Energy Conversion and Management 215 (2020) 112895

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Comprehensive evaluation of low-grade solar trigeneration system by T

photovoltaic-thermal collectors
⁎
Zeyu Lia,b,c, , Hongkai Chena,b,c, Yongrui Xua,b,c, Kim Tiow Ooid
a
School of Electric Power, South China University of Technology, Guangzhou 510640, China
b
Guangdong Province Key Laboratory of High Efficient and Clean Energy Utilization, South China University of Technology, Guangzhou 510640, China
c
Guangdong Province Engineering Research Center of High Efficient and Low Pollution Energy Conversion, Guangzhou 510640, China
d
School of Mechanical and Aerospace Engineering, Nanyang Technological University, 639798 Singapore, Singapore

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

Keywords: Solar trigeneration systems based on photovoltaic-thermal collectors can play a positive role in promoting re-
Low-grade newable energy integration in buildings. Usually, low-grade solar heat produced by photovoltaic-thermal col-
Photovoltaic-thermal collector lectors does not meet the temperature requirement for driving absorption chillers, leading to unsatisfactory
Solar trigeneration system performance. In this study, the solar absorption-subcooled compression hybrid cooling system, which is
Absorption chiller
capable of utilizing heat at temperatures as low as 60 °C, is coupled with photovoltaic-thermal collectors to be a
Energy saving
low-grade solar trigeneration system. In spite of heat recovery for heating and cooling, the electrical output of
collectors decreases, due to the losses caused by the additional glass cover and elevated operating temperature.
Therefore, through quasi-steady simulations based on annual meteorological data of three subtropical cities, this
system is assessed from energy, economic and environmental viewpoints. The results show that in Guangzhou,
the specific annual electricity saving of this system is up to 170.6 kWh/m2, which is 17.3% higher than that of a
comparable photovoltaic system. The parametric analysis suggests that increasing the solar filed area gives rise
to a shorter payback period. With the maximum available solar filed area of 600 m2, the minimum payback
period and the maximum electricity saving can be achieved by the absorption subsystem capacities of around
40 kW and 180 kW, respectively. The most economically viable system is obtained in Zhuhai where annual solar
irradiation is close to 1400 kWh/m2 and electricity prices are relatively high. This study is useful for the de-
velopment of solar trigeneration systems based on photovoltaic-thermal collectors and energy saving in modern
cities.

1. Introduction worldwide [1]. Consequently, enhancing solar utilization efficiency is

highly important for the widespread application of PV technology.
Recently, attention has been paid to sustainable development due to Typically, the peak efficiency of crystalline silicon PV modules is
the serious environmental issues associated with the extensive use of close to 0.16 at Standard Test Conditions (STC, i.e., 1000 W/m2 solar
fossil fuel. The Paris Agreement, which aims to limit the global tem- irradiance, 25 °C cell temperature and AM1.5 spectral distribution) [2].
perature increase to 1.5 °C, is agreed by all the parties in 2015 United This implies that nearly 85% of solar energy is not harvested during PV
Nations Climate Change Conference. Therefore, the transition of tra- power generation, but wasted as rejected heat. Moreover, the heat ac-
ditional fossil energy systems to renewable energy is inevitable and cumulation occurring in PV cells, to some extent, decreases the elec-
urgent. Solar photovoltaic (PV) technology has become increasingly trical efficiency, especially in the case of concentrating PV modules [3].
popular, owing to its simple installation, low transmission losses, high Therefore, photovoltaic-thermal (PVT) collectors are proposed to re-
reliability, and the abundance of solar energy. China’s 13th Five-Year markably increase solar energy utilization efficiency. This technology
Plan for Renewable Energy states that the annual power generation of combines the advantages of traditional PV modules and solar thermal
PV facilities is planned to go up from 4318 million kWh in 2015 to collectors into a single system which provides heat and electricity at the
10,500 million kWh in 2020. Till today, the relatively expensive cost of same time [4]. In this regard, PVT collectors are promising for com-
PV generation leads to a strong dependence on government subsidies, bined cooling, heating and power provision, especially in densely-po-
which is the primary obstacle for the development of solar PV systems pulated urban regions where the available solar field area is restricted

⁎
Corresponding author at: School of Electric Power, South China University of Technology, Guangzhou 510640, China.
E-mail address: [email protected] (Z. Li).

https://doi.org/10.1016/j.enconman.2020.112895
Received 6 January 2020; Received in revised form 13 March 2020; Accepted 23 April 2020
0196-8904/ © 2020 Elsevier Ltd. All rights reserved.
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

Nomenclature Q heat transfer rate (kW)

t temperature (°C), time (s)
Abbreviations Δt time step (s)
UA overall heat transfer coefficient (kW/K)
3E energy, economic and environmental T temperature (K)
AS absorption subsystem ΔT temperature difference
CHW chilled water ΔΔT characteristic temperature difference (K)
CW cooling water V volume (m3)
CS compression subsystem W compressor work (kW)
DHW domestic hot water
DHWT domestic hot water tank Greek symbols
HP heat pump
HW hot water α absorptance (–)
PHE preheating heat exchanger (τα) transmittance-absorptance product (–)
PV photovoltaic η efficiency (–)
PVT photovoltaic-thermal
SASCHCS solar absorption-subcooled compression hybrid cooling Subscripts
system
ST storage tank abs absorber
STC Standard Test Conditions av average
b bottom section
Symbols c closure frame
col collector
A area (m2) con condenser
b temperature coefficient (K−1) dis discharge
c specific heat capacity [kJ/(kg·K)] eva evaporator
CEF carbon dioxide emission factor (ton/MWh) el electrical
CER carbon dioxide emission reduction (ton) g glass cover
CECS carbon dioxide emission cost saving ($) gen generator
CI capital investment ($) in inlet
COP coefficient of performance (–) inv inverter
ECS electricity cost saving ($) m motor
ES electricity saving (kWh) min minimum
h heat transfer coefficient [J/(m2 K)], specific enthalpy (J/ nom nominal
kg) out outlet
I solar irradiance (W/m2) p PVT laminate
IR interest rate (–) prim-e converting primary energy into electricity
LMTD logarithmic mean temperature difference (K) ref reference system, refrigerant
LT system lifetime (year) s isentropic
m mass flow rate (kg/s) suc suction
M mass (kg) sur surrounding
NPV net present value ($) th thermal
OCS operating cost saving ($/year) tri trigeneration system
P electric power (kW) tot total
PBP payback period (year) w water
PES primary energy saving (kWh)

[5]. London, a PVT system can cover half of the annual electrical demand
Mainly, the heat produced by PVT collectors is used for heating and its payback period is 9–11 years. Long et al. [11] proposed a
purposes. Ahmed-Dahmane et al. [6] designed a building-integrated building-integrated heat pipe PVT system for electricity and DHW
photovoltaic-thermal (BIPVT) system for preheating fresh air, and this provision, concluding that this system achieves the electrical and
system is able to reduce 24.2% of thermal energy consumption in thermal efficiencies of 0.1 and 0.35, respectively, with a specific annual
winter. Shahsavar et al. [7] also investigated a BIPVT system coupled electricity saving of 315 kWh/m2. Lari et al. [12] pointed out that
with thermal wheels, indicating that the thermal energy recovered from 27.3% of DHW demand in residential buildings can be met by thermal
exhaust air increases by 4.4% but the generated electricity decreases by batteries which overcome the time lag between supply and demand.
7.3%, as compared to the BIPVT system without thermal wheels. Yu The experimental results of Cen et al. [13] showed that the electrical
et al. [8] proposed a hybrid photocatalytic, photovoltaic and thermal and thermal efficiencies of PVT solar field producing DHW are 13.4%
recovery system for solar gradient utilization. This system achieves an and 53.4%, respectively. Furthermore, PVT collectors are used for food
electrical efficiency of 0.174 and a total solar utilization efficiency of drying as well. Yakobiuk et al. [14] reported that 40–50% of incident
0.644. Apart from air preheating, PVT collectors are also employed to solar radiation can be converted into heat requirements in a BIPVT
provide domestic hot water (DHW). The simulation [9] showed that the dehydration system. The numerical study [15] concluded that the op-
daily useful heat for water heating is estimated to be 3.18 kWh/m2 timal air gap thickness of BIPVT collectors for drying applications is
under Tunisian climate conditions. The techno-economic analysis [10] nearly 8 cm. Furthermore, given that PVT collectors usually provide
concluded that for domestic heat and power provision in homes of low-grade solar heat, they can be coupled with heat pumps to elevate

