fig. 1, has estimated the number of solar cooling installations world- heating loads on the grid [15]. Another advantage of thermal-driven
wide to be ∼1200 systems in 2014 [6]. systems is that by removing the compressor, they are characterized by
Solar energy can be used to produce a cooling effect via either low-vibration and low noise operation. To date, the vast majority of
electricity-driven or thermally-driven cooling processes [7]. Fig. 2 illus- existing solar air-conditioning systems are driven by solar thermal heat
trates a classification of main solar cooling technologies. The most [6,16]. In regions that require both cooling and heating throughout a
common solar electricity-driven cooling technology is based on driving year, these systems represent a year-round solution, improving the
high COP vapor compression chillers (COPe ≈ 4–6) connected to solar system efficiency and economics as compared to those producing either
PV modules. This technology is relatively simple as it has low main- cold or heat alone. Solar thermal cooling systems are less likely to be
tenance and is suitable for small-scale applications [5]. Although the taken up at residential scales (5–15 kWc) due to their significantly
price of solar PV cells has dramatically decreased in recent years, the higher price tag compared to conventional grid-connected split systems
high cost of battery storage has limited PV-driven cooling production to [17]. For large-scale applications (> 50 kWc), however, economies of
scale can make larger units more financially viable [18].
As shown in Fig. 2, the available technologies on the market for
thermally-driven cooling systems are desiccant cooling systems, ejector
refrigeration cycles, adsorption and absorption chillers [19]. A de-
siccant cooling cycle utilizes liquid or solid desiccant material to absorb
water from an incoming air stream – using thermal heat as the driving
source [20]. Water is then sprayed into the resulting dehumidified air
stream, thereby lowering its temperature and providing a cooling effect
(evaporative cooling). While there are relatively few suppliers of these
systems, desiccant cooling systems have been used extensively in cer-
tain niche applications (e.g. supermarkets and hot and humid climates)
[21], where the ability to independently control air humidity provides
additional benefits. An ejector cooling cycle is basically the same as a
conventional vapor compression chiller, but uses an ejector – a thermal
compressor consisting of a supersonic nozzle, mixing chamber, and
diffuser − to compress a refrigerant instead of an electrically-driven
mechanical compressor [22]. The ejector requires thermal heat as a
driving source in order to increase the pressure of the refrigerant. To
date, this technology has not been widely used due to its relatively low
efficiency [23]. An adsorption cooling cycle is based on the phenom-
enon of physical adsorption between the refrigerant vapor and a solid
adsorbent to achieve a cooling effect. When heated, the solid absorber
Fig. 2. Classification of main solar cooling technologies. desorbs the vapor and pressurizes the vessel in which the vapor is

60
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

contained [24]. This, in effect, creates a thermal compressor that re- higher than the low-pressure level, allowing the heat rejection from the
places a conventional mechanical compressor. While there are only a refrigerant to occur at commonly useable temperatures. The operating
limited number of adsorption chiller manufacturers [24], adsorption principle of this cycle is briefly explained as follows:
chiller technology is able to operate well with a low-temperature heat The working principle of the cycle is simply based on different
source and is more suitable for operation with a dry cooling tower. boiling temperatures of the refrigerant and absorbent. First, low-grade
However, they are generally heavy, bulky and relatively expensive, heat from a conditioned space (e.g. a building) is transferred to the
especially in large-scale ranges [25,26]. evaporator through a chilled water loop, causing the refrigerant to
The last type of thermally-driven cooling systems is the absorption evaporate. The low-pressure refrigerant vapor is then passed into the
cooling cycle which uses a liquid absorbent to absorb the refrigerant absorber, where it is absorbed by a concentrated absorbent liquid and
vapor in order to thermally compress the refrigerant and therefore forms a weak solution. The heat released during this process is removed
create a cooling effect. Having been commercially available for many by the cooling water stream coming from a cooling tower. The diluted
years, absorption chillers are mature, reliable and more energy efficient solution of absorbent and refrigerant is pumped to the generator, where
compared to other thermal cooling processes, meaning less thermal an external heat source is utilized to boil off the refrigerant from the
heat is required to achieve a given amount of cooling [27,28]. In ad- solution while the concentrated absorbent stays in the liquid. The rich
dition, absorption chillers are available for large-scale applications and solution then passes through an expansion valve and returns to the
their cost is lower than the rest of thermally-activated cooling systems absorber. The regenerated refrigerant vapor flows to the condenser
for such uses [28,29]. Thus, absorption chiller technology is considered where heat is rejected to the circulating cooling water loop, condensing
as the most desirable method for solar thermal cooling and expected to the refrigerant vapor to liquid. The high-pressure liquid is then passed
compete at scale with conventional air-conditioning systems in the through an expansion valve, reducing its pressure and temperature to
foreseeable future. However, solar-driven absorption chillers are not yet the evaporator levels and completing the cycle. The only mechanical
widely commercially available, mainly due to their relatively high in- work input to the cycle is the work required to run the solution pump,
itial investment of solar field, system complexity and the low number of which is significantly less than that of a vapor compressor.
demonstration plants [30]. Thus, more research is required in order to
evaluate the energy efficiency, reliability and cost-competitiveness of
3.2. Crystallization
this technology compared to typical HVAC systems. More details on the
performance of solar absorption chillers are presented in the following
Crystallization can occur in LiBr-H2O chillers when the concentra-
section.
tion of the hot solution, rich in LiBr salt becomes too high or, alter-
natively, if the solution is cooled down to too low of temperatures. In
3. Solar cooling via absorption chillers the crystallization region, a two-phase mixture of LiBr-H2O solution and
LiBr crystals exist in equilibrium. This phenomenon is most likely to
3.1. Absorption cooling cycle take place in the solution heat exchanger, and can block pipes and
valves and can even lead to dry operation of the solution pump due to
Absorption cooling technology has been widely used for air-con- an empty absorber sump. If this occurs, the temperature of the con-
ditioning applications, especially in large-scale buildings [31]. The centrated solution needs to be increased above its saturation point so
working principle of an absorption cycle is similar to that of a vapor that the salt crystals re-dissolve [32]. The process of recovering the
compression cycle with two major differences. The first difference is absorber operation after crystallization is labor intensive and time-
that unlike electrically-driven mechanical vapor compression chillers, consuming [33]. Thus, it is essential for LiBr-H2O chillers to operate
the absorption cycle is a heat-driven thermal system with little me- well away from the crystallization region. Crystallization of the LiBr-
chanical energy input for liquid pumps [31]. The second difference is H2O solution is more problematic in air-cooled chillers since the air-
the existence of a secondary fluid in addition to the working refrigerant, cooled absorber tends to operate at a higher temperature and con-
which is known as absorbent. The absorbent solution is used to absorb centration level than a water-cooled absorber due to the relative heat
the refrigerant vapor, allowing its pressure to be economically in- transfer characteristics of the coolant [33]. To illustrate the crystal-
creased by a liquid pump, rather than a vapor compressor that requires lization region, phase boundaries are usually included on the working
much more mechanical work input. Fig. 3 illustrates a basic absorption fluid pressure-temperature-mass fraction diagram, as presented in the
cooling cycle. The main components of the cycle are evaporator, ab- LiBr-H2O Duhring chart in Fig. 4 [31]. As shown in this Fig, in theory,
sorber, solution pump, generator, condenser and expansion valves. Si- the mass fraction of LiBr should be kept under about ∼70% so that
milar to a vapor compression cycle, the absorption cycle operates at two
pressure levels. The low-pressure level corresponds to the evaporator
and absorber, while the condenser and generator operate at the high-
pressure level. The high-pressure level is approximately ten times

Fig. 3. Schematic diagram of a basic absorption cooling cycle. Fig. 4. Duhring chart for LiBr-H2O solution with crystallization line [31].

61
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

LiBr-H2O solution can remain miscible, while this value should be much • The temperature difference between the boiling points of the re-
lower in practice (e.g. 50%) [34]. frigerant and the absorber should be high.
The three most frequent causes of crystallization are [35]: • The refrigerant should be volatile, allowing it to be easily separated
from the absorbent inside the generator.
• Air leakage into the machine, which results in an increase in the
evaporator pressure level. Non-absorbable gases such as hydrogen, Among numerous working fluid pairs used in absorption cycles, li-
produced during corrosion, which can also reduce the performance thium bromide-water (LiBr-H2O) and water-ammonia (H2O-NH3) are
of the chiller absorber and condenser. Therefore, a vacuum pump is the two most common pairs [36]. Other pairs have also been reported
used to maintain the vacuum inside the machine and eliminate in the literature such as NH3-LiNO3 [37], LiBr-HO (CH2) OH [38] and
unwanted gases. others [36,39], but they are not competitive as much, mainly because of
• If the cooling water temperature becomes too low especially at high their poor performance. LiBr-H2O is mainly used for air-conditioning
cooling load conditions, it can cause crystallization. applications, whereas H2O-NH3 is the most commonly used working
• Electric power failure can also lead to crystallization inside the fluid for refrigeration purposes due to the capability of ammonia to
chiller. During normal shutdowns, the machine undergoes a dilution achieve sub-zero temperatures. The H2O-NH3-based absorption ma-
cycle, thereby lowering the concentration level of LiBr-H2O solution chines generally require higher temperatures at their generators and
in the chiller. If the machine is suddenly stopped during the full load have lower COP levels than those using LiBr-H2O as working fluid [40].
operation, when the highly concentrated solution is present in the Since the difference between the boiling temperature of ammonia and
solution heat exchanger, crystallization is definitely a big risk. water is not very high, H2O-NH3 absorption chillers also need a deph-
legmator to separate ammonia and water vapor inside the generator,
To prevent crystallization, the hot and cooling water temperatures which adds to the complexity and cost of the system. In LiBr-H2O
need to be measured and controlled within the certain range re- chillers, crystallization may occur if the solution concentration is too
commended by the chiller’s manufacturer. Chemical additives such as high. To avoid this, an internal control system is normally installed in
2-Ethyl Hexanol can also be used to shift the crystallization line to the chiller to monitor the temperature at which heat rejection takes
higher temperatures. The limitations placed upon the operating tem- place inside the absorber. As mentioned earlier, crystallization may
peratures at the generator, condenser and evaporator loop of the chiller occur in LiBr-H2O chillers, and to avoid this, an internal control system
as a result of preventative measures to avoid crystallization can impose is normally installed inside the chiller to monitor the temperature at
operational constraints on the performance of absorption chillers, which heat rejection takes place inside the absorber. Although the
which may limit or even prevent their utilization in extreme weather aforementioned working fluid pairs have been dominantly used in ab-
conditions. For example, in hot and humid climates, it may not be sorption chiller technologies, research into discovering new working
possible to maintain the cooling water temperature at the desired fluids still continues [41,42].
range, and therefore, the chiller should be switched off to prevent any
damage inside the machine. 3.3.3. Number of effects
Absorption chillers can also be categorized by the number of ‘ef-
3.3. Classification of absorption chillers fects’ they employ, a designation which refers to the number of times
heat is recycled inside the chiller to produce cooling. Currently, there
Absorption chillers can be classified based on three main criteria: are three common types of absorption chillers commercially available
the firing method, the working fluid pair, and the number of effects. on the market − single-, double-, and triple-effect chillers [31]. Moving
These are briefly described in the following sections. toward a higher effect, more complicated, cycle leads to a higher
coefficient of performance (COP), but requires higher driving tem-
3.3.1. Firing method peratures [43].
Depending on the type of driving heat input available, absorption
chillers can be classified into two groups as follows: 3.3.3.1. Single-effect absorption chiller. Single-effect chillers are the
simplest configuration of absorption chillers. As demonstrated in
• Direct-fired chillers in which the driving heat directly comes from Fig. 5a single-effect chiller consists of an evaporator (E), an absorber
the combustion of fossil fuels. As a result, this type of absorption (A), a solution heat exchanger (SHX), solution pump, a generator (G), a
chiller runs on a liquid fuel or natural gas. condenser (C) and two expansion valves. The main function of the
• Indirect-fired chillers in which the driving heat comes from another solution heat exchanger is to preheat the weak absorbent solution
source and is normally delivered to the chiller through an inter- before entering the generator by receiving heat from the strong
mediate element (i.e. a heat exchanger). This heat can be in the form absorbent solution returning from the generator, leading to an
of hot water, steam, or hot exhaust gas. enhanced COP of the chiller. Single-effect chillers operate in the
driving temperature range of 80 °C to 100 °C, achieving thermal COPs
Since solar-derived thermal heat is usually stored in water, indirect- of around 0.7–0.8 as shown in Fig. 6 [44].
fired absorption chillers are preferred in solar cooling applications.
3.3.3.2. Double-effect absorption chiller. Double-effect chillers employ
3.3.2. Working fluid pair an additional generator to produce more refrigerant vapor from a given
Another classification of absorption chillers is based on the type of heat input, and thus operate between three pressure levels − the low-
working fluid pair used in the chiller. The performance and efficiency of pressure level at the evaporator, the medium-pressure level at the low-
an absorption chiller is completely governed by thermo-physical temperature generator (LTG), and the high-pressure level at the high-
properties of the working fluid pair. Miscibility of the refrigerant and temperature generator (HTG), where the external heat input is used to
absorbent in liquid phase within the operating temperature range of the boil off the refrigerant from the absorbent solution [45]. As illustrated
absorption cycle is the most important requirement for a suitable in Fig. 7 Fig. 8 and Fig. 9 there are three design variations for a double-
working fluid pair. In addition, the following criteria should be taken effect absorption chiller (due to the preference of manufacturers) −
into account when selecting the working fluid pair [36]: parallel, series and reverse-series flow cycles [43].
The difference between these cycles is in the way the solution
• The working fluid pair should be chemically stable, non-toxic, non- stream is distributed between the two generators. In parallel flow, the
corrosive, and inexpensive. weak solution is divided between the low and high-temperature

62
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

QC QG

C G

SHX

E A

QE QA
Fig. 5. Schematic diagram of a single-effect absorption chiller.

Fig. 7. Schematic diagram of a parallel flow double-effect absorption chiller.

refrigerant vapor streams produced at the high- and medium-


Fig. 6. COP levels for LiBr-H2O single-, double-, and triple-effect absorption temperature generators are condensed at the low-temperature
chillers as a function of heat supply temperature [44]. condenser, thereby providing heat to the low-temperature generator.
The liquid refrigerant eventually flows to the evaporator and completes
generators, while in series flow the solution is directly pumped from the the cycle. Triple-effect chillers can produce cooling at COPs of around
absorber to the high-temperature generator. In the reverse-series flow 1.8, but require a 210–240 °C heat input (Fig. 6) which is difficult to
cycle, the solution is first partially concentrated in low-temperature achieve for solar thermal collectors [47]. To the author’s knowledge,
generator before being pumped to the high-temperature generator. In only two companies − Thermax Ltd and Kawasaki Thermal
all three designs, the hot refrigerant vapor leaving the high-temperature Engineering Ltd – currently manufacture triple-effect chillers (using
generator enters the high-temperature condenser (HTC), where the heat LiBr-H2O) in the world [48,49].
released during the condensation process is reused to drive the low- To summarize these options, Table 1 presents the main design
temperature generator. In practice, the high-temperature condenser characteristics of absorption chiller technologies available on the
and the low-temperature generator are incorporated into one heat market.
transfer device (i.e. a heat exchanger), where one side is the high-
temperature condenser and the other side acts as the low-temperature
generator. The refrigerant vapor then passes through the low-tem- 3.4. Heat rejection methods
perature condenser (LTC), joining another refrigerant stream coming
from the low-temperature generator. After passing through the expan- The choice of a heat rejection system is critical to the electricity
sion valve, the refrigerant then flows back to the evaporator, where the consumption of absorption chillers, and therefore it can significantly
cooling effect takes place in the same way of the single-effect cycle. As affect the energy efficiency performance of the chiller. In general, dif-
shown in Fig. 6 double-effect chillers can achieve higher COPs, up to ferent heat sinks (e.g. air, ground, and water) can be used to reject heat
1.4, but require significantly higher driving temperatures of around from the chiller. Ground sinks may be a good choice in some locations
180 °C [46]. It should be mentioned that all the commercially available because, for the most part, they change on slower, seasonal time-scales,
double-effect chillers use LiBr-H2O as working fluid (since H2O-NH3 whereas air and water heat sinks highly depend on the daily climatic
requires much higher pressures) [43]. variations (i.e. dry/wet bulb temperature and humidity). In the air-
conditioning industry, air heat sinks have dominated the market: in the
form of dry cooling and wet cooling towers [31], which are discussed in
3.3.3.3. Triple-effect absorption chiller. Triple-effect chillers, as the the following. Other methods of heat rejection such as adiabatic pre-
name implies, benefit from three cascading generators, utilizing the cooling of the air in dry coolers or hybrid cooling towers also exist, but
external heat input three times in order to produce refrigerant vapor. their installations are limited to areas with specific climate conditions
Fig. 10 shows the schematic diagram of a triple-effect chiller. The [28].