2
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

the quality and quantity of heat for household electricity, space heating field [27]. Cheng et al. [28] proposed a double-stage electrodialysis
and DHW demands [16]. Vaishak et al. [17] reported that PVT-assisted regeneration system with PVT collectors for liquid desiccant cooling,
heat pump systems can achieve a significantly higher electrical effi- finding that it is two to three times more energy-efficient than the
ciency as compared to independent PV systems, and their coefficients of single-stage one. The experimental results of Buker et al. [29] showed
performance (COP) are acceptable. The experimental results of Wang that a liquid desiccant cooling system driven by 140 m2 of BIPVT col-
et al. [18] showed that the COP of water source heat pumps coupled lectors is capable of producing a cooling capacity of 5.2 kW and annual
with PVT collectors is 42.6% higher than that of air source heat pumps. electricity of 10.3 MWh. Additionally, Su et al. [30] concluded that the
In addition, the combination of thermoelectric devices and PVT col- annual energy saving ratio of a liquid desiccant cooling system with
lectors gained some attention. The experimental results of He et al. [19] PVT collectors for trigeneration is 73.3%. Ren et al. [31] introduced
showed that thermoelectric modules powered by PVT collectors can thermal energy storage to match the collector heat production and
achieve a COP of 1.7 in the heating mode, and the electrical and energy demand for desiccant wheel regeneration, aiming to achieve
thermal efficiencies of PVT collectors are 16.7% and 23.5%, respec- more flexible operation. Furthermore, adsorption chillers, whose
tively. Liu et al. [20] modeled a photovoltaic thermal-compound ther- driving temperature is around 65 °C, can also be driven by PVT col-
moelectric ventilator (PVT-TEV) system, and found that its average COP lectors. It was exhibited that the cooling output of adsorption chillers
is up to 13.85. Riahi et al. [21] investigated a concentrating PVT system coupled with BIPVT solar field nearly covers all the cooling demand of a
combined with thermoelectric generators, and found that it can pro- three-story building located in Naples [32].
duce 5.9% more annual electrical energy than the PVT system without Since absorption chillers are more suitable for large-scale cooling
thermoelectric generators. application and their COP is relatively high among thermal cooling
Although PVT collectors for heating purposes remarkably improve machines [33], the energy saving potential of absorption chillers driven
solar utilization, the coupling of PVT collectors and refrigeration sys- by PVT collectors is more pronounced (e.g., an electrical efficiency of
tems for building cooling is even more suitable and attractive for re- 10% along with a primary energy saving ratio of 50% [34]). Differently
gions with rather high cooling demand and long cooling period (e.g., from desiccant and adsorption cooling, the operating temperature
subtropical and tropical cities). It was reported that PVT collectors for mismatch between PVT collectors and absorption chillers has to be
night radiative cooling yield an annual cooling capacity of 55 kWh/m2 overcome for reliable cooling. Generally, the hot water temperature is
under Shanghai climate conditions [22]. Gupta et al. [23] presented a required to be more than 80 °C to drive absorption chillers (at a cooling
passive building cooling system equipped with semitransparent BIPVT water inlet temperature of 32 °C and an evaporation temperature of
collectors. It was exhibited that the indoor air temperature reduces by 5 °C), but the PV panel temperature of PVT collectors exceeds 80 °C
about 30 °C through evaporative cooling and natural ventilation. only in midday [35]. Accordingly, one of the solutions is to reheat the
Moreover, such a system with cooling water increases its electrical collector hot water by auxiliary heaters to maintain the operation of
output by 16.6% [24]. Hu et al. [25] developed photovoltaic-photo- absorption chillers. However, it should be noted that the electrical ef-
thermic-radiative cooling collectors whose average electrical and ficiency of PVT collectors drops, due to the higher operating tempera-
thermal efficiencies reach 10.3% and 55.3%, respectively, along with a ture [36]. Furthermore, a large amount of thermal energy is consumed
net cooling capacity of 72.4 W/m2. In order to evaluate the active and by backup heaters, especially under weak solar irradiance and high
passive effect, a new dynamic model of air open-loop PVT collectors cooling demand conditions, leading to the poor economic feasibility of
based on detailed transient finite-difference thermal network was de- the above-mentioned solution in high-rise buildings [37]. Shirazi el al.
veloped by Athienitis et al. [26]. In addition to radiative cooling, PVT [38] reported that a solar absorption-compression hybrid cooling
collectors can also be adopted to drive desiccant systems, because their system, in which its absorption subsystem is exclusively driven by solar
low driving temperature matches the heat source feature of PVT solar heat and its compression subsystem provides auxiliary cooling, is

Fig. 1. Schematic of low-grade solar trigeneration system.