63
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Fig. 8. Schematic diagram of a series flow double-effect absorption chiller. Fig. 9. Schematic diagram of a reverse-series flow double-effect absorption
chiller.
3.4.1. Dry cooling
Dry heat rejection systems consist of finned heat exchangers, where stream through the evaporative cooling. During this process, a small
the cooling water stream returning from the chiller rejects heat to the portion of water evaporates into the air stream as it is drawn/blown
ambient air which is directed to the heat exchangers via fans [53]. The through the tower. As heat is transferred to the air flow, it increases its
performance of dry cooling systems is strongly dependent upon the dry temperature up to the saturation point (i.e. ∼100% relative humidity),
bulb temperature of the ambient air. Dry heat rejection systems cannot and is then discharged into the ambient. The evaporated (and any blow
cool down the cooling water flow below the ambient air dry bulb down) water must be continuously replaced by make-up water. The
temperature. In practice, a minimum approach temperature of 3 K is cooling water supply temperature will always be approximately 3 K
considered between the cooling water outlet temperature and the am- higher than the wet bulb temperature of the ambient air, which itself
bient dry bulb temperature [54]. depends on both the dry bulb temperature and the humidity of ambient
Dry cooling systems generally have higher up-front costs but lower air [6].
maintenance costs compared to wet cooling tower technology. Wet cooling towers have lower investment costs and are able to
Chemical water treatment costs are avoided, since the cooling water reject more heat with lower parasitic electricity consumption as com-
does not need to be in direct contact with air and there are no hygienic pared to dry coolers [58]. This is particularly important given that the
problems related to legionella or mold growth in dry cooling tech- cooling tower fan and the cooling water pump are normally responsible
nology. The main disadvantages of dry cooling systems compared to for the major part of the total parasitic power consumption of the chiller
wet cooling tower systems include the higher heat rejection tempera- [55]. In addition, wet cooling towers enable heat rejection to take place
tures, higher investment cost, and higher parasitic electricity con- at lower temperatures over a wider range of ambient conditions. This
sumption [55,56]. The performance of dry cooling systems declines as increases the cooling capacity and efficiency of the chiller, and also
the ambient temperature rises. This can result in a significant drop in enhances the annual solar cooling energy savings [59]. The main dis-
the effectiveness of heat transfer from the cooling water to the ambient advantages of wet cooling towers are hygienic problems, water con-
air, especially in hot climates. In addition, it can lead to high cooling sumption and higher maintenance costs. However, the economics and
water temperatures supplied to the chiller [54]. To mitigate this pro- performance of absorption chiller-based solar cooling installations is
blem, the incoming air to the heat exchanger can be cooled by spraying almost always improved by using wet cooling towers, particularly in the
water into the air flow (adiabatic evaporative pre-cooling). It should be larger size ranges [55,59]. For the aforementioned reasons, wet cooling
noted that this approach may not work well in extremely humid cli- towers are generally well suited to solar abortion cooling applications.
mates, as the humidity of the ambient air is already high [54,57]. In the case of extremely high relative humidity in areas, the evaporative
cooling in the cooling tower becomes ineffective, and thus the heat
rejection performance drops significantly. Therefore, it may be desir-
3.4.2. Wet cooling towers able to switch to a dry cooling system or to use water as a temporary
Wet cooling towers extract waste heat from the cooling water

64
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Cooling
Tower

Solar
Collector
Thermal Auxiliary
Chiller
Storage Heater

Solar Pump Cooling


Coils

Fig. 11. A generic absorption chiller-based solar cooling system.

years, demonstration projects have shown the potential to use solar


thermal energy to run these chillers [8]. As shown in Fig. 11 a typical
solar-driven absorption system consists of four basic components: a
solar thermal collector, an absorption chiller, an auxiliary heater, and a
storage tank.
The incident solar radiation absorbed by solar thermal collectors
increases the temperature of a storage medium (thermal storage)
through a heat transfer fluid circulated by a pump in the solar loop. The
absorption chiller then converts the collected solar-derived thermal
heat into useful cooling, which is delivered from the chiller as chilled
water into cooling coils to cover the building cooling load. The driving
heat to the chiller and the building thermal load are rejected through a
cooling water loop to the ambient. Since solar energy is not always
available (or in phase with building load), the plant is usually equipped
with an auxiliary heater and thermal storage unit, thereby decoupling
the intermittent availability of solar heat from the variable cooling
demand of the building. The thermal storage also provides residence
time buffering so that the absorption chiller can continue to operate
smoothly when solar radiation varies (e.g. due to cloud cover). This is
particularly important for absorption chillers, which have slow startup
times and a dilution cycle when going into shutdown [31]. It should be
noted that although the system in Fig. 11 looks simple, many critical
choices are available regarding the specific components and operational
strategies. To complicate matters, transients in the local climate and
building loads significantly influence performance.

3.5.1. Solar thermal collectors


A solar thermal collector is a device which absorbs the incoming
solar irradiation, transforms it to useful thermal energy and transfers
Fig. 10. Schematic diagram of a triple-effect absorption chiller.
this energy to a fluid (e.g. air, water, or oil) circulating through the
collector [61]. The collected thermal energy can be either directly fed
Table 1 into a thermally-driven application (e.g. heating, cooling, or power
Absorption chiller technical comparison [48–52]. generation cycles) or stored in a thermal storage tank to be utilized later
Working fluid LiBr-H2O H2O-NH3 [61]. Depending on the application, this collected heat can be delivered
at different temperatures, ranging from 30 °C to 1000 °C [61,62]. At
Number of effect Single Double Triple Single high operating temperatures, the amount of heat loss from the collector
can be significantly reduced by increasing the concentration ratio (i.e.
Nominal cooling ∼5–10,000 ∼20–10,000 ∼350–10,000 ∼10–1500
capacity (kW) reducing the absorber area) and, in some cases, by limiting the con-
COPth (–) 0.7–0.8 1.2–1.4 1.6–1.8 0.5–0.7 vective heat losses of the collector (e.g. vacuum insulation around the
Heat source 80–100 160–180 210–240 100–120 absorber).
temperature (°C) In general, solar thermal collectors can be classified into two broad
groups: non-concentrating and concentrating. In non-concentrating
collectors, the collector aperture area − the area that intercepts the
spray over an otherwise dry cooler on peak hot days to avoid prolonged
solar irradiation – is approximately the same as the absorber area. Thus,
shutdown periods of the absorption chiller [59]. Other heat rejection
the whole solar panel in this type of collectors absorbs sunlight [63].
methods such as hybrid cooling towers or the use of an air dehumidifier
Non-concentrating collectors are easy to install, do not require a me-
may also be adopted in such extreme climatic conditions [60].
chanical tracking system, and are nominally cost-effective. The sim-
plicity of these systems, however, limits their operational temperature
3.5. Solar-powered absorption chillers to well below 200 °C and, thus, their potential use in ‘high-value’ ap-
plications [64].
Absorption chillers have been traditionally powered by natural gas Concentrating collectors use mirrored surfaces to focus sunlight
or industrial waste heat in large buildings for decades [31]. In recent from a large area (the aperture area) to a much smaller area (the

65
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Table 2
Characteristics of typical, commercially available solar thermal collectors [61,62].
Tracking system Collector type Absorber type Concentration ratio Operating temperature range (°C)

– Flat plate collector Flat 1 30–120


Evacuated tube collector Flat 1 50–150
Compound parabolic collector Tubular 1–5 60–240

Single-axis Linear Fresnel reflector Tubular 10–40 60–250


Parabolic trough collector Tubular 15–50 60–300
Cylindrical trough collector Tubular 10–50 60–300

Two-axis Parabolic dish reflector Point 100–1000 100–500


Heliostat field collector Point 100–1500 150–2000

receiver area) where the absorber is located [65]. The ratio of the hot water and space heating, while some types of flat-plate collectors,
collector aperture area to the absorber (receiver) area is known as the such as double-glazed collectors (operating at 80–120 °C), could be
geometric concentration ratio (CR) of the collector. As the concentra- used to drive single-effect absorption chillers [67,68]. It should be
tion ratio increases, these collectors can deliver significantly higher noted that the heat losses associated with these collectors are
temperatures (200–1000 °C), without suffering from dramatic heat prohibitively high at higher operating temperatures – a factor that
losses. This, however, comes at the price of optical losses and compli- adversely affects the amount of useful heat gain and thus the thermal
cated tracking systems – resulting in a much higher capital cost per unit efficiency of the collector.
area of the collector [62]. Unlike non-concentrating collectors, con-
centrators collect little or no diffuse radiation (e.g. missing out on
3.5.1.2. Evacuated tube collector (ETC). A non-concentrating evacuated
15–100% of the available radiation), which limits their solar gain per
tube collector consists of parallel rows of vacuum-sealed glass tubes
unit area [61]. The general characteristics of common types of solar
that fit into a horizontal manifold located at the top of the collector, as
thermal collectors available on the market are presented in Table 2
shown in Fig. 13. Each tube contains a copper heat pipe and a dark
[61,62].
absorber coating which transfers solar-derived heat into a heat transfer
The type of solar thermal collector required to drive an absorption
fluid (e.g. purified water) inside the heat pipe. As this fluid receives
chiller critically depends on its number of effects. Low-temperature
heat, it evaporates and rises to the top of the heat pipe, where the
solar thermal collectors such as flat plate or evacuated tube collectors
collected heat is transferred to a cold water/oil stream flowing through
(FPCs and ETCs) are suitable to provide hot water to single-effect ab-
the manifold. As this process takes place, the vapor inside the heat pipe
sorption chillers. High-temperature collectors, capable of delivering
condenses and turns back into liquid returning to the bottom of the heat
temperatures around 200–240 °C with an acceptable thermal efficiency,
pipe [70]. Similar to flat plate collectors, evacuated tube collectors can
must be used to drive multi-effect absorption chillers. Based on current
technology, there are three types of high temperature solar thermal
collectors on the market, which could be potentially suitable for solar
cooling systems using double- and triple-effect absorption chillers:
linear Fresnel reflector collectors (LFRs), parabolic trough collectors
(PTCs), and evacuated flat plate collectors (EFPCs) [66]. The following
presents a brief description of the most common, commercially avail-
able solar thermal collectors which could be integrated with absorption
cooling technologies.

3.5.1.1. Flat plate collector (FPC). Flat plate collectors are the simplest
and probably cheapest way to harvest solar energy and produce thermal
heat. As illustrated in Fig. 12 a flat plate collector mainly consists of a
transparent cover that allows solar irradiation in, a dark, selective
absorber plate that converts the incoming radiation to heat and
transfers it to the tubing system attached to it, and a heat-insulating
structure on the backside to minimize heat losses [62]. FPCs are usually
employed for low-temperature applications such as providing domestic

Fig. 13. Schematic diagram and operating principle of an evacuated tube col-
Fig. 12. Schematic of a typical flat plate collector [69]. lector [71].

66
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Fig. 14. Schematic of a parabolic trough collector [75].

harness both beam and diffuse radiation. However, they have lower Fig. 15. Schematic representation of a linear Fresnel reflector collector [80].
heat losses as compared to flat plate collectors, due to the vacuum
insulation around the hot absorber tubes and the cylindrical geometry
of absorbers which has less dependence on solar incidence angle [70].
This results in ETCs achieving higher temperatures (up to 150 °C) with
an acceptable efficiency level than FPCs. This enables ETCs to be
principally suitable to drive a single-effect absorption chiller.
Employing ETCs to drive multi-effect chillers, however, is a
challenge, as the required temperatures in such cases are well above
the nominal operating temperature range of the collector, which can
dramatically reduce the thermal efficiency of the collector.

3.5.1.3. Parabolic trough collector (PTC). A parabolic trough collector


consists of an evacuated receiver tube positioned along the focal line of
a parabola-shaped reflector (Fig. 14). The tube is fixed to the mirror
structure and used to transfer solar irradiation reflected by the linear
concentrator to the heat transfer fluid (e.g. water or oil) passing
through the tube. To ensure the incoming sunlight strikes the tube,
the reflector and tube assembly needs to be equipped with a tracking
system. Unlike FPCs and ETCs, PTCs can only utilize the beam
component of global solar irradiance and must be spaced farther
apart to avoid shading, leading to a lower solar gain per unit area
[72]. This deficiency, however, may be partly compensated by their low Fig. 16. Schematic and exploded view of the Chromasun MCT solar thermal
heat losses, only occurring at the external surface of the tube which is collector [79].
relatively small compared to the collector aperture area. Having a
relatively high concentration ratio (around 15–50) and low heat loss applications [78]. Recently, a low-profile, roof-mounted micro-
levels, PTCs can achieve high operating temperatures up to 300 °C with concentrating collector (MCT) has been developed by Chromasun Inc.
a satisfactory thermal efficiency [62]. Such temperatures are sufficient [79], which uses linear Fresnel reflectors to concentrate beam radiation
to achieve pressurized high-temperature hot water in order to drive a to a stationary receiver. Fig. 16 shows a schematic and exploded view of
double or triple-effect absorption chiller. It should be mentioned that to this collector. As shown in this Fig, the entire optical system is enclosed
this date a number of PTC manufactures such as NEP Solar and in a sealed glazed canopy to reduce convective heat losses. MCT
SOLITEM GmbH have successfully installed PTC-based double-effect collectors can achieve temperatures up to 200 °C and can be
absorption chiller systems in a few demonstration projects [73,74]. seamlessly integrated into the architecture of buildings. Therefore,
they seem suitable to deliver thermal heat at around 180 °C in order to
3.5.1.4. Linear Fresnel reflector (LFR). Linear Fresnel reflectors, as drive double-effect absorption chillers.
shown in Fig. 15 use long, thin segments of mirrors to focus sunlight
onto a fixed absorber tube located at the common focal line of the 3.5.1.5. Evacuated flat plate collector (EFPC). Recently, an innovative
reflectors. A small parabolic mirror is used on top of the absorber tube evacuated flat plate collector has been developed by TVP Solar Inc.
as the secondary concentrator to further focus the beam radiation [81]. The TVP MT-Power, in particular, can capture both beam and
reflected from the reflectors to the absorber tube. This concentrated diffuse sunlight and achieve temperatures up 200 °C without any solar
energy is then transferred through the absorber tube into a heat transfer concentration or a tracking mechanism [82]. A schematic of the TVP
fluid, thereby increasing its temperature. Linear Fresnel concentrating MT-Power collector is shown in Fig. 17. Due to the high-vacuum
collectors are mostly manufactured for large-scale power production achieved inside this collector, the convective heat losses are reduced to
purposes [76] and therefore cannot be easily integrated on rooftops. a negligible level. For this reason, EFPCs offer significantly higher
Industrial Solar GmbH Ltd. [77], for example, has developed a thermal efficiency levels at temperatures around 180 °C as compared to
commercial cascading NH3/H2O absorption chiller driven by linear typical flat plate or evacuated tube collectors. Therefore, EFPCs can be
Fresnel collectors to produce cooling for frozen food (∼−12 °C) a promising design choice for integration with double-effect absorption

67
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

utilization of the solar field [84]. In addition, cold thermal storage


systems require a better insulation, as the energy available in the cool
state is more expensive than the heat available in the hot storage tank
[59,86].
Thermal storage systems can be based on the sensible heat of a solid
or liquid medium or the latent heat of a phase change material (PCM)
[84]. To date, most solar cooling installations have used sensible sto-
rage medium due to their relatively low cost, and there is limited design
experience for the use of latent heat storage systems in this field [15].
Nonetheless, research on new PCM candidates with competitive prop-
erties is underway, aiming to reduce the size of the storage tank in solar
Fig. 17. Schematic of TVP high-vacuum, flat plate solar collector [83]. absorption cooling systems [15,87]. The operating temperature of the
chiller is an important factor when selecting a suitable storage material
for solar absorption cooling systems. Water is the most common storage
Table 3
medium used for low-temperature solar absorption cooling systems
Absorption chiller: solar thermal collector matching.
using single-effect chillers. This is because water has high specific heat
Chiller type Nominal operating temperature Collector type capacity, and is readily available, inexpensive and chemically stable
(°C) [15]. Due to its boiling point constraint (100 °C at 1 bar), the use of
Single-effect ∼80–100 Flat plate collector water as sensible heat storage medium for high-temperature solar
Evacuated tube collector multi-effect chillers requires increasing the system pressure to keep
Double-effect ∼180–200 Linear Fresnel reflector water in liquid phase. As an alternative to pressurized hot water,
Evacuated flat plate thermal oil could also be used as the storage medium in such applica-
collector
tions. The advantages and disadvantages associated with these two
Parabolic trough collector
Triple-effect ∼210–240 Parabolic trough collector media are presented in Table 4. It should be noted depending on the
design conditions and standards, the solar cooling industry in different
jurisdictions may prefer either pressurized hot water or thermal oil.
chillers.
In summary, Table 3 lists the type(s) of solar thermal collectors that
could be potentially used to drive single-, double-, and triple-effect 3.5.3. Backup options
LiBr-H2O-based absorption chillers. Due to the intermittent nature of solar energy, a (fossil fuel-based)
backup system is required in solar absorption chiller systems to meet
the building load demand when solar inputs are insufficient. The
3.5.2. Thermal storage tank backup system can use either gas or electricity (or rarely, a ground
The basic function of a thermal storage system is to store energy in a source) as the auxiliary energy source. Gas-fired heaters and/or elec-
storage medium to be used later [84]. Thermal storage is a vital part of trically-driven mechanical vapor compression chiller are the most
solar absorption chiller systems, acting as a buffer to overcome the commonly used. It should be noted that due to the high capital costs of
intermittent nature of solar energy and the variation in load demand of solar absorption chillers, using two backup systems at the same time is
buildings. It also provides residence time buffering to the chiller, pre- unlikely to be economical [59,88].
venting it from frequent on/off cyclings. Generally, there are two pos- An auxiliary heater (AH) can be used either in series or parallel with
sible ways to integrate a thermal storage unit into a solar absorption the storage tank as shown in Fig. 18.
chiller system. One way is to store the solar-derived heat in a thermal In the series arrangement, the auxiliary heater is used to boost the
storage tank, which can be later used to drive the chiller. The second temperature of the hot water (HW) exiting the storage tank (ST). Thus,
option is to store the cold energy produced by the chiller in a cold the flow always goes through the auxiliary heater regardless of whether
storage unit. While the use of hot storage tank in the solar field is an it is in use. In the parallel arrangement, the flow bypasses the storage
inseparable part of a solar cooling system [6,70], adding a cold storage tank when the auxiliary heater is in used and the entire energy re-
unit after the chiller could reduce both the chiller capacity and the quirement of the chiller is met by the auxiliary system whenever the
mismatch between the supply and demand [85]. It can also improve the tank temperature is too low to be useful.

Table 4
Comparison of thermal storage media for high temperature solar thermal cooling applications.
Storage medium Advantages Disadvantages

Pressurized hot water • Potential for one fluid/ no secondary heat exchanger • Must comply with pressure vessel/ steam codes
system:
– No temperature reduction over secondary heat
• Possible need for frost protection required
exchangers
– Less fluid streams to pump/ less parasitic pumping
power
• High heat transfer coefficients
• Low cost, environmentally friendly fluid
• High chemical stability

Thermal oil •– Atmospheric pressure • Cost of fluid


Industry comfort • Environmental management in case of possible leakage
– Low cost • Large difference in viscosity over the operating temperature range
• Inherent frost protection • Not recommended for direct use in the absorption chiller (requires an extra heat
exchanger to transfer heat to hot water)
• Effects, on the oil, of possible contamination with water and air
• Special (expensive) pumping system
68
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

AH variables inside the chiller (e.g. temperatures, pressures and COP)


which played the biggest role in determining the thermal performance
T

HW Recirculation
of the system.
Bypass

ST Controller 4.1.2. IEA SHC Task 38


On/Off (from the building The main objective of Task 38 − Solar Air-Conditioning and
A/C control system)
Refrigeration – which was initiated in 2006 was to implement measures
T
for an accelerated market introduction of small and medium-sized solar
thermal air-conditioning and refrigeration systems [96]. The distribu-
Fig. 18. Series and parallel arrangements of auxiliary heater, upstream of the
tion system, the building, and the interaction of both with the technical
absorption chiller.
equipment were not considered in detail in the Task. An overview of
worldwide installed solar cooling systems was also conducted in this
CW loop
task, which showed the predominance of absorption chillers among the
installed systems. While the notion of solar-assisted multi-effect ab-
sorption chillers was not investigated in this task, the use of high-
temperature solar thermal collectors to drive double- and triple-effect
absorption chillers was suggested as a potential next step in the de-
velopment of solar absorption chillers.
Bypass

HW loop Controller 4.1.3. IEA SHC Task 45


On/Off (from the building
A/C control system)
The main objective of Task 45 − Large Scale Solar Heating and
CW loop
Cooling Systems – (duration: 2011–2014) was to assist in the devel-
opment of a sustainable market for large-scale solar heating and cooling
Fig. 19. Series and parallel arrangements of a mechanical vapor compression systems by focusing on cost-effectiveness and performance reliability
chiller as the cooling backup system, downstream of the absorption chiller.
[97]. The scope covered in this task was relatively broad, including
improved collector testing, guidelines for materials and construction for
Similar to the gas heater, a backup mechanical chiller can also be seasonal storage systems, and a simple analysis of low temperature
arranged in either parallel or series with respect to the absorption solar thermal cooling systems. This task lacks control strategies on a
chiller. These arrangements are illustrated in Fig. 19. In the parallel system level or solar-driven multi-effect absorption chillers.
configuration, only the absorption chiller or the mechanical chiller
operates at any given time to satisfy the required cooling demand of the 4.1.4. IEA SHC Task 48
building. Thus, both the absorption and mechanical chillers should be Task 48 (Quality Assurance and Support Measures for Solar
sized such that either can meet the maximum cooling demand of the Cooling) is a project conducted by a group of researchers and practi-
building. To achieve a more economical design of the system, the ab- tioners from 9 countries around the world [98]. This task is aimed at
sorption chiller (and solar field) can be undersized. This requires a finding solutions to enable the solar cooling industry to deliver systems
series arrangement between the two chillers, as the (undersized) ab- that are efficient, reliable, and cost competitive. These three major
sorption chiller is not able to supply the whole cooling demand alone. targets are to be achieved through activities grouped into four subtasks:
Therefore, in the series configuration, the absorption and conventional
chillers can operate simultaneously to satisfy the demand of the
building.
• Subtask A: Development of tools and procedures that characterize
the performance of the main components of solar absorption sys-
tems;
4. Solar-powered absorption chiller in the literature • Subtask B: Creating practical and unified procedures for specifying
the best technical configurations for complete integrated solar ab-
There have been wide-ranging studies in the literature dedicated to sorption chillers;
design, simulation, experimental analysis and optimization of solar
absorption chiller systems [8,26–28,55,89–93]. As such, this section
• Subtask C: Development of standards and procedures to identify and
validate the quality of solar absorption chillers; and
presents a review on the recent development in the field of solar-
powered absorption chiller technologies as follows.
• Subtask D: Production and dissemination of information to promote
solar thermal driven cooling and heating systems

4.1. International Energy Agency (IEA) SHC tasks It should be mentioned that the optimal control and design of solar-
assisted multi-effect absorption chillers in large buildings is an im-
The International Energy Agency (IEA) has organized several portant target of this task. This aspect was not investigated in the
dedicated projects as part of Solar Heating and Cooling (SHC) Program, previous IEA SHC tasks, and may prove particularly useful in climates
aiming to enhance collective knowledge and application of solar with high levels of direct normal irradiation. In addition, as part of
heating and cooling through international collaboration [94]. The fol- Subtasks A and B, a solar double-effect chiller prototype was built at
lowing is a brief review of the most relevant IEA SHC Tasks to the topic CSIRO, Newcastle, Australia to elucidate some of the system complexity
of solar-powered absorption chiller: associated with solar multi-effect absorption chillers.

4.1.1. IEA SHC Task 25 4.1.5. IEA SHC Task 53


Task 25 − Solar Assisted Air Conditioning of Buildings – was set by Task 53 − New Generation Solar Cooling and Heating Systems (PV
the IEA in 1999 to improve conditions for the market entry of solar- or Solar Thermally Driven Systems) – (2014–2018) was focused on
assisted cooling systems, thereby promoting primary energy savings assisting market development of small to medium-sized PV and solar
and reductions in electricity peak loads with regard to cooling [95]. As thermal-driven cooling systems, ranging from 1 to a few tens of kW (i.e.
for system level simulation, Subtask B of this Task investigated a residential applications) [99]. Neither the building nor its interaction
small–scale (5 kW) NH3/H2O single-effect absorption chiller prototype with the technical equipment was the main focus of the Task. Thus, the
for residential applications. The results included some thermodynamic control strategy analysis and optimization components of this task

69
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

mostly focused on PV-driven systems. In addition, there is no mention The results revealed that either increasing the heat source temperature
of multi-effect absorption chillers in this task. or decreasing the cooling water inlet temperature increases the risk of
crystallization. However, the authors discovered that the chiller has a
4.1.6. IEA SHC Task 55 better performance when the hot water source temperature is selected
Task 55 − Towards the Integration of Large SHC Systems into as the manipulated variable to adjust the delivered cooling load. Labus
District Heating and Cooling (DHC) Network – is focused on storage et al. [110] also proposed a control strategy for absorption chillers
systems, industrial waste heat, and heat pumps in solar district heating using artificial neural networks. Their model took into account inlet
and cooling [100]. This Task was initiated very recently (i.e. September and outlet temperatures as wells as flow rates of external water circuits
2016), so little in the way of tangible results was available for review at to determine the optimal performance of the chiller and achieve the
the time of writing this paper. required cooling capacity It should be noted that changing the mass
flow rates in the chiller’s external loops may not be a feasible option as
4.2. System arrangement and control a control strategy since commercial chillers are often designed to op-
erate at fixed flow rates, thereby preventing the use of VSD pumps
As mentioned above, solar absorption air-conditioning systems can [59–61]. In addition, Albers [111] developed a new control strategy to
be designed with a number of different configurations and components, govern the cooling capacity of absorption chillers by changing the hot
resulting in quite different performance. Although there is a substantial and cooling water temperature simultaneously, leading to a 5% re-
body of research in the field of solar absorption chillers, surprisingly duction in the operating cost of the system. Nowadays, almost all large-
little has been written from the perspective of how to design the ‘best’ scale absorption chillers are equipped with a microprocessor-based
system. control panel which monitors and control all operations of the machine
A basic, yet controversial difference of opinion seen in the literature [48–52].
is whether the auxiliary heater and storage tank should be in series or in Although an increasing amount of research is devoted to the control
parallel. Initially introduced by Ward et al. [101] in 1977, the question strategies of absorption machines [112,113] and heat rejection devices
of which arrangement between the auxiliary burner and the storage [58,114,115], less emphasis has been placed on the control approaches
tank (i.e. series or parallel) is appropriate has not been answered by within the solar collector loop in solar absorption chiller systems. For
many studies in the field. As a case in point, Hang et al. [94,95], and Li example, many authors have used constant speed pumps on the solar
and Sumathy [96] together with a few other researchers [97,98] have collectors [116–120], but only a few works have stressed the im-
reported parallel auxiliary heater arrangements in their work. In these portance of variable speed pumps [105,121–123]. Since the cost of
studies, the auxiliary heater was arranged in a way to supply the full variable speed devices has significantly dropped in recent years, the use
energy requirements of the building when the solar-driven energy of variable speed pumps and fans has become much more widespread in
stored in the storage tank was not sufficient to drive the absorption air-conditioning systems [124,125]. Due to the transient nature of solar
chiller. It should be noted that no further analysis on the behavior of the energy, a variable speed pump would appear to be even more important
tank and the performance of the solar cooling plant as a whole under in solar HVAC than in conventional HVAC. This way, a desired solar
this arrangement was reported in these studies. On the other hand, collector outlet temperature can be achieved, thereby enabling the
there are several other studies [99,100,102–105] in which the auxiliary absorption chiller to operate for extended periods of time thus in-
heater was implemented in series, where the auxiliary heater boosts the creasing its utilization.
energy level of the tank if the storage temperature is too low to be able Recently, Shirazi et al. [126] conducted a comprehensive analysis to
to meet the heating and cooling demand. Similarly, these studies did determine which alternative design option is the best configuration and
not provide any reasoning as to why this arrangement was adopted and control strategy among a long list of design alternatives presented in the
whether it leads to an energy-efficient performance of the system. field of solar absorption chillers in the literature. The authors proposed
The control strategy is also a topic of interest in recent literature. three control scenarios in the solar collector loop as follows: (i) a
Yeung et al. [106] developed a simple control mechanism to adjust the constant speed pump without a set-point temperature, (ii) and (iii) a
chilled and cooling water temperatures relative cooling load profiles for variable speed pump with a fixed and variable set-point temperature at
a solar-powered absorption air-conditioning system constructed at the the collector outlet. Considering these three control strategies, the
University of Hong Kong. In this control scheme, the hot water supply performance of a solar-assisted single-effect LiBr-H2O absorption
was switched off once the chilled water temperature dropped below a chiller was analyzed under both series and parallel arrangements of the
specified set-point, while the cooling water temperature was controlled auxiliary burner and the storage tank. The yearly performance results
by a differential controller in an on/off mode. This control strategy led from the parametric study of the proposed configurations showed that
to increased operational fluctuations and thermal heat losses as a result the total solar fraction of the plant was increased by up to 11% when a
of frequent start-up and shut-down procedures of the chiller. Koeppel variable speed solar loop pump was used to achieve a collector set-point
et al. [107] determined an optimal supervisory control of a natural gas- temperature adjusted according to the building load demand. Another
driven, direct-fired, double-effect LiBr-H2O absorption chiller using a significant finding of this study was that a parallel configuration for the
simulated annealing method − a probabilistic technique for approx- auxiliary heater out-performs a conventional series configuration. The
imating the global optimum of a given function – in order to minimize yearly performance of an auxiliary heater in parallel with the storage
the total operating cost of the chiller. The authors found that depending tank enhances the plant solar fraction, and the average collector effi-
on the control options, the optimal supervisory control resulted in a ciency, by up to 13% and 9%, respectively (as compared to the same
15% to 21% reduction in the total energy cost compared to the base components in series). Similar results have also reported by Ahmed
case control scheme. Liao and Radermacher [33] proposed a control Khan et al. [127], showing that the parallel arrangement led to higher
strategy based on regulating the chilled water or condenser tempera- primary energy savings.
tures to avoid crystallization in LiBr-H2O air-cooled absorption chillers.
Bujedo et al. [108] conducted a similar study, where they proposed 4.3. System simulation and modeling
three control strategies with regard to the chiller operation through
regulating the mass flow rate and temperature of cooling and hot water The performance of a solar-powered absorption chiller can also be
streams. Xu et al. [109] evaluated the dynamic performance of a single- affected by both design-related parameters such as the characteristics of
effect LiBr-H2O absorption chiller under two control strategies: (i) set- solar field, storage and chiller as well as variable factors like weather
ting the chilled water outlet temperature as the manipulated variable, conditions and the building load profile [70]. Mathematical modeling
and (ii) setting the hot water temperature as the manipulated variable. offers the possibility to study a physical system and discover ways to