3
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

economically viable in high-rise buildings, owing to the significant re- consumption of the air source heat pump. A heat exchanger for DHW
duction of operating costs. The simulation results of Buonomano et al. preheating is placed at the bottom of the DHW tank. Similar to the
[39] not only confirmed this viewpoint but also revealed that the heat arrangement in Ref. [45], the compression subsystem is connected to
source temperature for absorption chillers should be decreased as low the electric grid, and consumes the electricity generated by PVT col-
as possible to prevent the significant degradation of electrical effi- lectors as well. During the non-cooling season, PV electricity is supplied
ciency, because the amount of PV generation is dominant in energy for the air source heat pump, elevators, lighting and so on. In addition,
saving profitability. Accordingly, an absorption-compression hybrid the harvested solar heat is entirely used to preheat DHW. Given that the
refrigeration cycle, whose absorption subcycle can be driven by solar PVT system is insufficient to fully supply the electricity, cooling and
heat at lower temperatures, is attractive to utilize low-grade heat from heating demands of the entire hotel, the electric grid, compression
PVT collectors. subsystem and air source heat pump serve as the backup systems.
The solar absorption-subcooled compression hybrid cooling system The operation strategy of the low-grade solar trigeneration system is
(SASCHCS) can utilize low-grade solar heat at temperatures as low as as follows. The pump P1 is turned on at 7:00 and the water in the
60 °C. In this system, the cooling output of its absorption subsystem storage tank is heated up gradually. This pump is switched off if the
serves as the subcooling power for its compression subsystem. Thus, it bottom temperature of the storage tank exceeds the PVT laminate
becomes a better solution for solar cooling in high-rise buildings [40]. temperature. This is to prevent the performance deterioration caused by
The notable advantage of the SASCHCS lies in the elevated evaporation heat transfer from hot water to PV cells. During the cooling season, the
temperature in its absorption subsystem (usually close to 20 °C), which pump P2 is activated when the top temperature of the storage tank
is attributed to the higher chilled water temperature in its absorption reaches 70 °C, and then hot water is supplied for the absorption sub-
subcycle. The experiment [41] showed that 22.2% of compressor work system. The absorption subsystem is switched off when the top tem-
can be saved by the SASCHCS on the sunny day in subtropical perature of the storage tank drops below 55 °C. Subsequently, the hot
Guangzhou. Compared with the solution of Shirazi el al., the product water is used to preheat DHW. In the non-cooling season, the pump P2
cost flow rate of the SASCHCS decreases by 21.7%, implying its better is switched on when the top temperature of the storage tank reaches
exergoeconomic performance [42]. Given that the SASCHCS exhibits 43 °C (only for March) or 50 °C (for other months). The relatively low
great potential in low-grade heat utilization, the authors further in- activation temperature in March is attributed to the weak solar irra-
vestigate a low-grade solar trigeneration (electricity, cooling and diance. Note that the compression subsystem and air source heat pump
heating) system based on coupling PVT collectors with the SASCHCS in are kept operating to fulfill cooling and heating demands during the
this study. In spite of heat recovery for cooling and heating, the cou- entire period.
pling of PVT collectors and SASCHCS adversely affects PV generation,
which is attributed to not only the optical losses caused by the addi- 3. System modeling
tional glass covers in PVT collectors but also the elevated operating
temperature of PV cells. Therefore, it is highly important to estimate This section introduces the modeling for the low-grade solar tri-
whether the capital investment of the trigeneration system can be re- generation system. The models of system components are developed
turned by its total electricity saving within an acceptable payback based on the following assumptions: (1) quasi-steady state; (2) the
period. In this regard, the authors attempt to carry out the energy, water flow through the fluid channels of PVT collectors is uniform; (3)
economic and environmental (3E) study for this system to evaluate its the cooling output of the absorption subsystem is equal to the heat
feasibility and potential. The thermodynamic model of this system is transfer rate in the subcooler; (4) the outlet refrigerant of the eva-
established, and then the 3E performance is assessed in detail through porator and condenser in the compression subsystem is saturated; (5) in
quasi-steady simulations based on the annual meteorological data of the compression subsystem, the heat loss and pressure drop in the heat
three subtropical cities (i.e., Guangzhou, Shenzhen and Zhuhai). The exchangers and pipelines are negligible.
novelty of this paper is the thorough evaluation and parametric analysis
of the novel low-grade solar trigeneration system based on coupling 3.1. Photovoltaic-thermal collector
PVT collectors with the SASCHCS. This investigation is helpful to en-
hance the energy utilization of PV technology and energy saving in The cross-section of a PVT collector is depicted in Fig. 2. Its struc-
modern cities. ture is similar to that in Ref. [16]. This PVT collector is glazed in order
to reduce heat loss, although the glass cover deteriorates its electrical
2. Description of low-grade solar trigeneration system efficiency. Moreover, the PVT laminate, consisting of ethylene-vinyl-
acetate (EVA) layers, PV cells and absorber sheets, is coated with low-
The schematic of the low-grade solar trigeneration system based on emissivity materials [46] to further enhance the thermal efficiency. The
coupling PVT collectors with the SASCHCS is shown in Fig. 1. This
system is installed in a high-rise hotel and serves as a distributed energy
system. Electricity and heat are simultaneously produced by PVT col-
lectors when solar radiation is available. During the cooling season, the
collected solar heat is employed to drive the absorption subsystem and
the absorption cooling output is delivered as subcooling power for the
compression subsystem to save compressor work. Noted that the ab-
sorption subsystem is a single-effect lithium bromide/water (LiBr/H2O)
absorption chiller and the compression subsystem is a traditional vapor
compression chiller with R410a refrigerant. The conversion of sub-
cooling power into cooling output increment in the compression sub-
system depends on the flow rates of cooling water and chilled water, as
well as the compressor speed [43]. Moreover, the experiment [44] re-
vealed that the ratio of cooling output increment in the compression
subsystem to subcooling power is 0.9–1.2 at fixed superheating condi-
tions. In terms of step utilization principles, when the storage tank
water temperature is too low to drive the absorption subsystem, the
remaining solar heat is applied to preheat DHW to reduce the electricity Fig. 2. Cross-section of PVT collector.

4
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

energy balance of PVT collectors is described according to the one-di- that allows absorption chillers to produce the cooling effect, derived by
mension simplified approach [16]: the temperatures of external water loops:
dT dT
⎧ Mg cg g = Mg cg g = APVT [I ·αg + hsur − g (Tsur − Tg ) + hg − p (Tp − Tg )] ⎧ ΔΔT = Tw, gen − Tw, abs − B·(Tw, con − Tw, eva )
dt dt
⎪ ⎪ Tw, gen = (THWin, AS + THWout , AS ) 2
⎪ Mp cp dTp = APVT [I ·(τα )av + hg − p (Tg − Tp) + hp − c (Tc − Tp)] − Pel − Qth ⎪
⎪ dt Tw, abs = (TCWin, ASabs + TCWout , ASabs ) 2
dT ⎨
⎨ Mc cc dtc = APVT [hp − c (Tp − Tc ) + hsur − c (Tsur − Tc )] ⎪ Tw, con = (TCWin, AScon + TCWout , AScon) 2
⎪ ⎪ Tw, eva = (TCHWin, AS + TCHWout , AS ) 2
⎪ Pel = I ·APVT ηPV , STC [1 − b (Tp − TPV , STC )] ⎩ (5)
⎪ Qth = mcol c w (THWout , col − THWin, col )
⎩ where B = 1.15 for single-effect LiBr/H2O absorption chillers [50].
(1)
where mcol is the hot water flow rate through PVT collectors; h is the 3.5. Compression subsystem
heat transfer coefficient, calculated by the expressions in Ref. [16,47];
ηPV,STC = 0.127 (the influence of low-emissivity coatings on electrical The compression subsystem is modeled by the lumped parameter
efficiency is considered [46]); b = 0.004 K−1; TPV,STC = 298.15 K. method. The thermodynamic characteristic of its condenser, subcooler
It should be mentioned that the temperature difference between and evaporator is described by mass and energy conservation. For each
PVT laminate and rear heat exchangers is neglected because of the heat exchanger, the governing equations of internal fluid (i.e., re-
insignificant thermal resistance. Similarly, the temperature difference frigerant) are as follows:
between the closure frame and the nearest insulation layer is negligible ∑ min = ∑ mout
⎧
as well.
⎨
⎩ Q = ∑ min hin − ∑ mout hout (6)
3.2. Hot water storage tank For external fluids (i.e., cooling water and chiller water), the heat
transfer rates are determined by:
By considering thermal stratification, the hot water storage tank is
modeled as a cylindrical vessel which can be divided into three sections ⎧ Q = m w c w ΔTw
with equal volumes. The section 1 represents the top layer and the ⎩Q = UA·LMTD
⎨ (7)
section 3 donates the bottom layer. The energy balance of jth-section is The compressor work (WCS) depends on the refrigerant flow rate
described as follows: (mref):
dTST , j mref (hdis, s − hsuc )
Mj c w WCS =
dt ηs ηm (8)
= mcol c w (TST , j − 1 − TST , j ) + mHW c w (TST , j + 1 − TST , j ) + (UA)loss, j
(Tsur − TST , j ) (2) 3.6. Energy, economic and environmental analysis
where mHW is the hot water flow rate through the absorption subsystem
(for cooling) or DHW tank (for heating); TST,j−1 for the top section is The annual electrical, thermal and total efficiencies of PVT collec-
equal to the collector outlet water temperature; TST,j+1 for the bottom tors are calculated by:
section is equal to the hot water outlet temperature of the generator (for ∫ ηinv Pel dt
cooling) or preheating heat exchanger (for heating); (UA)loss,j is the ηel =
∫ I ·APVT dt (9)
overall heat loss coefficient, depending on the volume, aspect ratio and
insulation of the tank [48]. ∫ Qth dt
ηth =
∫ I ·APVT dt (10)
3.3. Preheating heat exchanger
∫ ηinv Pel dt + ∫ Qth dt
ηtot =
The preheating heat exchanger is placed at the bottom of the DHW ∫ I ·APVT dt (11)
tank. The heat transfer from the preheating heat exchanger to the
bottom water in the DHW tank is controlled by the following equations: For the calculation of electricity saving, a reference system is con-
sidered. It simply consists of a compression subsystem and an air source
dT
⎧ MDHWT , b c w DHWT
dt
,b
= QPHE − (UA)loss, DHWT (TDHWT , b − Tsur ) heat pump, satisfying the whole cooling and DHW demands of the
⎨ QPHE = hPHE APHE LMTDPHE = mHW c w (THWin − THWout ) building, respectively. Accordingly, the electricity saving (ES) is defined
⎩ (3)
as:
QPHE
3.4. Absorption subsystem ES = ∫ ηinv Pel dt + ∫ (WCS,ref − WCS,tri ) dt + ∫ COP HP
dt
(12)

The thermodynamic model of the absorption subsystem is based on where the first, second and third terms of the right-hand side represent
the characteristic equation method in order to simplify the iteration of the electricity savings through PV generation, solar cooling and DHW
solution processes [49]. The cooling capacity and coefficient of per- preheating, respectively.
formance are calculated by: Similar to Ref. [39], the primary energy saving (PES) is calculated
by:
⎧ Qeva, AS = s (ΔΔT − ΔΔTmin )
⎪ Q ΔΔT − ΔΔT
ES
COPAS = Qeva, AS = G·ΔΔT + (1 α − Gmin PES =
⎨ gen )·ΔΔTmin ηprim - e (13)
⎪ ΔΔ T = 1.9 + 0.1ΔΔ T (4)
⎩ min
where ηprim-e is the conversion efficiency for converting primary energy
where s and α depend on the product of heat transfer coefficients and into electricity in a conventional way. Here, the power generation ef-
areas of heat exchangers, reflecting the influence of chiller size on ficiency of a conventional power plant is used, which is typically 38%
cooling capacity [49]; G is equal to 1.178 [49]; ΔΔT is the driving force [51].

5
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

The simple payback period (PBP), which is the ratio of capital in- of Guangzhou, respectively. Such data are the average values based on
vestment (CI) to annual operating cost saving (OCS), is employed as the annual measurement [57]. The shape and size of this hotel are similar
economic indicator: to those in Ref. [58]. The length, width and height of each floor are
CI CI 54 m, 20 m and 3.6 m, respectively. The heat transfer coefficients of the
PBP = = external wall, roof and window are 2.336, 0.405 and 5.680 W/(m2 K),
OCS ECS + CECS (14)
respectively. The cooling load of this hotel is calculated by the DeST
where CI denotes the extra investment cost with respect to the reference software [59], and Fig. 7 shows the cooling load in Guangzhou. From
system; such extra investment includes the total cost for solar devices 8:00 to 20:00, the fall of cooling load in guest rooms (guests are out of
(i.e., PVT collectors, water storage tank, absorption subsystem, cooling the hotel) results in the drop of total cooling load in hot months (from
tower, preheating heat exchanger); 12% of the total cost for solar de- July to September). During other months, since the cooling load of
vices is taken into account as the cost for pumps, pipelines and valves guest rooms is much less than that of the public area, the variation of
[52]; OCS is comprised of electricity cost saving (ECS) and carbon di- the total cooling load mirrors that of the solar irradiance.
oxide (CO2) emission cost saving (CECS), calculated according to an- The design and operation parameters of the system are listed in
nual electricity saving, electricity prices and carbon tax for CO2 emis- Table 3, and the parameters of PVT collectors are summarized in
sion. Table 4. Note that in order to prevent interference among collectors, the
The net present value (NPV) is applied to assess the profitability of total available solar field area is nearly 600 m2 which is 56% of the roof
the system: area. Similar to Ref. [60], the design heat transfer areas and external
LT flow rates of heat exchangers are proportional to the rated cooling
OCS
NPV = ∑ (1 + IR) j
− CI capacity of the SASCHCS prototype [41].
j=1 (15)
In general, the comprehensive environmental evaluation is com- 5. Results and discussion
plicated and usually includes the emission in the entire life cycle of a
system. Since the 3E evaluation primarily focuses on the energy saving In this section, the results of energy, economic, and environmental
within the system lifetime, the environmental evaluation is based on evaluation are presented and analyzed in detail.
CO2 emission reduction (CER) through electricity saving (like other 3E
evaluation): 5.1. Quasi-steady simulation results for Guangzhou
CER = ES·CEF (16)
Figs. 8–14 illustrate the thermodynamic behavior of the in-
where the CO2 emission factor for electricity grid (CEF) [38]. vestigated solar trigeneration system in Guangzhou. The PVT laminate
The parameters for economic and environmental analysis are listed temperature is shown in Fig. 8. As expected, the variation of this tem-
in Table 1, respectively. The electricity prices for different cities are perature mirrors that of the solar irradiance. In the early morning, with
summarized in Table 2. the increasing solar irradiance, the PVT laminate temperature goes up
quickly. Subsequently, its growth rate is reduced as the pump P2 is
3.7. Numerical solution procedure activated. The PVT laminate temperature falls gradually in the after-
noon because of the declining solar irradiance. In most months, the
The simulation models are developed by programming with duration, in which a relatively high PVT laminate temperature is
FORTRAN language. The thermophysical properties of the refrigerant maintained, is about five hours after the activation of the pump P2. This
R410A are calculated with the Refprop software [55]. Fig. 3 shows the result implies that the adopted PVT collector type achieves a satisfac-
general solution process for these models. Note that the ordinary dif- tory thermal performance, because the glass cover reduces heat losses
ferential equations of the PVT collector model are numerically solved effectively. The radiation thermal resistance from the PVT laminate to
by Euler’s method. In order to ensure the convergence of iterative se- the surrounding can be divided into the heat resistance from the PVT
quence, the condenser and evaporate temperatures of the compression laminate to glass cover and the heat resistance from the glass cover to
subsystem are solved by the fixed-point iteration method, while ΔΔT is surrounding. Since the glass cover temperature is nearly 10–15 °C
solved by the secant method. A time step (Δt) of 2 min is considered for higher than the surrounding temperature, the heat loss from the PVT
simulation to represent a realistic operation of the system. laminate to surrounding is dramatically reduced, according to the law
of radiation heat transfer. Furthermore, the low-emissivity coatings
3.8. Model validation decrease the emissivity of the PVT laminate to 0.13, so that the linear
thermal loss coefficient of collectors can be reduced by 37.5% [46]. The
The above-mentioned simulation models are validated against the peak PVT laminate temperatures from January to December are 51.6,
corresponding experimental results, as reported in Ref. [56]. As de- 51.5, 43.4, 71.1, 72.1, 71.7, 76.1, 79.1, 86.6, 85.8, 60.5 and 59 °C,
monstrated in Fig. 4, the experimental and simulation results of the hot
water temperature (its calculation relies on the entire system model) Table 1
are compared. A good agreement is obtained and the deviation is within Input data for economic and environmental analysis [53,54].
10%. In addition, during the test period, the experimental and simu-
Parameter Value
lated average COPs of the absorption subsystem are 0.567 and 0.594,
respectively. The deviation is only 4.8%. The experimental and simu- Investment cost of system components ($)
lated average COPs of the compression subsystem are 6.41 and 6.59, PVT collector 270APVT
respectively. A slight deviation of 2.8% is obtained. Storage tank 578.8 + 944.79VST
Single-effect absorption chiller (absorption subsystem) 241.95Qnom,AS
Vapor compression chiller (compression subsystem) 125.04Qnom,CS
4. Case study Cooling tower 15.63Qnom
Cooling coil 18.6Qnom
A case study is performed to assess the 3E performance of the Preheating heat exchanger 268.45 + 516.62APHE
system. This system is built at a 20-storey high-rise hotel located in System lifetime (LT) 20 years
Interest rate (IR) 0.0435
subtropical cities. The cooling season is from April to October, and the
Carbon tax for CO2 emission 18.3 $/ton CO2
other months are the non-cooling season. For example, Figs. 5 and 6 CO2 emission factor for electricity grid (CEF) 0.756 ton/MWh
depict the annual typical solar irradiance and surrounding temperature