70
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

improve its overall performance where possible. Simulation modeling is libraries (Standard Library and TESS Library) available in this program,
a useful tool that can be employed to evaluate and understand the offering a wide range of components that are commonly used in applied
behavior and operation of these systems under various design circum- energy systems. TRNSYS also allows users to develop new components
stances [70,96,128–130]. There are a wide range of simulation pro- through programming languages such as FORTRAN and C++ and in-
grams that have been used in the field of energy systems and building tegrate them into the simulation environment [135]. This feature gives
simulation. SPARK (Simulation Problem Analysis and Research Kernel), TRNSYS the flexibility to accommodate the changing needs of re-
for example, is a generic, equation-based simulation tool that is mainly searchers in the realm of energy system simulation. After being com-
used to solve thermal processes in buildings [60]. SPARK has its own piled in the dynamic link library (DLL), the user-defined components
HVAC library based on simple thermodynamic models. In addition, become independent of the programming language in which they were
various user-defined components in the field of building and HVAC written, leading to a system simulation with significantly lower simu-
systems have been created in SPARK environment by students and re- lation time compared to other simulation tools (e.g. MATLAB/Simu-
searchers. Although many of these models were dedicated for internal link). TRNSYS also contains a comprehensive set of weather data files to
use, they could be distributed by their developers [60]. EnergyPlus is a facilitate the system performance comparisons at different locations
building energy simulation tool used to calculate the cooling and and climatic conditions [136]. The most recent weather files distributed
heating loads of buildings. Although it does have a number of pre-de- with TRNSYS 17 were obtained from two different data sources: (i)
fined conventional HVAC components, EnergyPlus is still not suitable typical meteorological year (TMY2) datasets for 239 locations in the US
for the integration of novel air-conditioning systems (e.g. solar cooling which have been collected by the US National Renewable Energy La-
technologies) in buildings, and therefore has to be coupled with a third- boratory (NREL), and (ii) weather files for selected worldwide stations
party program. INSEL, which stands for Integrated Simulation En- (including 1000 locations in more than 150 countries) generated by
vironment Language, is a general-purpose graphical programming Meteonorm software (Version 5) published by Meteotest Ltd. [134].
language used in the field of energy systems. This tool is based on The US TMY2 data files were derived from the 1961–1990 National
structured programming of block diagrams, and covers a wide range of Solar Radiation Data Base (NSRDB) [137], and they are based on hourly
renewable energy systems, including building simulation and commu- values of solar radiation and meteorological elements for a one-year
nication technologies [131]. The core component of INSEL is the in- period. The weather and radiation data from Meteotest are based on
selEngine – a compiler which is able to interpret and execute applica- monthly values that Meteonorm generates stochastically to hourly va-
tions written in the INSEL language. The use of DELAY blocks in this lues in its 1991–2010 database. The TMY2 output format was used in
program could be problematic in terms of affecting the functionality of Meteonorm software because it is consistent with the TRNSYS weather
control blocks [132]. In addition, it was reported that several compo- data file reader type (Type 15).
nents (e.g. the storage and pipe models) require a very short time-step As for the modeling of absorption chillers, both thermodynamic and
in order to properly predict the system thermal performance, which can empirical approaches have been well established in the literature.
increase the simulation’s computation time. Polysun [133] is another Kaynakli and Kilic [138] developed a thermodynamic model of a LiBr-
simple, component-based simulation program, which is mainly used to H2O absorption chiller system based on detailed mass and energy bal-
design renewable energy systems such as solar photovoltaic, solar hot ance, and heat transfer equations of the chiller components. The au-
water, and heat pumps. In addition to pre-defined components, Polysun thors analyzed the effects of operating temperatures and heat ex-
has an economic analysis package, which can provide useful technical changer effectiveness on the thermal loads of the components and the
reports and the information necessary to obtain government subsidies COP of the chiller. Grossman and Zaltash [139] developed a modular
in some locations in Europe and the USA. However, this economic thermodynamic model of an absorption chiller system in order to
model is suitable only for solar thermal heating systems and rooftop PV evaluate various absorption cycle configurations working with different
systems [132]. Although being intuitive and user-friendly, Polysun is working fluid pairs. Although accurate, this model requires many in-
less appropriate for solar cooling systems [132]. EasyCool is a simula- ternal parameters such as heat transfer coefficients, heat transfer areas,
tion tool which was originally designed to perform simple and fast solution flow rate, and working fluid properties, which in most of the
energetic and economic analysis of solar thermal cooling systems as cases are not provided by the chiller manufacturer. This is why ther-
part of Solar Air-Conditioning in Europe (SACE) projects [60]. This si- modynamic models are more appropriate at the design stage of ab-
mulation package consists of 11 pre-defined system configurations for sorption chillers, where more information about the behavior of the
solar thermally driven cooling applications. After reading the annual chiller with respect to component characteristics and the cycle internal
time series and weather data file, this program calculates the annual conditions is required [35]. In addition, the computation time of ther-
energetic and economic performance of these systems. Due to the modynamic models is very long, as they normally perform multiple
limited features of EasyCool, it is not possible to define new system non-linear iterations simultaneously. This can dramatically increase the
components and configurations in this simulation environment. Transol time which takes to run annual simulations which usually require time-
is another easy-to-use simulation tool, consisting of 40 predefined steps of less than an hour. Empirical models, on the other hand, are a
configurations for solar thermal systems (based on TRNSYS models), simple and relatively fast method for obtaining an accurate re-
which allow the user to conduct simple energy calculations and opti- presentation of, for example, an absorption chillers’ behavior based
mization runs. This program works with meteorological data generated only on external parameters (i.e. experimental measurements or man-
by Meteonorm software [134], based on which the building heating and ufacturer catalog data). Although they provide little in the way of
cooling demands can be calculated. physical understanding, empirical models require less time and effort to
Most of the theoretical studies on solar absorption chillers have develop and can easily be incorporated in simulation programs with
been conducted using Transient Systems Simulation Program (TRNSYS) much lower computation time as compared to thermodynamic models.
[8]. TRNSYS has been highlighted by Subtask C of the IEA’s SHC Task Kühn and Ziegler [140] developed a model called characteristic equa-
38 (modeling and fundamental analysis of solar air-conditioning and tion method, which predicts the performance of the absorption chiller
refrigeration) [96], as an appropriate tool to assess detailed system- by two simple algebraic equations: one for cooling capacity and the
level modeling and simulation of solar cooling systems. TRNSYS is a other for the driving heat input into the chiller. Puig-Arnavat et al.
component-based simulation program which is widely used to simulate [141] used the same approach to estimate and compare the perfor-
the behavior of transient systems such as renewable energy systems mance of some commercial single- and double-effect absorption chillers
[135]. The software consists of sub-routines that represent system through fitting the catalog and experimental data of the chillers. The
components (real physical devices), types, which work as a module and authors also presented a number of recommendations for applying the
can be linked together to represent a system. There are two extensive characteristic equation method to any absorption chiller. To

71
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

demonstrate the capability of the characteristic equation model, the perspective. Their simulation results suggested that evacuated tube
authors calculated the hourly cooling production of the modeled single- collectors should be preferred over flat-plate collectors in solar-driven,
effect chiller integrated with solar thermal collectors, and then com- single-effect absorption chillers due to their higher thermal efficiency.
pared the results with experimental data. The results showed that the Gomri [44] analyzed a solar-driven single-effect absorption chiller with
characteristic equation method successfully predicted the part load and a gas-fired auxiliary heater at various generator and condenser tem-
full load behavior of the chiller with an acceptable accuracy, and peratures and obtained a broad range of energetic performance levels of
therefore could be implemented in simulation programs, thereby the system. While the author concluded that the solar-assisted chiller
avoiding the long computation time associated with thermodynamic backed up by a gas burner was environmentally cleaner than a con-
models. Moreover, Labus et al. [142] used four different empirical ventional gas-fired absorption chiller, he did not investigate whether
modeling methods, including the adapted Gordon-Ng method, poly- this configuration would actually consume less primary energy or if it
nomial regression, artificial neural networks, and the adapted char- would be economically feasible as compared to a standard electrically-
acteristic equation model, to model an absorption chiller based on the driven vapor compression chiller under the same conditions. In addi-
performance curves provided by the chiller manufacturer. The authors tion, this study lacks the economic performance of the considered
concluded that the characteristic equation method was highly accurate system. Mammoli et al. [147] analyzed the performance of a solar
and fast in predicting the chiller thermal performance using only the single-effect absorption chiller under different control strategies for
variables of external water circuits as model input parameters. summer daytime, summer nighttime, winter daytime, and winter
There have also been numerous studies in the literature with regard nighttime operations. The plant was designed to provide cooling and
to modeling and simulation of solar absorption chiller systems heating demands of the Mechanical Engineering Building at University
[28,91,143,144]. Florides et al. [100,145] developed simulation of New Mexico in USA. The results showed that the plant solar fraction
models of solar single-effect absorption chillers under Cyprus climate can be boosted by 60% if the hot water storage tank is well-insulated.
using TRNSYS program. The authors carried out a parametric study by The authors also suggested a set of design recommendations for the
varying the main design parameters of the system to determine an solar collector array, hot water storage, cold water storage, and air
optimal design of the plant. Atmaca and Yigit [129] simulated a single- handling units. Recently, Reda et al. [89] developed a TRNSYS model of
effect solar absorption cooling system to cover a constant cooling load a solar-single effect absorption chiller using district heating systems
of 10.5 kW. The authors developed a modular computer program to based on cogeneration heating plants (CHPs) as an auxiliary heater for
simulate various cycle configurations and solar energy parameters for an office building in Finland. The results showed that this system led to
Antalya, Turkey. It was found that a solar collector area of 50 m2 and a peak electricity shaving during the summertime. The authors, however,
3.75 m3 storage volume led to an optimal energetic performance of the did not mention whether using a gas-driven backup system was efficient
plant. It should be noted that the authors did not investigate the in- from primary energy perspective, nor did they discuss the economic
tegration of the modeled plant in a building. Joudi and Abdul-Ghafour viability of such systems under European climates. Sokhansefat et al.
[98] developed a simulation program of a solar cooling system using a [148] conducted a transient simulation of a 17.6 kW solar absorption
LiBr-H2O absorption chiller system. Based on the simulations results, cooling system installed in Tehran, Iran. In addition to the collector
the authors created a general design procedure in a graphical form area and the tank volume, the auxiliary burner set-point temperature
called the cooling f-chart. Assilzadeh et al. [118] modeled a solar- was also considered as a design variable. The authors determined the
powered single-effect absorption chiller designed for a tropical climate optimal design point of the system through a parametric study, but did
in Malaysia using TRNSYS software. The authors investigated the en- not investigate the economic performance of the system. Recently,
ergetic performance of the plant through varying the collector area and Hirmiz et al. [149] developed an analytical framework to analyze the
slope, the storage tank volume, and the pump flow rate. Due to the fact performance of low-temperature PCM systems coupled with solar
that the units for the storage tank volume and collector area were not single-effect absorption chillers. The results showed that PCM-based
normalized, it is difficult to compare their approach in sub-optimal storage can reduce the storage volume by 43%.
conditions. Mazloumi et al. [146] modeled a single-effect LiBr-H2O The major disadvantage to solar-driven, single-effect chillers is the
absorption chiller integrated with parabolic trough collector (PTC) in low COP of the chiller, leading to large (read: expensive) solar collector
order to provide the cooling demand of a residential house with a areas required to supply the thermal heat demand of the chiller.
maximum cooling load of 17.5 kW. They obtained the minimum re- Further, this may be a significant limiting factor in buildings with
quired collector area and optimum storage tank capacity for various limited available rooftop area. Combining high-temperature solar
flow rates within the collector loop. The results revealed that the col- thermal collectors and multi-effect absorption chillers can be more
lector mass flow rate has a negligible effect on the minimum required energy-efficient due to the higher COP of these chillers, resulting in less
collector area, while it has a significant effect on the optimal capacity of solar-derived thermal energy and (potentially) less collector area re-
the storage tank. As the operating temperature of PTCs is well above quired to supply a given amount of cooling [150,151]. While many
that of single-effect chillers, it is not clear whether the choice of high studies in the literature have been dedicated to simulation and analysis
temperature PTC-driven single-effect chillers make sense from both of solar-assisted single-effect absorption chillers, little emphasis was
energetic and environmental perspectives. A similar study was carried placed on the use of multi-effect absorption chillers integrated with
out by Baniyounes et al. [116] for an office building under subtropical high-temperature solar thermal collectors [152]. Kaushik and Arora
climates in Australia, aiming at improving the system energetic per- [153] modeled and analyzed a single- and double-effect LiBr-H2O
formance. Since the study was focused on technical aspects of the chiller based on the 1st and 2nd laws of thermodynamics. The results
system, it is unclear if the systems would be economically feasible. from the 1st law analysis indicated that the COP of the double-effect
Balghouthi et al. [120] investigated the feasibility of solar-powered chiller was 60–70% higher than that of the single-effect chiller, while
single-effect absorption chillers under Tunisian climates and obtained the optimum heat input temperature to the double-effect chiller was
an optimal design for reliable operation of the system from an energetic found to be 150 °C as opposed to 91 °C for the single-effect chiller. The
perspective. The authors, however, did not investigate the economic results obtained from the 2nd law, however, suggested that the optimal
feasibility of the proposed plant when compared to conventional operating temperature of the chillers should be reduced by ∼12% to
cooling systems. Martinez et al. [117] developed a TRNSYS model of a achieve maximum exergetic efficiencies. Cabrera et al. [154] carried
low-capacity solar single-effect absorption chiller using different types out a comprehensive literature review on the use of parabolic trough
of solar thermal collectors in order to gain new insights into the op- solar thermal collectors in solar air-conditioning applications and
eration of the system and found the optimal system design parameters summarized the existing experiences and realizations for the potential
(i.e. collector area and storage tank volume) from an energy efficiency application of parabolic trough collectors to feed double-effect

72
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

absorption chillers. They found that the yearly installation rate of this exergetic efficiency of the system. The author found that the maximum
type of systems is still low − four installations per year – as compared COP of the single-, double-, and triple-effect chillers can be within the
to those fed by FPCs and ETCs. However, according to the market po- range of 0.73–0.79, 1.22–1.42, and 1.62–1.90, while the maximum
tential, this rate is expected to increase in the near future. Tierney [155] exergetic efficiency of these chillers were in the range of 12.5–23.2%,
investigated the potential of the combination of PTCs and a small-sized 14.3–25.1%, and 17.7–25.2% respectively. Khamooshi et al. [163]
double-effect absorption chiller as an alternative for typical solar single- carried out a similar study, considering six different configurations of
effect chillers in order to reduce the gas-firing requirements. The results triple-effect heat transformer absorption chillers integrated into a de-
suggested that integrating PTCs and double-effect absorption chillers salination system for waste heat recovery from industrial processes.
can achieve an energy saving of up to 86% as compared to a conven- They found that the condensation temperature should always be kept at
tional gas-fired double-effect chiller. The author, however, did not in- a minimum value in order to achieve higher COPs. Wang et al. [164]
vestigate how the performance of the plant would be affected at dif- proposed the integration of an exhaust heat recovery generator with
ferent climates. Qu et al. [156] analyzed the performance of a small- conventional gas-fired triple-effect LiBr-H2O absorption cooling cycles
scale solar cooling prototype based on PTC modules and a double-effect to improve the energy efficiency of the system. The results showed that
LiBr-H2O absorption chiller installed at Carnegie Mellon University, the COP of the integrated chiller was enhanced by 3%. The authors
USA. The plant consisted of 52 m2 of parabolic trough collectors, a concluded that this technology could be easily implemented and pro-
16 kW double-effect LiBr-H2O absorption chiller, and a heat recovery vide promising economic prospects in the development of triple-effect
heat exchanger with circulation pumps and control valves. The authors absorption cycles. Recently, Lizarte and Marcos [165] developed a si-
performed a parametric study in order to determine the optimal col- mulation code of a triple-effect parallel-flow LiBr-H2O absorption
lector orientation, collector receiver area, storage tank volume, and chiller, aiming to determine the optimal COP of the chiller under design
pipe lengths and diameters from a thermodynamic perspective. They and off-design conditions. To achieve this goal, computer simulations
found that a properly designed system could potentially supply 39% of were conducted for different evaporation and condensation tempera-
the cooling and 20% of the heating demand of the building. The eco- tures and a wide range of solution concentration in all generators. The
nomic aspect of this prototype was not discussed by the authors. A si- authors finally presented useful tips for optimizing the control regula-
milar study was carried out by Balghouthi et al. [157] under Tunisia’s tion of future parallel-flow triple-effect machines driven by a variable
climate. They found that solar-assisted double-effect chillers can sig- temperature heat source like solar energy.
nificantly contribute to CO2 emissions reduction as compared to a The potential of triple-effect absorption systems fed by solar thermal
conventional cooling system. Calise [158] developed a dynamic model energy has been rarely discussed in the literature. Agrawal et al. [160]
of a high temperature solar cooling plant based on coupling a double- proposed a solar-powered triple-effect absorption chiller integrated
effect LiBr-H2O absorption chiller with parabolic trough collectors. He with an ejector and cascaded cycles for refrigeration application at
performed a detailed parametric study to evaluate the influence of such temperatures ranging from −80 °C to −50 °C. A heliostat and central
important design parameters as the collector area, storage tank volume receiver were employed as the concentrator-receiver system in the solar
and set-point temperatures on the system performance. The results field. The authors investigated the effect of key design parameters on
showed that high temperature SHC absorption systems can be con- the energetic and exergetic efficiencies of the plant and concluded that
sidered as one of the most profitable renewable energy technology in solar-driven triple-effect absorption refrigeration systems are suitable
Mediterranean countries due to the higher COP of the double-effect for this application from thermodynamic viewpoint. An integrated PV/
absorption chillers. The author, however, did not mention how the size thermal and triple-effect absorption chiller was studied by Ratlamwala
of solar field or the capital cost of such plants compared with those of et al. [166] from energetic and exergetic perspectives for hydrogen
solar single-effect chillers. Moreover, Chemisana et al. [159] in- production and cooling. The electricity produced in PV modules was
vestigated the feasibility of a double-effect absorption chiller integrated supplied into an electrolyzer to produce hydrogen which is later burned
with linear Fresnel concentrating collectors. The results showed a re- and provides the driving heat into the absorption chiller. The results of
duction in the size of solar field as compared to standard low-tem- their parametric study showed that the highest COP of the chiller was
perature solar thermal collectors. The authors, however, did not ana- achieved for months when both solar radiation and the air inlet tem-
lyze the performance of the plant from an economic standpoint. Li et al. perature to the PV panels were high relative to the required cooling
[160,161] investigated the influence of the operating temperature of load. The authors did not consider any building model nor implement a
solar collector on the performance of a solar-powered 20 kW air-cooled, cost analysis to evaluate the economic feasibility of the plant. They also
LiBr-H2O double-effect absorption chiller in a subtropical climate in did not consider the possibility of using the concentrating solar thermal
Guangzhou, China. The authors developed a simulation model of the collectors to directly drive the triple-effect chiller. Shirazi et al. [66]
plant with a 27 m2 compound parabolic concentrator array in the solar investigated the feasibility of SHC absorption systems based on com-
field and suggested a suitable range of collector temperatures for the bining three types of LiBr-H2O absorption chillers (single-, double-, and
optimal operation of the proposed solar cooling system. Since no triple-effect) with common solar thermal collectors available on the
backup system was considered in this system, a continuous operation of market. A simulation model for each configuration was developed in a
the plant outside sunshine hours was not possible. transient system simulation environment (TRNSYS 17). The authors
Triple-effect absorption chillers can operate more efficiently than introduced the ratio of solar beam radiation to global radiation as a
single- and double-effect chillers when coupled with high-temperature unique parameter for analyzing the techno-economic performance of
solar thermal collectors [47]. Although triple-effect chillers driven by the proposed SHC plants for a wide variety of climatic regions world-
natural gas or waste heat are already commercially available on the wide. The results of parametric study suggested that a storage volume
market [48,49], testing data and standardization of such systems cou- of around 70 L m−2 was a good choice for the solar single-effect chiller,
pled with solar energy are still scarce. Matsushima et al. [162] devel- while 40–50 L m−2 storage capacity was sufficient for the multi-effect
oped a dynamic simulation program to predict the transient behavior of chiller configurations. The authors also found that when the fraction of
absorption chillers with different cycle configurations. This simulation direct normal irradiance (DNI) is less than 50%, the multi-effect chillers
program was used in the development phase of a high-efficiency triple- using concentrating collectors require larger collector area compared to
effect absorption chiller. The results obtained showed that this machine the solar single-effect chiller. However, in climates with DNI fractions
can operate smoothly and reached COP values of > 1.6. Gomri [47] above 60%, the smallest solar field is achieved by the configurations
analyzed the performance of gas-fired triple-effect LiBr-H2O chillers for using concentrated solar thermal collectors. Recently, Pintaldi et al.
chilled water production by performing a parametric study. He eval- [167] investigated the performance of sensible (i.e. thermal oil) and
uated the influence of various operating parameters on the COP and latent (i.e. PCM) thermal storage systems in a solar-driven absorption