6
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

Table 2
Annual solar irradiation and electricity prices in three subtropical cities.
Location Annual horizontal solar irradiation (kWh/m2) Period of time Electricity price ($/kWh)

Guangzhou 1136.6 0:00–8:00 0.053

8:00–14:00; 17:00–19:00; 22:00–0:00 0.102
14:00–17:00; 19:00–22:00 0.165
Shenzhen 1307.5 0:00–7:00; 23:00–0:00 0.032
7:00–9:00; 11:30–14:00; 16:30–19:00; 21:00–23:00 0.100
9:00–11:30; 14:00–16:30; 19:00–21:00 0.134
Zhuhai 1392.8 0:00–8:00 0.053
8:00–9:00; 12:00–19:00; 22:00–0:00 0.102
9:00–12:00; 19:00–22:00 0.165

respectively. Note that the PVT laminate temperatures in the cooling Since these temperatures are directly affected by the PVT laminate
season are 15–25 °C higher than those in the non-cooling season, which temperature, similar variation trends are detected. The temperature
is attributed to the relatively high temperature requirement for driving difference between the collector outlet water and PVT laminate is
the absorption subsystem. within 1 °C and it is positively related to solar irradiance. The maximum
In addition, the collector outlet water temperature and storage tank temperature difference between the PVT laminate and storage tank top
top section temperature are presented in Appendix A (Figs. A1 and A2). section is within 4 °C.

Fig. 3. Flow chart of simulation program.

7
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

Fig. 4. Comparison of experimental and simulation results. Fig. 7. Monthly cooling load of the hotel building in Guangzhou.

Table 3
Design and operation parameters of solar trigeneration system.
Parameter Value

PVT collector
Aperture area of PVT collector unit (m2) 1.63
Number of PVT collectors 372
Azimuth angle (°) 0
Slope angle (°) 20
Hot water flow rate (kg/s) 4.69
Activation time of pump P1 7:00 a.m.
Inverter efficiency 0.9
Hot water storage tank
Volume (m3) 3.04
Aspect ratio 3.5
Heat loss coefficient [W/(m2 °C)] 0.83
Domestic hot water loop
Heat transfer area of preheating heat exchanger 8.2
(m2)
Hot water flow rate (kg/s) 5.56
Fig. 5. Monthly solar irradiance for Guangzhou. Activation temperature of pump P2 in heating 43 (for March)/50 (for other
mode (°C) months)
COP of air source heat pump 4.0
Absorption subsystem
Nominal cooling capacity (kW) 90
Hot water flow rate (kg/s) 5.56
Cooling water flow rate in condenser/absorber 4.44/5.0
(kg/s)
Chilled water flow rate (kg/s) 5.83
Cooling water inlet temperature (°C) 32
Activation/shutdown temperature (°C) 70/55
Compression subsystem
Nominal cooling capacity (kW) 2300
Cooling water flow rate (kg/s) 130.6
Cooling water inlet temperature (°C) 32
Chilled water inlet/outlet temperature (°C) 12/7
Isentropic efficiency of compressor 0.7
Motor efficiency of compressor 0.9

solar irradiation in November is 19.7% lower than that in September.

Note also that the low electrical output in March is due to the extremely
Fig. 6. Monthly surrounding temperature for Guangzhou. low solar irradiance.
The thermal output of PVT collectors is presented in Fig. 10. As
The electrical output of PVT collectors is shown in Fig. 9. It is ob- expected, the trend of thermal output is similar to that of electrical
served that the trend of electrical output generally mirrors that of the output. It should be mentioned that the thermal output decreases to
PVT laminate temperature. The daily mean electrical outputs from zero when the pump P1 is turned off. The daily average thermal outputs
January to December are 6.40, 7.12, 3.96, 6.44, 8.32, 8.52, 9.74, 9.85, from January to December are 16.29, 21.40, 7.35, 15.05, 23.34, 24.34,
10.79, 10.24, 9.36 and 8.76 kW, respectively. These results imply that 29.65, 31.32, 34.4, 30.22, 34.02 and 28.02 kW, respectively. These
in the non-cooling season, the lower PVT laminate temperatures is results imply that the lower PVT laminate temperature can notably
slightly beneficial to electrical output. For example, the electrical improves the thermal output. For example, the thermal output in Feb-
output in November is 13.3% less than that in September, although the ruary is just 8.3% lower than that in May, while the solar irradiation in

8
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

Table 4
Parameters of PVT collectors [16,46,61].
Parameter Value

Glass cover
Absorptance 0.009
Transmittance 0.962
Reflectance 0.0029
Emissivity 0.88
Density (kg/m3) 2515
Thickness (mm) 3.2
PVT laminate
Absorptance 0.862
Emissivity 0.13
Density (kg/m3) 2500
Thickness (mm) 1.71
Air gap thickness (mm) 25
Heat exchanger
Material Aluminum
Thickness (mm) 1.5
Unit heat transfer area of channel (m2) 0.093 Fig. 10. Thermal output of PVT collectors.
Number of channels 40
Insulation
Material Mineral wool
Thickness (mm) 50
Closure frame
Material Aluminum
Emissivity 0.03
Thickness (mm) 3.0

Fig. 11. Cooling output of absorption subsystem.

Fig. 8. Temperature of PVT laminate.

Fig. 12. Electrical efficiency of PVT collectors.

February is 21.3% lower than that in May. As compared to electrical

output, the thermal output is more sensitive to the PVT laminate tem-
perature. Moreover, the ratios of thermal output to electrical output
from January to December are 2.55, 3.01, 1.86, 2.34, 2.81, 2.86, 3.04,
3.18, 3.19, 2.95, 3.63 and 3.20, respectively. By contrast, such a ratio is
Fig. 9. Electrical output of PVT collectors. only 1.27 in the case of the solar trigeneration system with a simple
absorption-compression hybrid configuration [62]. Such a difference

9
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

well-known that the thermal efficiency depends on the collector tem-

perature and solar irradiance, and it increases with the fall in collector
temperature and the rise in solar irradiance. Therefore, the increase of
heat losses, caused by the elevated collector temperature, diminishes
the thermal efficiency in the morning. Subsequently, the thermal effi-
ciency increases as a result of higher solar irradiance. Finally, the
thermal efficiency decreases in the afternoon with solar irradiance de-
clining. It is also observed that the thermal efficiency in the non-cooling
season is higher than that in the cooling season, because of the lower
collector temperature. In particular, the thermal efficiency reaches up
to 0.45 at noon in November. The daily mean thermal efficiencies from
January to December are 30.29%, 36.05%, 22.62%, 26.14%, 31.91%,
33.01%, 34.05%, 34.58%, 33.28%, 30.79%, 40.51% and 36.57%, re-
spectively. Such results reveal that the glass cover and low-emissivity
coatings of PVT collectors are highly beneficial to their thermal per-
formance. Conversely, the thermal efficiency of unglazed PVT collectors
without such low-emissivity coatings decreases by nearly 50% [39].
Fig. 13. Thermal efficiency of PVT collectors. The COP of the absorption subsystem is shown in Fig. 14. Similar to
the authors’ previous study [63], its variation mirrors those of the
driving hot water temperature and cooling output. The daily mean
COPs of the absorption subsystem from April to October are 0.562,
0.600, 0.604, 0.619, 0.624, 0.630 and 0.619, respectively. Corre-
spondingly, the daily average solar COPs (i.e., the ratio of total ab-
sorption cooling output to total solar radiation) from April to October
are 0.080, 0.142, 0.148, 0.171, 0.178, 0.180 and 0.157, respectively. In
addition, the annual mean COP and solar COP of the absorption sub-
system are 0.615 and 0.156, respectively. Such results are close to the
measured data [41], implying that the absorption subsystem of
SASCHCS can be driven efficiently by low-grade solar heat from PVT
collectors.

5.2. Energy, economic and environmental evaluation for Guangzhou

The monthly electricity savings of the solar trigeneration system and

a comparable PV system are shown in Fig. 15. This PV system has the
same specifications of cell efficiency and solar filed area as the trigen-
Fig. 14. COP of absorption subsystem. eration system. Although the thermal energy output of PVT collectors is
2–3 times as much as the electrical energy output (Appendix A Fig. A3),
the amount of electricity saving through PV generation is still dominant
implies that the investigated system significantly increases the thermal
in the total electricity saving of the trigeneration system. This is because
output of PVT collectors, owing to the relatively lower driving tem-
the generated PV electricity is a high energy form, while the solar
perature for the absorption subsystem.
cooling and preheating energy is used to reduce the electricity con-
The cooling output of the absorption subsystem during the cooling
sumption of the relatively high-efficiency vapor compression chiller
season is shown in Fig. 11. Although the top section temperature of the
and air source heat pump. It should also be noted that the electricity
storage tank is seldom greater than 85 °C, the absorption subsystem
savings through cooling and preheating generally compensate the
operates efficiently, because of its elevated evaporation temperature.
According to simulation results, the evaporation temperature of the
absorption subsystem is mainly 15–20 °C. Since the cooling output is
strongly dependent on the driving hot water temperature, its variation
is similar to that of the storage tank top section temperature. Moreover,
the absorption cooling durations from April to October are 188, 388,
398, 392, 394, 418 and 392 min, respectively. The higher monthly solar
irradiation results in the longer duration of absorption cooling.
The electrical efficiency of PVT collectors is depicted in Fig. 12. It
can be seen that the variation of electrical efficiency is inverse to that of
PVT laminate temperature, because the temperature rise in PVT lami-
nate is unfavorable for the electrical efficiency. Notably, the electrical
efficiency drops to 0.087 during the September noon. Moreover, the
electrical efficiency of these PVT collectors is lower than that of un-
glazed PVT collectors, due to the reflection and absorption losses
caused by the additional glass cover and low-emissivity coatings. That
is, the electrical efficiency of glazed PVT collectors with low-emissivity
coatings is 0.127 at Standard Test Conditions, and this value is 15.3%
lower than that of conventional PV modules with the equivalent spe-
cification. Fig. 15. Comparison of monthly electricity savings between solar trigeneration
The thermal efficiency of PVT collectors is presented in Fig. 13. It is system and PV system in Guangzhou.

10
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

decrease in PV generation, making the total electricity saving of the absorption subsystem, the total investment cost of the entire system is
trigeneration system superior to that of the PV system. This is further increased by 1.3 times as the solar field area increases from 200 m2 to
analyzed with the following annual results. 600 m2. However, the increased solar field area gives rise to 1.9 times
The annual results of energy analysis are summarized in Table 5. increase in annual operating cost saving. On the other hand, increasing
These results confirm that this solar trigeneration system has a certain the solar field area results in diminishing return, because the resultant
advantage over conventional PV systems. Based on the same simulation higher collector operating temperature slightly deteriorates the elec-
conditions, an annual electrical efficiency of 0.124 can be achieved by trical and thermal efficiencies. It should also be noted that the limiting
the PV modules with the equivalent specification [53]. By contrast, the factor for collector installation includes space availability. In the pre-
annual electrical efficiency of PVT collectors is 21% lower than that of sented case study, the high-rise hotel has a roof area of 1080 m2. Thus,
PV modules, which is mainly attributed to the optical losses caused by the available area for the solar field is considered as 600 m2 in order to
the additional glass cover and low-emissivity coatings. The glass cover avoid shading effects.
generally receives off-normal solar radiation, and causes averagely 10% With the maximum available solar filed area of 600 m2, the annual
of the total incident solar radiation to be reflected or absorbed. The low- electricity savings for different absorption subsystem capacities and
emissivity coatings decrease the absorptance of PVT laminate from subtropical cities are compared in Fig. 17. Obviously, the electricity
0.901 to 0.862 [46]. Moreover, during the whole year, the irradiance- saving initially increases with the absorption subsystem capacity, since
weighted average cell temperature of PVT collectors is 20 °C higher a high generator heat consumption rate is favorable for both the
than that of PV modules (i.e., 58.3 °C for PVT collectors versus 38.7 °C thermal and electrical efficiencies of PVT collectors. In Guangzhou, as
for PV modules), which also results in electrical efficiency degradation. the rated cooling capacity increases from 30 kW to 180 kW, the average
Nevertheless, PVT collectors improve solar energy utilization by har- thermal and electrical efficiencies are increased by 3.1% and 30%, re-
vesting heat, so the total efficiency of PVT collectors is 3.4 times that of spectively. However, after a certain high rated cooling capacity, a
PV modules. In addition, a specific annual power generation of plateau and a subsequent decay are detected. Although a large-size
145.4 kWh/m2 can be achieved by PV modules [53], while the specific absorption subsystem allows high instantaneous cooling output, the
annual electricity saving of the solar trigeneration system is 17.3% resultant low evaporator temperature reduces its COP. The average COP
higher than that of PV modules. This improvement is attributed to heat decreases by 3.2% and the average thermal efficiency of PVT collectors
recovery for cooling and preheating. just slightly increases by 1.0%, as the rated cooling capacity increases
Furthermore, the investigated system can also be compared with the from 180 kW to 250 kW. Thus, the maximum annual electricity savings
PVT-based trigeneration system based on a simple absorption-com- can be obtained by a 180-kW absorption subsystem in these cities. In
pression hybrid configuration [62]. In this simple configuration, the particular, the highest electricity saving is achieved in Zhuhai, which is
chilled water for space cooling is successively cooled in the evaporators mainly attributed to the highest local solar irradiation (Table 2).
of its absorption and compression chillers. Given that the supply and With the maximum available solar filed area of 600 m2, the payback
return chillers temperatures of air-conditioning systems are usually 7 °C periods for different absorption subsystem capacities and subtropical
and 12 °C, respectively, the evaporator temperature of the absorption cities are compared in Fig. 18. The lowest payback periods can be
chiller needs to be maintained at 5–10 °C for effective cooling. In achieved by a 40-kW absorption subsystem in these cities. The ab-
contrast, the absorption-subcooled compression hybrid configuration sorption subsystem with a too low rated cooling capacity leads to ex-
enables the investigated system to provide better energy performance. cessive heat accumulation in the storage tank, which dramatically
In this way, the absorption subsystem operates with a high evaporator elevates the average collector inlet water temperature and significantly
temperature of around 20 °C, so the required generator temperature can decreases the electrical and thermal gains. Thus, the lower electricity
be decreased from around 75 °C to 55 °C. The required heat input saving and operating cost saving are obtained. On the other hand, an
temperature is accordingly decreased from around 80 °C to 60 °C or excessively large-capacity absorption subsystem is not a good choice
below, remarkably lowering the collector operating temperature. The from economic viewpoint, because the increase in operating cost saving
annual solar cooling output of the investigated system is 2.8 times that cannot compensate the added investment cost. The total investment
of the trigeneration system with the simple hybrid configuration, be- cost of the entire system increases linearly with the absorption sub-
cause of the similar increase in collector thermal efficiency [62]. system capacity, while the resultant increase of electricity saving tends
Table 6 shows the economic and environmental results for to level off (Fig. 17). As expected, the high solar irradiation in Zhuhai
Guangzhou. This economic evaluation does not include the subsidies gives rise to the shortest payback period of 11.8 years and a positive net
for PV power generation, because they vary from city to city and their present value of 23.6 k$. Note that the best economic performance in
continuing availability is uncertain. Although the payback period is less Zhuhai also benefits from the higher daytime electricity prices.
than the system lifetime, the obtained net present value is negative. As discussed above, the presented solar trigeneration system based
Thus, in the case without subsidies, the economic feasibility of the on coupling PVT collectors with the SASCHCS shows an energy-saving
trigeneration system based on the initial design parameters is not sa- advantage over the conventional system configurations (i.e., traditional
tisfactory enough in Guangzhou. The relatively low annual operating PV systems and PVT-based trigeneration systems with simple
cost saving leads to these results. As indicated in Table 2, the annual
solar irradiation in Guangzhou is the lowest, resulting in a low elec-
tricity saving. In this regard, the effect of different subtropical cities and Table 5
various design parameters on system performance is further analyzed in Annual results of energy analysis for Guangzhou.
the following subsection. Parameter Value