73
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Table 5
Summary of the reviewed literature on system simulation and modeling of solar absorption chillers.
Publication Application Building simulation 1-effect based 2-effect based 3-effect based

Li and Sumathy [96] Air-conditioning ✓


Atmaca and Yigit [129] Air-conditioning ✓
Kim and Park [130] Refrigeration ✓
Florides et al. [100,145] Air-conditioning ✓ ✓
Joudi and Abdul-Ghafour [98] Air-conditioning ✓ ✓
Assilzadeh et al. [118] Air-conditioning ✓ ✓
Mazloumi et al. [146] Air-conditioning ✓ ✓
Baniyounes et al. [116] Air-conditioning ✓ ✓
Reda et al. [89] Air-conditioning ✓ ✓
Balghouthi et al. [157] Air-conditioning ✓ ✓
Martinez et al. [117] Air-conditioning ✓ ✓
Gomri [44] Air-conditioning ✓ ✓
Mammoli et al. [147] Air-conditioning ✓ ✓
Kaushik and Arora [153] Air-conditioning ✓ ✓
Tierney [155] Air-conditioning ✓
Qu et al. [156] Air-conditioning ✓ ✓
Balghouthi et al. [157] Air-conditioning ✓ ✓
Calise [158] Air-conditioning ✓ ✓
Chemisana et al. [159] Air-conditioning ✓ ✓
Li et al. [160,161] Air-conditioning ✓
Agrawal et al. [160] Refrigeration ✓
Ratlamwala et al. [166] Cooling ✓
Shirazi et al. [66] Air-conditioning ✓ ✓ ✓ ✓
Pintaldi et al. [167] Cooling ✓ ✓

chiller for a typical commercial office building in Sydney, Australia. A study on a 4.7 kW single-effect chiller coupled with 38 m2 flat plate
triple-effect absorption chiller was used and the required solar thermal collectors and a partitioned hot-water storage tank, and they compared
heat was supplied by parabolic trough collectors. The overall energy the performance results of the plant with traditional whole-tank mode
performance of the proposed systems was investigated under constant systems. Darkwa et al. [173] performed an experimental analysis of
and variable cooling load scenarios. The results indicated that the latent LiBr-H2O single-effect chiller powered by flat plate and evacuated tube
storage led to higher storage efficiencies due to the reduced storage collectors and found that a COP of ∼0.7 was achieved when the sup-
volume levels. However, the solar field thermal yield of the system with plied heat from the solar field was at 96 °C. A solar-powered single-
a sensible storage medium was found to be higher than the one with the effect absorption cooling system consisting of a 75 kW chiller and flat
latent storage medium. This was due to the losses attributed to the heat plate collectors with an aperture area of 160 m2 was tested by Rosiek
transfer between the heat transfer fluid (HTF) and the PCM as well as and Batlles [174] at the Solar Energy Research Center of Spain. The
limitations in the heat transfer rate. The authors also found that im- experimental results obtained from the operation of the plant during
proper insulation choices can result in severe heat losses at the storage summertime showed that the prototype was satisfactorily able to meet
tank. the cooling demand of the facility. Said et al. [175] conducted a si-
Finally, the summary of the reviewed literature on system simula- mulation and experimental study on a modified design of a solar-driven
tion and modeling of solar absorption chillers is presented in Table 5. single-effect aqua-ammonia absorption chiller system, attempting to
increase the overall COP of the chiller. The modification included the
recovery of waste heat from dephlegmator and the use of a refrigerant
4.4. Experimental studies
storage unit. The plant was installed and tested for two days, and the
obtained experimental data was used to analyze the overall perfor-
In addition to theoretical simulation and modeling of solar-powered
mance of the system. The results suggested an improvement in the COP
absorption chillers, a number of experimental studies have been carried
of the chiller by 10% and 8% as a result of dephlegmator heat recovery
out to test these systems with different design configurations under
and refrigerant storage unit, respectively. Recently, Marc et al. [176]
practical operating conditions [168]. Although there are some solar
conducted an experimental study of a 30 kW single-effect LiBr-H2O
absorption cooling systems in capacities of up to several hundred
absorption chiller fed by low-temperature solar collectors without any
kilowatts, most of the installations are still at experimental and de-
backup systems. The prototype was installed in Reunion Island and
monstration level (i.e. constructed on a small scale) [28]. Currently,
tested to check the accuracy and the reliability of their simulation
solar-driven absorption chillers have attracted significant attention in
model. The authors finally optimized the thermal COP of the chiller at
Europe − particularly in Germany, Spain, Italy and Greece – but China
different levels of refrigerating capacities. Overall, these existing ex-
(and broader Asia) are also taking up this technology and pioneering in
perimental studies show that solar-driven single-effect absorption chil-
this field [150,169,170]. Most solar cooling installations to date have
lers can operate using a heat source with a driving temperature ranging
been based on single-effect chillers and low-temperature solar thermal
from 70 to 100 °C, while COP levels below the rated value (i.e. ∼0.7)
collectors, while implementation of high-temperature solar cooling
can be achieved under partial load conditions.
plants using multi-effect absorption chillers is still infrequent [8,26,91].
Xu et al. [177] evaluated the performance of a solar-assisted 50 kW
Nakahara et al. [171] tested a 7 kW single-effect LiBr-H2O absorp-
variable-effect LiBr-H2O absorption chiller, where the chiller was de-
tion chiller driven by a 32.2 m2 array of flat plate solar collectors. A
signed in a novel way that could shift from the single- to double-effect
2.5 m3 hot water storage tank was used to store the solar-derived
cycle and vice versa, aiming to efficiently utilize solar energy with
thermal energy. The setup was designed to collect data to be used for
variable temperature. The authors analyzed the effects of chilled and
better design of large-scale solar absorption chillers. The experimental
cooling water temperature as well as the frequency of high-pressure
results showed that depending on the required cooling load, the COP of
absorber (HA) valve, generation pump, and hot water pump. The results
the chiller and the generator temperature varied from 0.4 to 0.8 and
showed that the performance of the chiller is enhanced at higher chilled
70 °C to 100 °C, respectively. Li and Sumathy [172] conducted a similar

74
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

water temperatures, low cooling water temperatures, while the HA 4.5. Economic and environmental assessments
valve and pumps operated at their optimized respective frequencies.
The results also indicated that the chiller COP increased from 0.69 to Many studies in the literature have investigated the performance of
1.08 under generation temperature from 95 °C to 120 °C as a result of solar absorption chillers only from an energetic perspective without an
the chiller shifting from the single-effect to the double-effect cycle. economic evaluation to determine whether or not it makes sense to
Bermejo et al. [178] developed a 174 kW gas/solar-powered double- actually build such plants [16,182]. Boopathi Raja and Shanmugam
effect LiBr-H2O absorption chiller using 352 m2 of linear concentrating [183] reviewed past theoretical and experimental investigations on
Fresnel collectors in Spain. The experimental investigation was focused solar cooling systems based on LiBr-H2O single-effect absorption chil-
on the solar collector size, piping heat losses, and the operation control. lers and low-temperature solar thermal collectors at residential scales
Their experimental results on a sunny representative day indicated that and suggested a number of ways to minimize the investment and op-
75% of the total heat input to the chiller was covered by solar energy, erating costs of these systems. Desideri et al. [184] conducted a similar
while the daily average efficiency of the collector and COP of the chiller study to determine technical solutions to achieve higher energy effi-
was reported to be ∼0.37 and ∼1.2, respectively. A similar study was ciency and lower costs in design of low-temperature solar absorption
conducted by Hang et al. [179], where a solar cooling system based on cooling systems for industrial refrigeration and air-conditioning appli-
a 23 kW double-effect absorption chiller and 54 m2 external compound cations. Al-Alili et al. [104] conducted a thermal and economic analysis
parabolic collectors was built and tested at the University of California, of solar-driven single-effect absorption cycles in order to assess their
Merced, USA. The authors reported a detailed energy performance of performance under Abu Dhabi’s climate. The results showed that the
the plant based on the collected experimental data. The results showed proposed system consumed 47% less electrical energy than the con-
that system could provide adequate cooling for the test facility between ventional vapor compression cycles, and the collector area was found to
11 AM and 5 PM on both sunny and cloudy days. The average daily be the key parameter impacting the payback time of initial investment
efficiency of the solar collectors and the COP of the chiller were around costs. The authors did not take into account the cost associated with
36–39% and ∼1, respectively, while the annual solar fraction of the equipment installation, integration, and piping (which can be even
plant was predicted to be 55–68%. Recently, Buonomano et al. [83] higher than the purchased cost of equipment) when performing the
carried out numerical and experimental analyses of a solar double-ef- economic analysis of the system. Eicker et al. [10,185], in two research
fect absorption cooling system using EFPC units and investigated the studies, analyzed the energy efficiency performance and cooling costs
technical and economic feasibility of the plant. The pilot system was of solar PV and absorption cooling systems for office buildings in sev-
constructed and tested in Saudi Arabia, and experimental data was eral climates worldwide. The results showed that the plant primary
collected over the course of 6 months of operation. The results showed energy use can be reduced by 40 to 70%, depending on system design
that the rated energy performance of the prototype solar cooling system and cooling load data. They also found that in order to achieve payback
is higher than that of a reference system which was based on con- periods of about 10 years with today's energy prices, the investment
centrating solar thermal collectors. The authors also found that the costs have to be reduced by 30–70%, depending on the location and
proposed prototype could achieve an acceptable profitability, provided dimensioning. A similar study was conducted by Mokhtar et al. [181]
that incentives were appropriately designed. for commercial buildings in UAE. They concluded that the cost of solar
Many researchers and industry from all around the world have also collectors and the performance of the refrigeration technologies are the
been working on solar absorption cooling systems, especially those two most significant parameters affecting solar cooling plant costs. A
using multi-effect absorption chillers driven by high temperature solar very similar study was conducted by Bellos and Tzivanidis [186], in
thermal collectors. A description of solar absorption cooling systems that it analyzed the energetic and economic performance of a solar
installed in around the world has been briefly presented in Table 6. It single-effect absorption chiller under various climates. The authors fi-
should be mentioned that to the author’s knowledge there has been no nally presented a general guideline for design of solar absorption chil-
published literature on experimentation of solar-powered triple-effect lers. Arsalis et al. [187] developed a mathematical model to simulate a
absorption chillers for air-conditioning applications. single-effect LiBr-H2O absorption chiller coupled with a FPC-based
array to satisfy the load demand of a single-family residential house in

Table 6
List of solar-assisted absorption chiller systems installed in around the world [150,180,181].
Location Cooling capacity Solar field size Chiller type Collector type Application Year

2
Ipswich, Australia 300 kW 574 m double-effect PTC Hospital 2007
Padstow, Australia 175 kW 165 m2 double-effect PTC Commercial building 2007
Newcastle, Australia 18 kW 50 m2 single-effect PTC Demonstration 2007
Echuca, Australia 200 kW 400 m2 double-effect ETC Hospital 2010
Alice Springs, Australia 230 kW 630 m2 double-effect PTC Art gallery 2011
Sydney, Australia 250 kW 500 m2 single-effect ETC Office building 2011
Charlestown, Australia 230 kW 350 m2 double-effect PTC Shopping center 2011
Sydney, Australia 19 kW 115 m2 single-effect PTC University 2014
Newcastle, Australia 10 kW 28 m2 double-effect LFR Demonstration 2016
Echuca, Australia – – – Micro LFR – Planned
Dammam, Saudi Arabia 23 kW 52.5 m2 double-effect EFPC Demonstration 2014
Dalaman/Turkey 140 kW 360 m2 double-effect PTC Park 2004
Alanya/Turkey 150 kW 360 m2 double-effect PTC Grand Kaptan 2005
Istanbul/Turkey 160 kW 324 m2 double-effect PTC Research Institute 2008
Antalya/Turkey 300 kW 423 m2 double-effect PTC Supermarket 2008
Dead Sea/Jordan 13 kW 126 m2 single-effect ETC Hotel 2010
Casablanca/Morocco 13 kW 108 m2 single-effect FPC Hospital 2010
Firenze/Italy 17 kW 108 m2 single-effect ETC Health facility 2010
Seville/Spain 174 kW 354 m2 double-effect PTC University 2010
Abu Dhabi/UAE 175 kW 334 m2 double-effect PTC Office 2010
Doha/Qatar 750 kW 1408 m2 double-effect PTC Stadium 2010
Island/USA 533 kW 315 m2 double-effect PTC Factory 2009

75
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

Nicosia, Cyprus. Through a series of parametric studies, the authors using a biomass fueled heater as an auxiliary heat source. The results
determined a set of design variables (i.e. collector area, slope, and the suggested that the optimal design of the solar-assisted plant led to a
storage tank volumetric capacity) which led to the minimum annual 20% reduction in energy consumption and carbon emissions compared
cost of the system. They concluded that the collector capital cost must to a conventional system. A similar study was carried out by Tsoutsos
be reduced to $360 per unit area of the collector in order for the et al. [196] to analyze the environmental benefits associated with a
modeled system to economically compete with conventional elec- solar-assisted single-effect absorption chiller to cover a fraction of the
trically-driven vapor compression heat pumps. Eicker et al. [188] in- total annual load demand of a hospital in Greece. Their results sug-
vestigated the primary energy use and economic performance of a solar- gested that the plant led to a 58.7% saving in the primary energy use as
assisted single-effect absorption chiller in several climatic regions in the compared to a conventional HVAC system. Moreover, Hang et al. [95]
world. The authors found that the unit cost of cooling for such systems carried out a parametric study to assess the energetic, economic, and
in European locations with limited operating hours strongly depended environmental prospects of a single-effect LiBr-H2O absorption chiller
on proper sizing of the system. The results suggested that under-sizing equipped with a gas-fired auxiliary heater to provide the cooling re-
the capacity of the chiller by 30–40% only slightly modified the solar quirements of a medium-sized office building in Los Angeles, California.
fraction of the plant in many regions, but significantly increased the The authors obtained the optimal design of the plant using the cost of
chiller’s operating hours and thus improved the economics of the CO2 emission reduction as an economic indicator. Recently, the en-
system. Eicker also [55] conducted one case study to compare the ergetic, economic, and environmental analyses of four configurations of
techno-economic performance of solar PV cooling systems with solar SHC systems based on coupling evacuated tube collectors with a single-
single- and multi-effect absorption chillers to supply cooling to a large effect LiBr-H2O absorption chiller have been carried out by Shirazi et al.
office building in Cairo, Egypt. In this work, ETCs were used to supply [88]. In the first configuration, a gas-fired heater was used as the
heat to the single-effect chiller, and linear concentrating Fresnel col- backup system, while a mechanical compression chiller was employed
lectors were used to drive the double- and triple-effect chillers. No hot as the auxiliary cooling system in the second configuration. The third
water storage was considered for the triple-effect chiller configuration. and fourth configurations were similar to the second configuration, but
Eicker varied the storage tank volume and the chiller capacity to de- the absorption chiller size was reduced to 50% and 20%, respectively.
termine the optimal size of each configuration, but no information was The results showed that the highest primary energy saving was
given about how the optimal size of these plants was obtained. In ad- achieved by the second configuration using the mechanical chiller as a
dition, one should note that in a parametric study, the effect of varia- backup system. The authors found that the economic performance of all
tion of one parameter on the system performance is evaluated at a time, configurations was still unsatisfactory (without subsidies) due to their
while the rest of design parameters are fixed – a drawback which can high capital costs. However, if a government subsidy of 50% was con-
result in local optimal designs. It was also assumed that the additional sidered, the undersized configurations can be economically feasible.
cooling load beyond the absorption chillers capacity was provided by a
mechanical compression chiller, while an auxiliary heater was also used 4.6. System optimization
at the same time for the double- and triple-effect chiller configurations.
Given the high capital costs of solar absorption chillers, the choice of The studies cited above have mainly performed a parametric study
two backup systems can negatively affect the economic performance of on various configurations of solar-driven absorption chillers, where
these systems. While a constant speed pump with an on/off controller only one design parameter was varied at a time while the rest of
was used in the solar field, the author did not report how the collectors parameters were fixed. The main drawback to this approach is that
performed during low solar radiation periods to achieve desired tem- design parameters can be trapped in local optimum regions, with the
peratures, which need to reach > 250 °C for the triple-effect chiller (a trend changing if any of the fixed variables are altered [197]. Thus, this
Kawasaki Sigma Ace model). The results of the energy analysis showed necessitates the implementation of a better optimization approach to
that the triple-effect chiller had the highest primary energy ratio, while obtain a better optimal design of solar absorption chiller systems. There
the PV-based system led to the lowest primary energy savings. It was have been several studies to develop tools for optimizing the perfor-
also found that the cost of annual cooling in all solar cooling options mance of solar absorption cooling cycles and therefore overcome the
was higher than that of a mechanical compression chiller due to the drawbacks associated with parametric studies in this field. Calise et al.
heavily subsidized energy prices in Egypt. Recently, Bellos et al. [189] [198–200] performed a single-objective optimization process on solar-
assessed the energetic, exergetic and economic performance of a assisted single- and double-effect absorption chillers to determine their
100 kW solar single-effect absorption chiller in Athens, Greece. A optimal economic performance. The system total cost or simple pay-
parametric study of the system was conducted by the authors to achieve back period was selected as an objective function which was mini-
the highest system exergetic efficiency levels. A cost analysis of the mized. The authors used the TRNOPT optimization program, a TRNSYS
optimal cases was carried out to determine the system payback time. component, which linked the TRNSYS simulation with an optimization
The results indicated that such systems would be economically attrac- algorithm called GenOpt capable of performing single-objective opti-
tive only when the environmental factors are taken into account. The mization problems. In these studies, the authors used a simple economic
results also suggested that with a decrease in the system capital costs model, where the cost of equipment installation, integration, and piping
and an increase in the electricity price, this technology can become and the escalation rate of expenditure and fuel price were not taken into
economically viable in the near future, especially in countries with high account. Hang et al. [94,121] modeled and optimized the design of a
solar energy potential. solar-assisted double-effect LiBr-H2O absorption chiller in small- and
In addition to the economics of solar-powered absorption chillers, medium-sized office buildings. The authors applied a linear regression
the emissions produced by these systems should also be taken into ac- analysis to a series of data points obtained from the parametric study of
count, as environmental concerns with regards to energy systems are the plant, yielding three equations representing the present worth cost,
accelerating [188,190–193]. Bukoski et al. [194] developed a life cycle life cycle energy, and life cycle carbon dioxide emissions of the plant as
assessment method to study the environmental impacts of a solar- functions of the system key parameters. Assigning different weights to
powered absorption chiller compared to a conventional air-con- these equations and combining them into one equation, a single-ob-
ditioning system in a stadium in Bangkok, Thailand. The results showed jective optimization problem was conducted to determine the optimal
that solar absorption chillers are environmentally advantageous due to design of the plant. Since the objectives were formulated through a
significantly lower electricity consumption relative to conventional typical parametric study, this optimization model could also be vul-
chillers. Ghaith and Abusitta [195] evaluated the thermal and en- nerable to being trapped in local optimum zones. Al-Alili et al. [201]
vironmental performance of a solar single-effect absorption chiller carried out a similar study to determine the optimal design of a solar-

76
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

powered 10 kW single-effect H2O-NH3 absorption chiller designed tariff in Spain. The authors, however, did not investigate how the op-
under Abu Dhabi’s climate. Two single-objective optimization problems timal design of the modeled system would compare against that of solar
were formulated and run in parallel to minimize the energy use of the multi-effect absorption chillers. Recently, Xu et al. [212] performed a
electrically-driven auxiliary heater and the total cost of the system se- multi-objective optimization of a solar absorption chiller in a medium-
parately. The results showed potential improvements of the plant per- sized office building in Atlanta, USA. The authors implemented a sto-
formance as compared to that of a conventional base system. Recently, chastic model to incorporate the uncertainties into the optimal design
Jing et al. [202] carried out an exergo-economic optimization of a of the plant and to validate the use of deterministic optimization ap-
solar-driven, single-effect absorption chiller in low-rise buildings to proach. They suggested that the deterministic optimization is sufficient
achieve a cost-effective and energy efficient design of the cooling ca- if the designer wishes to improve the average performance of the
pacity in absorption sub-systems. The authors reported that the col- system more than to control the uncertainties. Stochastic optimization
lector area should be determined by the optimal cooling capacity of the was recommended when the designer is extremely risk-averse. The
absorption chiller, not vice versa. The authors also found that the op- optimal results showed that the solar-assisted plant could cost between
timal cooling capacity of the absorption subsystem strongly depends on 60% and 120% of a conventional system, leading to 45% to 75% re-
solar irradiance and cooling demands. It was also found that the op- duction in the plant energy use. Recently, a systematic simulation-
timal size of the chiller should be designed according to the minimum based, multi-objective optimization of commercially available LiBr-H2O
product cost flow rates or the lowest relative cost difference, with the single-, double- and triple-effect absorption chillers powered by ETCs,
latter being sensitive to the local mean solar irradiance. EFPCs, and PTCs was conducted by Shirazi et al. [213]. A simulation
Multi-objective optimization methods offer an efficient framework model of these three configurations was developed in the TRNSYS 17
to simultaneously deal with conflicting objectives (e.g. cost and effi- environment. A combined energy, economic, and environmental ana-
ciency) in an optimization problem. Several studies have employed a lysis of the modeled systems was carried out to calculate the primary
multi-objective optimization approach in the design of various energy energy use as well as the levelized total annual cost of each plant, which
systems [203–208], but very few studies have applied this approach to were considered as two conflicting objective functions. The perfor-
solar-assisted absorption chillers. Indeed, since the optimal design of a mance of the proposed systems at their optimal was compared to that of
stand-alone absorption machine does not necessarily lead to an op- a reference conventional system. The authors also carried out a sensi-
timum performance of the whole solar-assisted absorption chiller tivity analysis to assess the influence of fuel cost, capital cost of in-
system, it is therefore essential to take a holistic optimization approach. novative components, and the annual interest rate on the Pareto front of
Iranmanesh and Mehrabian [209] carried out a multi-objective opti- optimal solutions. Overall, the optimization results revealed that of the
mization study on a double-effect absorption chiller coupled with proposed configurations, the SHC double-effect chiller has the best
evacuated tube collectors. The auxiliary energy consumption and the trade-off between the energetic, economic and environmental perfor-
net profit obtained from the solar-derived energy were considered as mance of the system, having a total cost of ∼0.7–0.9 M$ per year and
objectives for system optimization, while the environmental aspect of reducing the annual primary energy use and CO2 emissions by
the system was not taken into account. It was concluded that optimum 44.5–53.8% and 49.1–58.2% respectively (relative to the reference
mass flow rates had an important role in reducing the auxiliary energy. conventional system). The authors found that with the high capital cost
Since ETCs are not usually suggested for high temperature applications, associated with these systems, government subsidies and incentives are
due to high heat losses, they would have had a negative impact on the still required in order for this technology to achieve satisfactory pay-
size of the solar field and – consequently – on the economics of the back times and become cost-competitive with conventional HVAC
proposed system. Gebreslassie et al. [210] developed a mathematical systems.
model of solar-assisted single-effect absorption cooling system and Finally, Table 7 lists the summary of the reviewed literature on
conducted an optimization process to minimize the system total cost system-level optimization of solar absorption chillers.
and environmental impacts over the life cycle of the plant. The authors
only focused on industrial applications by assuming constant cooling 5. Challenges and pathways for development
demand, so the model was based only on a monthly simulation. In
another study, Gebreslassie et al. [211] developed an optimization With only ∼2000 installation worldwide as of 2017, there are still
model combined with life cycle assessment principles to optimize the several areas of techno-economic improvement needed before solar-
performance of a single-effect H2O-NH3 absorption chiller system from driven absorption systems can truly compete with conventional HVAC
economic and environmental aspects in Barcelona. No building model on a global scale [214]. Based on the literature above, the authors
was considered in their study, assuming the end user required a con- would like to highlight the following key areas for future research and
stant cooling load throughout the year. The total cost of the system and development, noting that much of the development is likely to be di-
the amount of CO2 emissions released into the atmosphere were treated rected towards system components and design/operation of the system,
as two objective functions and were minimized. The results suggested rather than the chillers themselves:
that considering government subsidies on solar technologies, the
modeled system could be economically appealing under the energy • Backup systems: The use of gas-fired backup in solar single-effect
Table 7
Summary of the literature which provides system-level optimization of solar absorption chillers.
Publication Single-objective optimization Multi-objective optimization

1-effect 2-effect 3-effect 1-effect 2-effect 3-effect

Calise et al. [198–200] ✓ ✓


Hang et al. [94,121] ✓
Al-Alili et al. [201] ✓
Iranmanesh and Mehrabian [209] ✓
Gebreslassie et al. [210] ✓
Gebreslassie et al. [211] ✓
Xu et al. [212] ✓
Shirazi et al. [213] ✓ ✓ ✓

77
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

chillers is inefficient due to its very low primary energy savings. Gas based on single-effect chillers and low-temperature solar thermal col-
backup system should ONLY be considered for double or triple-ef- lectors, while less emphasis has been placed on the combination of
fect chillers, in order to avoid a negative primary energy balance. high-temperature solar thermal collectors and multi-effect absorption
Backup cooling, instead, is recommended for solar single-effect chillers, especially triple-effect chillers. It is concluded while solar-
chillers to achieve an acceptable primary energy savings. powered absorption chillers can be promising in mitigating environ-
• Energy storage: The storage tank represents a major potential source mental footprints, their economic performance needs to improve sig-
of solar (and auxiliary) input heat losses. This is particularly re- nificantly to penetrate the market successfully. In addition, much of the
levant for double and tipple-effect systems, where standard, un- literature to date has focused on the chiller and collector performance,
pressurized, hot water storage is not applicable (since the required but the next stage of development will likely require more emphasis on
temperature is above 100 °C). Therefore, efficient tank design, other system components and design, maintenance, and operation is-
which includes appropriate insulation, and detailed system design sues. Thus, the challenge for the solar industry is to reduce the whole
and analysis which account for tank losses is strongly suggested for lifetime cost of solar-powered absorption chiller systems in order to
future work. Left unchecked, parasitic losses can completely offset improve the economic viability of this technology, aiming to eventually
any advantage gained from the solar field. be competitive without subsidies.
• Other parasitic losses: Any given cross-section of the literature on
absorption cooling has a large fraction of studies which neglect Acknowledgements
parasitic losses from pumping energy and pipe heat loss. However,
special care should be taken to ensure the piping has sufficient in- This research is funded by the for Low Carbon Living Ltd (Project ID
sulation since a large temperature drop along the pipeline can result (RP1002)) supported by the Cooperative Research Centres program, an
in too low of a temperature to run the absorption chiller. Again, for Australian Government Initiative.
piping systems operating at > 100 °C, waterproof insulation is a ne-
cessity since any water that infiltrates the insulation (e.g. from rain) References
will be boiled off during operation, at the cost of ∼2200 kJ/kg!
• Location: Although obvious, the choice of location is very important [1] World Energy Resources: 2013 Survey, URL: < https://www.worldenergy.org/
publications/2013/world-energy-resources-2013-survey/ > ; 2013 [accessed June
for solar-driven absorption systems. In regions with very low solar
2016].
irradiation, where the heating demand is dominant, solar-driven [2] Kannan N, Vakeesan D. Solar energy for future world: – a review. Renew Sustain
multi-effect chillers are not an efficient option due to under-utili- Energy Rev 2016;62:1092–105.
zation of the high-temperature solar heat in summer. [3] Kumar SahuB. A study on global solar PV energy developments and policies with

• Heat rejection: In hot and humid regions the cooling tower has to
special focus on the top ten solar PV power producing countries. Renew Sustain
Energy Rev 2015;43:621–34.
deal with high ambient wet bulb temperatures. Thus, other design [4] Renewables 2015 global status report, URL: < http://www.ren21.net/wp-content/
adaptations or employing a completely different heat rejection uploads/2015/07/REN12-GSR2015_Onlinebook_low1.pdf > ; 2015 [accessed
June 2016].
system may be required in extreme climates.

[5] Zeyghami M, Goswami DY, Stefanakos E. A review of solar thermo-mechanical
Advanced system design: The design of solar double- and triple-effect refrigeration and cooling methods. Renew Sustain Energy Rev 2015;51:1428–45.
absorption chillers is more complex than a typical solar single-effect [6] Kohlenbach P, Jakob U. Solar cooling: the earthscan expert guide to solar cooling
systems. 1st ed. UK: Taylor & Francis Ltd; 2014.
chiller system. This results primarily from the high-temperature
[7] Henning H-M. Solar assisted air conditioning of buildings – an overview. Appl
operating conditions and pressurized hot water in the chiller hot Therm Eng 2007;27:1734–49.
water loop. Solar multi-effect absorption chillers also require regular [8] Nkwetta DN, Sandercock J. A state-of-the-art review of solar air-conditioning
maintenance. These systems also need to carefully consider storage systems. Renew Sustain Energy Rev 2016;60:1351–66.
[9] Henning H-M, Döll J. Solar systems for heating and cooling of buildings. Energy
mediums which are: (i) simple and cost-effective and (ii) have Proc 2012;30:633–53.
minimized heat losses (as above). [10] Eicker U, Pietruschka D, Schmitt A, Haag M. Comparison of photovoltaic and solar

• Cost: The capital cost of solar field and absorption chiller are both thermal cooling systems for office buildings in different climates. Sol Energy
2015;118:243–55.
critical to the economic performance of system. Public funding and [11] Jordehi AR. Parameter estimation of solar photovoltaic (PV) cells: a review. Renew
subsidies were found to be a critical requirement in order for solar Sustain Energy Rev 2016;61:354–71.
absorption chillers to achieve a satisfactory economic performance. [12] Sarbu I, Sebarchievici C. Review of solar refrigeration and cooling systems. Energy
Build 2013;67:286–97.
Mass production and the associated learning curves for these sys- [13] Huang B-J, Hou T-F, Hsu P-C, Lin T-H, Chen Y-T, Chen C-W, et al. Design of direct
tems might help tunnel through this cost barrier, but right now solar PV driven air conditioner. Renew Energy 2016;88:95–101.
solar-powered absorption chillers have long payback times. To [14] Li Y, Wang RZ. 10 – Photovoltaic-powered solar cooling systems. Advances in Solar
Heating and Cooling. Woodhead Publishing; 2016. p. 227–50.
overcome this challenge, if only a small level of government subsidy [15] Pintaldi S, Perfumo C, Sethuvenkatraman S, White S, Rosengarten G. A review of
is available, the solar field should be down-sized to achieve lower thermal energy storage technologies and control approaches for solar cooling.
solar fractions, thereby improving the economic viability of this Renew Sustain Energy Rev 2015;41:975–95.
[16] Otanicar T, Taylor RA, Phelan PE. Prospects for solar cooling – an economic and
technology.
environmental assessment. Sol Energy 2012;86:1287–99.
[17] Kohlenbach P, Dennis M. Solar cooling in Australia: the future of air-conditioning?
The challenge for the solar industry is to reduce the cost associated Ecolibrium: Aust Inst Refrig, Air Condition Heat (AIRAH) J 2010:32–8.
with solar field (including properly insulated piping/plumbing), which [18] Beccali M, Cellura M, Finocchiaro P, Guarino F, Longo S, Nocke B. Life cycle as-
sessment performance comparison of small solar thermal cooling systems with
could significantly improve the economic viability of solar absorption conventional plants assisted with photovoltaics. Energy Proc 2012;30:893–903.
chiller systems. [19] Kalkan N, Young EA, Celiktas A. Solar thermal air conditioning technology redu-
cing the footprint of solar thermal air conditioning. Renew Sustain Energy Rev
2012;16:6352–83.
6. Conclusions [20] Mujahid Rafique M, Gandhidasan P, Rehman S, Al-Hadhrami LM. A review on
desiccant based evaporative cooling systems. Renew Sustain Energy Rev
This paper presented a detailed literature review of the recent ad- 2015;45:145–59.
[21] Daou K, Wang RZ, Xia ZZ. Desiccant cooling air conditioning: a review. Renew
vances on solar-powered absorption chillers for air-conditioning ap- Sustain Energy Rev 2006;10:55–77.
plications. A wide range of topics including the background theory, [22] Chen J, Jarall S, Havtun H, Palm B. A review on versatile ejector applications in
system arrangement, control designs, system modeling and simulation, refrigeration systems. Renew Sustain Energy Rev 2015;49:67–90.
[23] Besagni G, Mereu R, Inzoli F. Ejector refrigeration: a comprehensive review.
experimental studies, energetic-economic-environmental (3E) assess- Renew Sustain Energy Rev 2016;53:373–407.
ments and system optimization were covered. The review showed that [24] Li XH, Hou XH, Zhang X, Yuan ZX. A review on development of adsorption
the majority of solar absorption chillers installed around the world are cooling—Novel beds and advanced cycles. Energy Convers Manage

78
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

2015;94:221–32. review. Renew Sustain Energy Rev 2015;49:192–210.