5.3. Systematic parametric study Annual electrical efficiency of PVT collectors 0.098
Annual thermal efficiency of PVT collectors 0.325
Annual total efficiency of PVT collectors 0.423
The key design parameters such as solar field area and absorption Specific annual cooling output of absorption subsystem (kWh/m2) 112.8
chiller capacity affect not only the electrical and thermal efficiencies of Specific annual heating output (kWh/m2) 142.8
PVT collectors but also the economic performance of the system. Fig. 16 Specific annual electricity saving through PV generation (kWh/m2) 108.0
shows that if more PVT collectors are installed, the payback period of Specific annual electricity saving through solar cooling and heating 62.6
(kWh/m2)
the system decreases. The same variation trend is found in the case of
Specific annual total electricity saving (kWh/m2) 170.6
all the considered cities. This is because the higher operating cost Specific annual total primary energy saving (kWh/m2) 448.9
saving compensates the higher investment cost. With a 90-kW

11
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

Table 6
Results of economic and environmental analysis for Guangzhou.
Result Value

Extra investment cost (k$) 213.53

Annual operating cost saving (k$) 13.7
Payback period (year) 15.6
Net present value (k$) −33.52
Specific annual CO2 emission reduction (ton/m2) 0.129

Fig. 18. Variations of payback period with absorption subsystem capacity

(APVT = 600 m2).

Fig. 16. Variation of payback period with sizes of PVT solar filed and absorp-
tion subsystem in Guangzhou.

Fig. A1. PVT collector outlet water temperature in Guangzhou.

Fig. 17. Variations of annual electricity saving with absorption subsystem ca-
pacity (APVT = 600 m2).

absorption-compression integration). However, this work is pre-

liminary. Considering time-of-use electricity pricing (Table 2), the
availability of solar cooling is often not coincident with on-peak periods
during which electricity prices are the highest. Thus, for future devel-
opments of this work, it is recommended to integrate energy storage to
improve operating cost savings.
Fig. A2. Top section temperature of storage tank in Guangzhou.

6. Conclusions
important factors (e.g., additional glass cover and collector tempera-
This paper presents the energy, economic and environmental eva- ture), which have a great influence on the electrical and thermal effi-
luation of a low-grade solar trigeneration system based on coupling PVT ciencies of PVT collectors, are analyzed. The feasibility in three sub-
collectors with a solar absorption-subcooled compression hybrid tropical cities with different solar irradiation and electricity prices is
cooling system (SASCHCS). By quasi-steady simulations, some investigated. The effect of key design parameters is also studied. The

12
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

contract No. 201904010218, (3) State Key Laboratory of Compressor

Technology under the contract No. SKL-YSJ201806, (4) Key Laboratory
of Efficient and Clean Energy Utilization of Guangdong Higher
Education Institutes under the contract No. KLB10004.