[25] Saha BB, El-Sharkawy II, Shahzad MW, Thu K, Ang L, Ng KC. Fundamental and [63] Real D, Johnston R, Lauer J, Schicho A, Hotz N. Novel non-concentrating solar
application aspects of adsorption cooling and desalination. Appl Therm Eng collector for intermediate-temperature energy capture. Sol Energy
2016;97:68–76. 2014;108:421–31.
[26] Allouhi A, Kousksou T, Jamil A, Bruel P, Mourad Y, Zeraouli Y. Solar driven [64] Ibrahim A, Othman MY, Ruslan MH, Mat S, Sopian K. Recent advances in flat plate
cooling systems: an updated review. Renew Sustain Energy Rev 2015;44:159–81. photovoltaic/thermal (PV/T) solar collectors. Renew Sustain Energy Rev
[27] Hassan HZ, Mohamad AA. A review on solar cold production through absorption 2011;15:352–65.
technology. Renew Sustain Energy Rev 2012;16:5331–48. [65] Tian Y, Zhao CY. A review of solar collectors and thermal energy storage in solar
[28] Zhai XQ, Qu M, Li Y, Wang RZ. A review for research and new design options of thermal applications. Appl Energy 2013;104:538–53.
solar absorption cooling systems. Renew Sustain Energy Rev 2011;15:4416–23. [66] Shirazi A, Taylor RA, White SD, Morrison GL. A systematic parametric study and
[29] Kim DS, Infante Ferreira CA. Solar refrigeration options – a state-of-the-art review. feasibility assessment of solar-assisted single-effect, double-effect, and triple-effect
Int J Refrig 2008;31:3–15. absorption chillers for heating and cooling applications. Energy Convers Manage
[30] Xu ZY, Wang RZ. 11 – Solar-powered absorption cooling systems. Advances in 2016;114:258–77.
Solar Heating and Cooling. Woodhead Publishing; 2016. p. 251–98. [67] Saleh A, Mosa M. Optimization study of a single-effect water–lithium bromide
[31] Herold KE, Radermacher R, Klein SA. Absorption chillers and heat pumps. 2nd ed. absorption refrigeration system powered by flat-plate collector in hot regions.
USA: CRC Press; 2016. Energy Convers Manage 2014;87:29–36.
[32] Gilani SI-U-H, Ahmed MSMS. Solution crystallization detection for double-effect [68] Bellos E, Tzivanidis C, Antonopoulos KA. Exergetic, energetic and financial eva-
LiBr-H2O steam absorption chiller. Energy Proc 2015;75:1522–8. luation of a solar driven absorption cooling system with various collector types.
[33] Liao X, Radermacher R. Absorption chiller crystallization control strategies for Appl Therm Eng 2016;102:749–59.
integrated cooling heating and power systems. Int J Refrig 2007;30:904–11. [69] Adam Solar Resources, URL: < http://www.adamsolarresources.com/
[34] Prasartkaew B. Performance test of a small size LiBr-H2O absorption chiller. solarthermal.html > ; 2016 [accessed March 2016].
Energy Proc 2014;56:487–97. [70] Duffie JA, Beckman WA. Solar engineering of thermal processes. 3rd ed. Hoboken,
[35] Florides GA, Kalogirou SA, Tassou SA, Wrobel LC. Design and construction of a NJ, USA: Wiley; 2006.
LiBr–water absorption machine. Energy Convers Manage 2003;44:2483–508. [71] Solar Sense, URL: < http://www.solarsense.co.za/ > ; 2016 [accessed March
[36] Sun J, Fu L, Zhang S. A review of working fluids of absorption cycles. Renew 2016].
Sustain Energy Rev 2012;16:1899–906. [72] Hossain MS, Saidur R, Fayaz H, Rahim NA, Islam MR, Ahamed JU, et al. Review on
[37] Ayala R, Heard CL, Holland FA. Ammonia/lithium nitrate absorption/compression solar water heater collector and thermal energy performance of circulating pipe.
refrigeration cycle. Part II. experimental. Appl Therm Eng 1998;18:661–70. Renew Sustain Energy Rev 2011;15:3801–12.
[38] Park Y, Kim J-S, Lee H. Physical properties of the lithium bromide + 1,3-propa- [73] NEP Solar Pty Ltd, URL: < http://www.nep-solar.com/ > ; 2016 [accessed June
nediol + water system. Int J Refrig 1997;20:319–25. 2016].
[39] Zhang X, Hu D. Performance simulation of the absorption chiller using water and [74] SOLITEM GmbH Ltd, URL: < www.solitem.de > ; 2016 [accessed December
ionic liquid 1-ethyl-3-methylimidazolium dimethylphosphate as the working pair. 2016].
Appl Therm Eng 2011;31:3316–21. [75] Home Power, URL: < http://www.homepower.com/ > ; 2016 [accessed March
[40] Li ZF, Sumathy K. Technology development in the solar absorption air-con- 2016].
ditioning systems. Renew Sustain Energy Rev 2000;4:267–93. [76] Facão J, Oliveira AC. Numerical simulation of a trapezoidal cavity receiver for a
[41] Emhofer J, Schneider M-C, Fleckl T, Dudzinski P, Monsberger M. Investigation and linear Fresnel solar collector concentrator. Renewable Energy 2011;36:90–6.
testing of new polymeric-based working pairs for absorption chillers. Sci Technol [77] Industrial Solar GmbH Ltd, URL: < http://www.industrial-solar.de/CMS/ > ;
Built Environ 2015;21:976–85. 2016 [accessed December 2016].
[42] Álvarez ME, Esteve X, Bourouis M. Performance analysis of a triple-effect ab- [78] Berger M, Weckesser M, Weber C, Döll J, Morgenstern A, Häberle A. Solar driven
sorption cooling cycle using aqueous (lithium, potassium, sodium) nitrate solution cold rooms for industrial cooling applications. Energy Proc 2012;30:904–11.
as a working pair. Appl Therm Eng 2015;79:27–36. [79] Chromasun Inc, URL: < http://www.chromasun.com/ > ; 2016[accessed June
[43] Gebreslassie BH, Medrano M, Boer D. Exergy analysis of multi-effect water–LiBr 2016].
absorption systems: from half to triple effect. Renew Energy 2010;35:1773–82. [80] Solar Euromed, URL: < http://www.solareuromed.com/ > ; 2016 [accessed
[44] Gomri R. Simulation study on the performance of solar/natural gas absorption March 2016].
cooling chillers. Energy Convers Manage 2013;65:675–81. [81] TVP Solar Inc, URL: < http://www.tvpsolar.com/ > ; 2016 [accessed March
[45] Fan Y, Luo L, Souyri B. Review of solar sorption refrigeration technologies: de- 2016].
velopment and applications. Renew Sustain Energy Rev 2007;11:1758–75. [82] Buonomano A, Calise F, Palombo A, Vicidomini M. Energy and economic analysis
[46] Gomri R, Hakimi R. Second law analysis of double effect vapour absorption cooler of geothermal–solar trigeneration systems: a case study for a hotel building in
system. Energy Convers Manage 2008;49:3343–8. Ischia. Appl Energy 2015;138:224–41.
[47] Gomri R. Investigation of the potential of application of single effect and multiple [83] Buonomano A, Calise F, d’Accadia MD, Ferruzzi G, Frascogna S, Palombo A, et al.
effect absorption cooling systems. Energy Convers Manage 2010;51:1629–36. Experimental analysis and dynamic simulation of a novel high-temperature solar
[48] Thermax Ltd, URL: < http://www.thermaxglobal.com/ > ; 2016 [accessed March cooling system. Energy Convers Manage 2016;109:19–39.
2016]. [84] Chidambaram LA, Ramana AS, Kamaraj G, Velraj R. Review of solar cooling
[49] Kawasaki Thermal Engineering Ltd., URL: < http://www.khi.co.jp/corp/kte/EN/ methods and thermal storage options. Renew Sustain Energy Rev 2011;15:3220–8.
> ; 2015 [accessed September 2015]. [85] Ibrahim NI, Khan MMA, Mahbubul IM, Saidur R, Al-Sulaiman FA. Experimental
[50] BROAD X Non-electric Chiller: Model Selection & Design Manual, URL: < http:// testing of the performance of a solar absorption cooling system assisted with ice-
www.broadusa.com/ > ; 2008 [accessed March 2016]. storage for an office space. Energy Convers Manage 2017;148:1399–408.
[51] York® Absorption Chillers, URL: < http://www.york.com/ > ; 2016 [accessed [86] Henning H-M, Motta M, Mugnier D. Solar cooling handbook: a guide to solar as-
June 2016]. sisted cooling and dehumidification processes. Berlin, Germany: De Gruyter; 2013.
[52] Trane® Absorption Liquid Chillers, URL: < http://www.trane.com/ > ; 2016 [ac- [87] Gil A, Oró E, Peiró G, Álvarez S, Cabeza LF. Material selection and testing for
cessed June 2016]. thermal energy storage in solar cooling. Renew Energy 2013;57:366–71.
[53] Kroger D. Air-cooled heat exchangers and cooling towers: thermal-flow perfor- [88] Shirazi A, Taylor RA, White SD, Morrison GL. Transient simulation and parametric
mance evaluation and design. Tulsa, USA: PennWell Corporation; 2004. study of solar-assisted heating and cooling absorption systems: an energetic, eco-
[54] IEA SHC Task 38 Solar Air-Conditioning and Refrigeration: Subtask C, Report C5: nomic and environmental (3E) assessment. Renew Energy 2016;86:955–71.
Heat rejection, URL: < http://archive.iea-shc.org/publications/downloads/IEA- [89] Reda F, Viot M, Sipilä K, Helm M. Energy assessment of solar cooling thermally
Task38-Report_C5_Heat%20rejection.pdf > ; 2016 [accessed December 2016]. driven system configurations for an office building in a Nordic country. Appl
[55] Eicker U. Energy efficient buildings with solar and geothermal resources. West Energy 2016;166:27–43.
Sussex, UK: John Wiley and Sons; 2014. [90] Bataineh K, Taamneh Y. Review and recent improvements of solar sorption cooling
[56] Izquierdo M, Lizarte R, Marcos JD, Gutiérrez G. Air conditioning using an air- systems. Energy Build 2016;128:22–37.
cooled single effect lithium bromide absorption chiller: results of a trial conducted [91] Siddiqui MU, Said SAM. A review of solar powered absorption systems. Renew
in Madrid in August 2005. Appl Therm Eng 2008;28:1074–81. Sustain Energy Rev 2015;42:93–115.
[57] Helm M, Hagel K, Pfeffer W, Hiebler S, Schweigler C. Solar heating and cooling [92] Sarbu I, Sebarchievici C. General review of solar-powered closed sorption re-
system with absorption chiller and latent heat storage – a research project sum- frigeration systems. Energy Convers Manage 2015;105:403–22.
mary. Energy Proc 2014;48:837–49. [93] Al-Alili A, Hwang Y, Radermacher R. Review of solar thermal air conditioning
[58] Eicker U, Pietruschka D, Pesch R. Heat rejection and primary energy efficiency of technologies. Int J Refrig 2014;39:4–22.
solar driven absorption cooling systems. Int J Refrig 2012;35:729–38. [94] Hang Y, Qu M, Ukkusuri S. Optimizing the design of a solar cooling system using
[59] Mugnier D, Neyer D, White SD. The solar cooling design guide case studies of central composite design techniques. Energy Build 2011;43:988–94.
successful solar air conditioning design: case studies of successful solar air con- [95] Hang Y, Qu M, Zhao F. Economical and environmental assessment of an optimized
ditioning design. USA: Wiley: Ernst and Sohn; 2017. solar cooling system for a medium-sized benchmark office building in Los Angeles,
[60] IEA SHC Task 38 Solar Air-Conditioning and Refrigeration: Subtask C, Description California. Renew Energy. 2011;36:648–58.
of simulation tools used in solar cooling, URL: < http://task38.iea-shc.org/data/ [96] Li ZF, Sumathy K. Simulation of a solar absorption air conditioning system. Energy
sites/1/publications/Report_C2A_Simualtion-tools-in-solar-cooling.pdf > ; 2016 Convers Manage 2001;42:313–27.
[accessed December 2016]. [97] Blinn JC, Mitchell JW, Duffie JA. Modeling of transient performance of residential
[61] Kalogirou SA. Solar thermal collectors and applications. Prog Energy Combust Sci solar air-conditioning systems. In: Proceedings of the Silver Jubilee Congress.
2004;30:231–95. Atlanta, Georgia; 1979. p. 705–9.
[62] Suman S, Khan MK, Pathak M. Performance enhancement of solar collectors—a [98] Joudi KA, Abdul-Ghafour QJ. Development of design charts for solar cooling

79
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

systems. Part I: computer simulation for a solar cooling system and development of [133] PolySun Solar-Calculator, URL: < http://www.polysunonline.com/PsoPublic/
solar cooling design charts. Energy Convers Manage 2003;44:313–39. app/home/access > ; 2016 [accessed December 2016].
[99] Alva LH, González JE. Simulation of an air-cooled solar-assisted absorption air [134] Meteotest Ltd, URL: < http://www.meteonorm.com/ > ; 2016 [accessed
conditioning system. J Sol Energy Eng 2002;124:276–82. December 2016].
[100] Florides GA, Kalogirou SA, Tassou SA, Wrobel LC. Modelling and simulation of an [135] TRNSYS 17: a transient system simulation program. Madison, USA: Solar Energy
absorption solar cooling system for Cyprus. Sol Energy 2002;72:43–51. Laboratory, University of Wisconsin.
[101] Ward DS, Smith CC, Ward JC. Operational modes of solar heating and cooling [136] TRNSYS Standard Library, Volume 8: Weather Data.
systems. Sol Energy 1977;19:55–61. [137] National Solar Radiation Data Base (NSRDB), URL: < http://rredc.nrel.gov/solar/
[102] Zambrano D, Bordons C, Garcia-Gabin W, Camacho EF. Model development and old_data/nsrdb/ > ; 2016 [accessed December 2016].
validation of a solar cooling plant. Int J Refrig 2008;31:315–27. [138] Kaynakli O, Kilic M. Theoretical study on the effect of operating conditions on
[103] Argiriou AA, Balaras CA, Kontoyiannidis S, Michel E. Numerical simulation and performance of absorption refrigeration system. Energy Convers Manage
performance assessment of a low capacity solar assisted absorption heat pump 2007;48:599–607.
coupled with a sub-floor system. Sol Energy 2005;79:290–301. [139] Grossman G, Zaltash A. ABSIM—modular simulation of advanced absorption
[104] Al-Alili A, Islam MD, Kubo I, Hwang Y, Radermacher R. Modeling of a solar systems. Int J Refrig 2001;24:531–43.
powered absorption cycle for Abu Dhabi. Appl Energy 2012;93:160–7. [140] Kühn A, Ziegler F. Operational results of a 10 kW absorption chiller and adapta-
[105] Calise F, Dentice d’Accadia M, Palombo A. Transient analysis and energy opti- tion of the characteristic equation. In: Proceedings of the International Conference
mization of solar heating and cooling systems in various configurations. Sol Solar Air Conditioning. Bad Staffelstein, Germany; 2005. p. 6–7.
Energy 2010;84:432–49. [141] Puig-Arnavat M, López-Villada J, Bruno JC, Coronas A. Analysis and parameter
[106] Yeung MR, Yuen PK, Dunn A, Cornish LS. Performance of a solar-powered air identification for characteristic equations of single- and double-effect absorption
conditioning system in Hong Kong. Sol Energy 1992;48:309–19. chillers by means of multivariable regression. Int J Refrig 2010;33:70–8.
[107] Koeppel EA, Klein SA, Mitchell JW, Flake BA. Optimal supervisory control of an [142] Labus J, Bruno JC, Coronas A. Performance analysis of small capacity absorption
absorption chiller system. HVAC&R Res 1995;1:325–40. chillers by using different modeling methods. Appl Therm Eng 2013;58:305–13.
[108] Bujedo LA, Rodríguez J, Martínez PJ. Experimental results of different control [143] Ibrahim NI, Al-Sulaiman FA, Ani FN. Solar absorption systems with integrated
strategies in a solar air-conditioning system at part load. Sol Energy absorption energy storage–a review. Renew Sustain Energy Rev 2018;82:1602–10.
2011;85:1302–15. [144] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by ab-
[109] Y-j Xu, S-j Zhang, Y-h Xiao. Modeling the dynamic simulation and control of a sorption and adsorption chillers driven by stationary and concentrating photo-
single effect LiBr–H2O absorption chiller. Appl Therm Eng 2016;107:1183–91. voltaic/thermal solar collectors: modelling and simulation. Renew Sustain Energy
[110] Labus J, Hernández JA, Bruno JC, Coronas A. Inverse neural network based Rev 2018;82:1874–908.
control strategy for absorption chillers. Renewable Energy 2012;39:471–82. [145] Florides GA, Kalogirou SA, Tassou SA, Wrobel LC. Modelling, simulation and
[111] Albers J. New absorption chiller and control strategy for the solar assisted cooling warming impact assessment of a domestic-size absorption solar cooling system.
system at the German federal environment agency. Int J Refrig 2014;39:48–56. Appl Therm Eng 2002;22:1313–25.
[112] Palau A, Velo E, Puigjaner L. Use of neural networks and expert systems to control [146] Mazloumi M, Naghashzadegan M, Javaherdeh K. Simulation of solar lithium
a gas/solid sorption chilling machine: utilisation des réseaux neuronaux et des bromide–water absorption cooling system with parabolic trough collector. Energy
systèmes experts pour réguler une machine frigorifique à sorption gaz/solide. Int J Convers Manage 2008;49:2820–32.
Refrig 1999;22:59–66. [147] Mammoli A, Vorobieff P, Barsun H, Burnett R, Fisher D. Energetic, economic and
[113] Shin Y, Seo JA, Cho HW, Nam SC, Jeong JH. Simulation of dynamics and control of environmental performance of a solar-thermal-assisted HVAC system. Energy
a double-effect LiBr–H2O absorption chiller. Appl Therm Eng 2009;29:2718–25. Build 2010;42:1524–35.
[114] Gao M, Sun F-Z, Zhou S-J, Shi Y-T, Zhao Y-B, Wang N-H. Performance prediction [148] Sokhansefat T, Mohammadi D, Kasaeian A, Mahmoudi AR. Simulation and para-
of wet cooling tower using artificial neural network under cross-wind conditions. metric study of a 5-ton solar absorption cooling system in Tehran. Energy Convers
Int J Therm Sci 2009;48:583–9. Manage 2017;148:339–51.
[115] Hosoz M, Ertunc HM, Bulgurcu H. Performance prediction of a cooling tower using [149] Hirmiz R, Lightstone MF, Cotton JS. Performance enhancement of solar absorption
artificial neural network. Energy Convers Manage 2007;48:1349–59. cooling systems using thermal energy storage with phase change materials. Appl
[116] Baniyounes AM, Rasul MG, Khan MMK. Assessment of solar assisted air con- Energy 2018;223:11–29.
ditioning in Central Queensland's subtropical climate, Australia. Renew Energy [150] Ayadi O, Aprile M, Motta M. Solar cooling systems utilizing concentrating solar
2013;50:334–41. collectors – an overview. Energy Proc 2012;30:875–83.
[117] Martínez PJ, Martínez JC, Lucas M. Design and test results of a low-capacity solar [151] Liu YL, Wang RZ. Performance prediction of a solar/gas driving double effect
cooling system in Alicante (Spain). Sol Energy 2012;86:2950–60. LiBr–H2O absorption system. Renew Energy 2004;29:1677–95.
[118] Assilzadeh F, Kalogirou SA, Ali Y, Sopian K. Simulation and optimization of a LiBr [152] Li Q, Zheng C, Shirazi A, Bany Mousa O, Moscia F, Scott JA, et al. Design and
solar absorption cooling system with evacuated tube collectors. Renew Energy analysis of a medium-temperature, concentrated solar thermal collector for air-
2005;30:1143–59. conditioning applications. Appl Energy 2017;190:1159–73.
[119] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by CPVT [153] Kaushik SC, Arora A. Energy and exergy analysis of single effect and series flow
and ET solar collectors: a novel transient simulation model. Appl Energy double effect water–lithium bromide absorption refrigeration systems. Int J Refrig
2013;103:588–606. 2009;32:1247–58.
[120] Balghouthi M, Chahbani MH, Guizani A. Feasibility of solar absorption air con- [154] Cabrera FJ, Fernández-García A, Silva RMP, Pérez-García M. Use of parabolic
ditioning in Tunisia. Build Environ 2008;43:1459–70. trough solar collectors for solar refrigeration and air-conditioning applications.
[121] Hang Y, Du L, Qu M, Peeta S. Multi-objective optimization of integrated solar Renew Sustain Energy Rev 2013;20:103–18.
absorption cooling and heating systems for medium-sized office buildings. Renew [155] Tierney MJ. Options for solar-assisted refrigeration—trough collectors and double-
Energy 2013;52:67–78. effect chillers. Renew Energy 2007;32:183–99.
[122] Calise F, Palombo A, Vanoli L. Maximization of primary energy savings of solar [156] Qu M, Yin H, Archer DH. A solar thermal cooling and heating system for a
heating and cooling systems by transient simulations and computer design of ex- building: experimental and model based performance analysis and design. Sol
periments. Appl Energy 2010;87:524–40. Energy 2010;84:166–82.
[123] Petela K, Manfrida G, Szlek A. Advantages of variable driving temperature in solar [157] Balghouthi M, Chahbani MH, Guizani A. Investigation of a solar cooling installa-
absorption chiller. Renewable Energy 2017;114:716–24. tion in Tunisia. Appl Energy 2012;98:138–48.
[124] Ma Z, Wang S. Energy efficient control of variable speed pumps in complex [158] Calise F. High temperature solar heating and cooling systems for different
building central air-conditioning systems. Energy Build 2009;41:197–205. Mediterranean climates: dynamic simulation and economic assessment. Appl
[125] Sanaye S, Shirazi A. Thermo-economic optimization of an ice thermal energy Therm Eng 2012;32:108–24.
storage system for air-conditioning applications. Energy Build 2013;60:100–9. [159] Chemisana D, López-Villada J, Coronas A, Rosell JI, Lodi C. Building integration of
[126] Shirazi A, Pintaldi S, White SD, Morrison GL, Rosengarten G, Taylor RA. Solar- concentrating systems for solar cooling applications. Appl Therm Eng
assisted absorption air-conditioning systems in buildings: control strategies and 2013;50:1472–9.
operational modes. Appl Therm Eng 2016;92:246–60. [160] Agrawal SK, Kumar R, Khaliq A. First and second law investigations of a new solar-
[127] Khan MSA, Badar AW, Talha T, Khan MW, Butt FS. Configuration based modeling assisted thermodynamic cycle for triple effect refrigeration. Int J Energy Res
and performance analysis of single effect solar absorption cooling system in 2014;38:162–73.
TRNSYS. Energy Convers Manage 2018;157:351–63. [161] Li Z, Ye X, Liu J. Optimal temperature of collector for solar double effect LiBr/H2O
[128] Kohlenbach P, Ziegler F. A dynamic simulation model for transient absorption absorption cooling system in subtropical city based on a year round meteor-
chiller performance. Part I: The model. Int J Refrig 2008;31:217–25. ological data. Appl Therm Eng 2014;69:19–28.
[129] Atmaca I, Yigit A. Simulation of solar-powered absorption cooling system. Renew [162] Matsushima H, Fujii T, Komatsu T, Nishiguchi A. Dynamic simulation program
Energy 2003;28:1277–93. with object-oriented formulation for absorption chillers (modelling, verification,
[130] Kim B, Park J. Dynamic simulation of a single-effect ammonia–water absorption and application to triple-effect absorption chiller). Int J Refrig 2010;33:259–68.
chiller. Int J Refrig 2007;30:535–45. [163] Khamooshi M, Parham K, Egelioglu F, Yari M, Salati H. Simulation and optimi-
[131] INSEL Programming Language, URL: < http://www.insel.eu/index.php?id=73& zation of novel configurations of triple absorption heat transformers integrated to
L=1 > ; 2016 [accessed December 2016]. a water desalination system. Desalination 2014;348:39–48.
[132] IEA SHC Task 49 Solar Heat Integration in Industrial Processes: Subtask C, [164] Wang L, You S, Zhang H, Li X. Simulation of gas-fired triple-effect LiBr/water
Overview and description of simulation tools for solar industrial process heat absorption cooling system with exhaust heat recovery generator. Trans Tianjin
systems, URL: < http://task49.iea-shc.org/data/sites/1/publications/IEA_ Univ 2010;16:187–93.
Task49_Deliverable_C2_Overview%20Simulation%20Tools-final-160525.pdf > ; [165] Lizarte R, Marcos JD. COP optimisation of a triple-effect H2O/LiBr absorption
2016 [accessed December 2016]. cycle under off-design conditions. Appl Therm Eng 2016;99:195–205.

80
A. Shirazi et al. Energy Conversion and Management 171 (2018) 59–81

[166] Ratlamwala TAH, Gadalla MA, Dincer I. Performance assessment of an integrated technologies. Energy Policy 2005;33:289–96.
PV/T and triple effect cooling system for hydrogen and cooling production. Int J [191] Flowers ME, Smith MK, Parsekian AW, Boyuk DS, McGrath JK, Yates L. Climate
Hydrogen Energy 2011;36:11282–91. impacts on the cost of solar energy. Energy Policy 2016;94:264–73.
[167] Pintaldi S, Sethuvenkatraman S, White S, Rosengarten G. Energetic evaluation of [192] Novacheck J, Johnson JX. The environmental and cost implications of solar energy
thermal energy storage options for high efficiency solar cooling systems. Appl preferences in renewable portfolio standards. Energy Policy 2015;86:250–61.
Energy 2017;188:160–77. [193] Martinopoulos G. 3 – Energy efficiency and environmental impact of solar heating
[168] Aliane A, Abboudi S, Seladji C, Guendouz B. An illustrated review on solar ab- and cooling systems. Advances in Solar Heating and Cooling. Woodhead
sorption cooling experimental studies. Renew Sustain Energy Rev Publishing; 2016. p. 43–59.
2016;65:443–58. [194] Bukoski J, Gheewala SH, Mui A, Smead M, Chirarattananon S. The life cycle as-
[169] Drosou V, Kosmopoulos P, Papadopoulos A. Solar cooling system using con- sessment of a solar-assisted absorption chilling system in Bangkok, Thailand.
centrating collectors for office buildings: a case study for Greece. Renew Energy Energy and Buildings. 2014;72:150–6.
2016;97:697–708. [195] Ghaith FA, Abusitta R. Energy analyses of an integrated solar powered heating and
[170] Ghafoor A, Munir A. Worldwide overview of solar thermal cooling technologies. cooling systems in UAE. Energy Build 2014;70:117–26.
Renew Sustain Energy Rev 2015;43:763–74. [196] Tsoutsos T, Aloumpi E, Gkouskos Z, Karagiorgas M. Design of a solar absorption
[171] Nakahara N, Miyakawa Y, Yamamoto M. Experimental study on house cooling and cooling system in a Greek hospital. Energy Build 2010;42:265–72.
heating with solar energy using flat plate collector. Sol Energy 1977;19:657–62. [197] Deb K. Multi-objective optimization using evolutionary algorithms. 2nd ed. New
[172] Li ZF, Sumathy K. Experimental studies on a solar powered air conditioning system York, USA: John Wiley & Sons; 2008.
with partitioned hot water storage tank. Sol Energy 2001;71:285–97. [198] Calise F. Thermoeconomic analysis and optimization of high efficiency solar
[173] Darkwa J, Fraser S, Chow DHC. Theoretical and practical analysis of an integrated heating and cooling systems for different Italian school buildings and climates.
solar hot water-powered absorption cooling system. Energy 2012;39:395–402. Energy Build 2010;42:992–1003.
[174] Rosiek S, Batlles FJ. Integration of the solar thermal energy in the construction: [199] Calise F, d’Accadia MD, Vanoli L. Thermoeconomic optimization of solar heating
analysis of the solar-assisted air-conditioning system installed in CIESOL building. and cooling systems. Energy Convers Manage 2011;52:1562–73.
Renew Energy 2009;34:1423–31. [200] Calise F. Design of a hybrid polygeneration system with solar collectors and a solid
[175] Said SAM, El-Shaarawi MAI, Siddiqui MU. Analysis of a solar powered absorption oxide fuel cell: dynamic simulation and economic assessment. Int J Hydrogen
system. Energy Convers Manage 2015;97:243–52. Energy 2011;36:6128–50.
[176] Marc O, Sinama F, Praene J-P, Lucas F, Castaing-Lasvignottes J. Dynamic mod- [201] Al-Alili A, Hwang Y, Radermacher R, Kubo I. Optimization of a solar powered
eling and experimental validation elements of a 30 kW LiBr/H2O single effect absorption cycle under Abu Dhabi’s weather conditions. Sol Energy
absorption chiller for solar application. Appl Therm Eng 2015;90:980–93. 2010;84:2034–40.
[177] Xu ZY, Wang RZ, Wang HB. Experimental evaluation of a variable effect [202] Jing Y, Li Z, Liu L, Lu S, Lv S. Exergoeconomic-optimized design of a solar ab-
LiBr–water absorption chiller designed for high-efficient solar cooling system. Int J sorption-subcooled compression hybrid cooling system for use in low-rise build-
Refrig 2015;59:135–43. ings. Energy Convers Manage 2018;165:465–76.
[178] Bermejo P, Pino FJ, Rosa F. Solar absorption cooling plant in Seville. Sol Energy [203] Aminyavari M, Najafi B, Shirazi A, Rinaldi F. Exergetic, economic and environ-
2010;84:1503–12. mental (3E) analyses, and multi-objective optimization of a CO2/NH3 cascade
[179] Hang Y, Qu M, Winston R, Jiang L, Widyolar B, Poiry H. Experimental based refrigeration system. Appl Therm Eng 2014;65:42–50.
energy performance analysis and life cycle assessment for solar absorption cooling [204] Shirazi A, Najafi B, Aminyavari M, Rinaldi F, Taylor RA.
system at University of Californian, Merced. Energy Build 2014;82:746–57. Thermal–economic–environmental analysis and multi-objective optimization of an
[180] Osborne JP, Kohlenbach P, Jakob U, Dreyer J, Kim J. The design and installation ice thermal energy storage system for gas turbine cycle inlet air cooling. Energy
of a combined concentrating power station, solar cooling system and domestic hot 2014;69:212–26.
water system. Energy Proc 2015;70:486–94. [205] Sánchez-Orgaz S, Pedemonte M, Ezzatti P, Curto-Risso PL, Medina A, Calvo
[181] Mokhtar M, Ali MT, Bräuniger S, Afshari A, Sgouridis S, Armstrong P, et al. Hernández A. Multi-objective optimization of a multi-step solar-driven Brayton
Systematic comprehensive techno-economic assessment of solar cooling technol- plant. Energy Convers Manage 2015;99:346–58.
ogies using location-specific climate data. Appl Energy 2010;87:3766–78. [206] Jain V, Sachdeva G, Kachhwaha SS, Patel B. Thermo-economic and environmental
[182] Allouhi A, Kousksou T, Jamil A, El Rhafiki T, Mourad Y, Zeraouli Y. Economic and analyses based multi-objective optimization of vapor compression–absorption
environmental assessment of solar air-conditioning systems in Morocco. Renew cascaded refrigeration system using NSGA-II technique. Energy Convers Manage
Sustain Energy Rev 2015;50:770–81. 2016;113:230–42.
[183] Boopathi Raja V, Shanmugam V. A review and new approach to minimize the cost [207] Wang M, Wang J, Zhao P, Dai Y. Multi-objective optimization of a combined
of solar assisted absorption cooling system. Renew Sustain Energy Rev cooling, heating and power system driven by solar energy. Energy Convers
2012;16:6725–31. Manage 2015;89:289–97.
[184] Desideri U, Proietti S, Sdringola P. Solar-powered cooling systems: Technical and [208] Falke T, Krengel S, Meinerzhagen A-K, Schnettler A. Multi-objective optimization
economic analysis on industrial refrigeration and air-conditioning applications. and simulation model for the design of distributed energy systems. Appl Energy.
Appl Energy 2009;86:1376–86. [209] Iranmanesh A, Mehrabian MA. Optimization of a lithium bromide–water solar
[185] Eicker U, Colmenar-Santos A, Teran L, Cotrado M, Borge-Diez D. Economic eva- absorption cooling system with evacuated tube collectors using the genetic algo-
luation of solar thermal and photovoltaic cooling systems through simulation in rithm. Energy Build 2014;85:427–35.
different climatic conditions: an analysis in three different cities in Europe. Energy [210] Gebreslassie BH, Guillén-Gosálbez G, Jiménez L, Boer D. A systematic tool for the
Build 2014;70:207–23. minimization of the life cycle impact of solar assisted absorption cooling systems.
[186] Bellos E, Tzivanidis C. Energetic and financial analysis of solar cooling systems Energy 2010;35:3849–62.
with single effect absorption chiller in various climates. Appl Therm Eng [211] Gebreslassie BH, Guillén-Gosálbez G, Jiménez L, Boer D. Solar assisted absorption
2017;126:809–21. cooling cycles for reduction of global warming: a multi-objective optimization
[187] Arsalis A, Alexandrou AN. Parametric study and cost analysis of a solar-heating- approach. Sol Energy 2012;86:2083–94.
and-cooling system for detached single-family households in hot climates. Sol [212] Xu D, Qu M, Hang Y, Zhao F. Multi-objective optimal design of a solar absorption
Energy 2015;117:59–73. cooling and heating system under life-cycle uncertainties. Sustain Energy Technol
[188] Eicker U, Pietruschka D, Haag M, Schmitt A. Systematic design and analysis of Assess 2015;11:92–105.
solar thermal cooling systems in different climates. Renew Energy [213] Shirazi A, Taylor RA, Morrison GL, White SD. A comprehensive, multi-objective
2015;80:827–36. optimization of solar-powered absorption chiller systems for air-conditioning ap-
[189] Bellos E, Tzivanidis C, Symeou C, Antonopoulos KA. Energetic, exergetic and fi- plications. Energy Convers Manage 2017;132:281–306.
nancial evaluation of a solar driven absorption chiller – a dynamic approach. [214] IEA SHC Annual Report 2017, URL: < http://www.iea-shc.org/Data/Sites/1/
Energy Convers Manage 2017;137:34–48. publications/IEA-SHC-Annual-Report-2017.pdf > ; 2017 [accessed April 2018].
[190] Tsoutsos T, Frantzeskaki N, Gekas V. Environmental impacts from the solar energy

81

Вам также может понравиться