References

[1] Shuai J, Cheng X, Ding L, Yang J, Leng Z. How should government and users share
the investment costs and benefits of a solar PV power generation project in China?
Renew Sustain Energy Rev 2019;104:86–94.
[2] Haedrich I, Eitner U, Wiese M, Wirth H. Unified methodology for determining CTM
ratios: systematic prediction of module power. Sol Energy Mater Sol Cells
2014;131:14–23.
[3] Wang H, Qu J, Sun Q, Kang Z, Han X. Thermal characteristic comparison of three-
dimensional oscillating heat pipes with/without sintered copper particles inside
flat-plate evaporator for concentrating photovoltaic cooling. Appl Therm Eng
2020;167.
[4] Jia Y, Alva G, Fang G. Development and applications of photovoltaic–thermal sys-
tems: a review. Renew Sustain Energy Rev 2019;102:249–65.
[5] Ramos A, Chatzopoulou MA, Guarracino I, Freeman J, Markides CN. Hybrid pho-
Fig. A3. Monthly thermal output of trigeneration system in Guangzhou. tovoltaic-thermal solar systems for combined heating, cooling and power provision
in the urban environment. Energy Convers Manage 2017;150:838–50.
[6] Ahmed-Dahmane M, Malek A, Zitoun T. Design and analysis of a BIPV/T system
main quantitative conclusions are summarized as follows: with two applications controlled by an air handling unit. Energy Convers Manage
2018;175:49–66.
[7] Shahsavar A, Khanmohammadi S, Khaki M, Salmanzadeh M. Performance assess-
(1) The absorption subsystem of the SASCHCS operates efficiently ment of an innovative exhaust air energy recovery system based on the PV/T-as-
when driven by low-grade heat from PVT collectors. The results for sisted thermal wheel. Energy 2018;162:682–96.
[8] Yu B, Jiang Q, He W, Liu S, Zhou F, Ji J, et al. Performance study on a novel hybrid
Guangzhou show that the annual average COP and specific annual solar gradient utilization system for combined photocatalytic oxidation technology
cooling capacity of the absorption subsystem are 0.615 and and photovoltaic/thermal technology. Appl Energy 2018;215:699–716.
112.8 kWh/m2, respectively. Additionally, the annual electrical and [9] Hazami M, Riahi A, Mehdaoui F, Nouicer O, Farhat A. Energetic and exergetic
performances analysis of a PV/T (photovoltaic thermal) solar system tested and
thermal efficiencies of PVT collectors reach 0.098 and 0.325, re- simulated under to Tunisian (North Africa) climatic conditions. Energy
spectively. The specific annual electricity saving of the solar tri- 2016;107:78–94.
generation system is up to 170.6 kWh/m2, in which PV generation [10] Herrando M, Markides CN. Hybrid PV and solar-thermal systems for domestic heat
and power provision in the UK: techno-economic considerations. Appl Energy
is 108 kWh/m2 and the electricity saving through cooling and 2016;161:512–32.
heating is 62.6 kWh/m2. Accordingly, the specific annual primary [11] Long H, Chow T-T, Ji J. Building-integrated heat pipe photovoltaic/thermal system
energy saving is 448.9 kWh/m2. for use in Hong Kong. Sol Energy 2017;155:1084–91.
[12] Lari MO, Sahin AZ. Effect of retrofitting a silver/water nanofluid-based photo-
(2) The limiting factor for the optimum system design is roof-space
voltaic/thermal (PV/T) system with a PCM-thermal battery for residential appli-
availability. It is recommended to install as many PVT collectors as cations. Renew Energy 2018;122:98–107.
possible in the limited space for a short payback period. With the [13] Cen J, du Feu R, Diveky ME, McGill C, Andraos O, Janssen W. Experimental study
maximum available solar field area of 600 m2, the lowest payback on a direct water heating PV-T technology. Sol Energy 2018;176:604–14.
[14] Yakobiuk D, Jagomagi A, Yakobiuk I. Development of autonomous food dehydra-
period can be obtained by the absorption subsystem capacities of tion system based on building integrated photovoltaic thermal technology. J Renew
around 40 kW. The most energy-efficient design for the absorption Sus Energy 2018;10:021002.
subsystem is around 180 kW rated cooling capacity, while it leads [15] Assoa YB, Sauzedde F, Boillot B. Numerical parametric study of the thermal and
electrical performance of a BIPV/T hybrid collector for drying applications. Renew
to a longer payback period. Energy 2018;129:121–31.
(3) When applied in Zhuhai where annual solar irradiation is close [16] Emmi G, Zarrella A, De Carli M. A heat pump coupled with photovoltaic thermal
to1400 kWh/m2 and electricity prices are high, this solar trigen- hybrid solar collectors: a case study of a multi-source energy system. Energy
Convers Manage 2017;151:386–99.
eration system achieves the best energy and economic performance. [17] Vaishak S, Bhale PV. Photovoltaic/thermal-solar assisted heat pump system: current
The minimum payback period and the maximum net present value status and future prospects. Sol Energy 2019;189:268–84.
in Zhuhai are 11.8 years and 23.6 k$, respectively. The highest [18] Wang G, Zhao Y, Quan Z, Tong J. Application of a multi-function solar-heat pump
system in residential buildings. Appl Therm Eng 2018;130:922–37.
annual electricity saving in Zhuhai is 26% greater than that in [19] He W, Zhou J, Chen C, Ji J. Experimental study and performance analysis of a
Guangzhou. thermoelectric cooling and heating system driven by a photovoltaic/thermal system
in summer and winter operation modes. Energy Convers Manage 2014;84:41–9.
[20] Liu Z, Zhang Y, Zhang L, Luo Y, Wu Z, Wu J, et al. Modeling and simulation of a
CRediT authorship contribution statement photovoltaic thermal-compound thermoelectric ventilator system. Appl Energy
2018;228:1887–900.
Zeyu Li: Conceptualization, Methodology, Writing - original draft. [21] Riahi A, Ben Haj Ali A, Fadhel A, Guizani A, Balghouthi M. Performance in-
vestigation of a concentrating photovoltaic thermal hybrid solar system combined
Hongkai Chen: Investigation, Writing - review & editing. Yongrui Xu: with thermoelectric generators. Energy Convers Manage 2020;205.
Formal analysis, Validation. Kim Tiow Ooi: Resources. [22] Eicker U, Dalibard A. Photovoltaic–thermal collectors for night radiative cooling of
buildings. Sol Energy 2011;85:1322–35.
[23] Gupta N, Tiwari A, Tiwari GN. A thermal model of hybrid cooling systems for
Declaration of Competing Interest building integrated semitransparent photovoltaic thermal system. Sol Energy
2017;153:486–98.
[24] Mishra GK, Tiwari GN, Bhatti TS. Effect of water flow on performance of building
The authors declare that they have no known competing financial integrated semi-transparent photovoltaic thermal system: a comparative study. Sol
interests or personal relationships that could have appeared to influ- Energy 2018;174:248–62.
ence the work reported in this paper. [25] Hu M, Zhao B, Ao X, Zhao P, Su Y, Pei G. Field investigation of a hybrid photo-
voltaic-photothermic-radiative cooling system. Appl Energy 2018;231:288–300.
[26] Athienitis AK, Barone G, Buonomano A, Palombo A. Assessing active and passive
Acknowledgement effects of façade building integrated photovoltaics/thermal systems: dynamic
modelling and simulation. Appl Energy 2018;209:355–82.
[27] Guo J, Lin S, Bilbao JI, White SD, Sproul AB. A review of photovoltaic thermal (PV/
This work is supported by: (1) Natural Science Foundation of T) heat utilisation with low temperature desiccant cooling and dehumidification.
Guangdong Province under the contract No. 2018A030313310, (2) Renew Sustain Energy Rev 2017;67:1–14.
Guangzhou Municipal Science and Technology Project under the [28] Cheng Q, Zhang X-S, Li X-W. Double-stage photovoltaic/thermal ED regeneration

13
Z. Li, et al. Energy Conversion and Management 215 (2020) 112895

for liquid desiccant cooling system. Energy Build 2012;51:64–72. conditioning and solar photovoltaic systems. Energy Convers Manage
[29] Buker MS, Mempouo B, Riffat SB. Experimental investigation of a building in- 2019;185:836–49.
tegrated photovoltaic/thermal roof collector combined with a liquid desiccant en- [46] Lämmle M, Kroyer T, Fortuin S, Wiese M, Hermann M. Development and modelling
hanced indirect evaporative cooling system. Energy Convers Manage of highly-efficient PVT collectors with low-emissivity coatings. Sol Energy
2015;101:239–54. 2016;130:161–73.
[30] Su B, Han W, Qu W, Liu C, Jin H. A new hybrid photovoltaic/thermal and liquid [47] Hu M, Zhao B, Li J, Wang Y, Pei G. Preliminary thermal analysis of a combined
desiccant system for trigeneration application. Appl Energy 2018;226:808–18. photovoltaic–photothermic–nocturnal radiative cooling system. Energy
[31] Ren H, Ma Z, Lin W, Fan W, Li W. Integrating photovoltaic thermal collectors and 2017;137:419–30.
thermal energy storage systems using phase change materials with rotary desiccant [48] Marrasso E, Roselli C, Sasso M, Tariello F. Analysis of a hybrid solar-assisted tri-
cooling systems. Sustain Cities Soc 2018;36:131–43. generation system. Energies 2016;9.
[32] Buonomano A, Calise F, Palombo A, Vicidomini M. Adsorption chiller operation by [49] Jing Y, Li Z, Chen H, Lu S, Lv S. Exergoeconomic design criterion of solar absorp-
recovering low-temperature heat from building integrated photovoltaic thermal tion-subcooled compression hybrid cooling system based on the variable working
collectors: modelling and simulation. Energy Convers Manage 2017;149:1019–36. conditions. Energy Convers Manage 2019;180:889–903.
[33] Zhai XQ, Wang RZ. A review for absorbtion and adsorbtion solar cooling systems in [50] Puig-Arnavat M, López-Villada J, Bruno JC, Coronas A. Analysis and parameter
China. Renew Sustain Energy Rev 2009;13:1523–31. identification for characteristic equations of single- and double-effect absorption
[34] Alobaid M, Hughes B, Calautit JK, O’Connor D, Heyes A. A review of solar driven chillers by means of multivariable regression. Int J Refrig 2010;33:70–8.
absorption cooling with photovoltaic thermal systems. Renew Sustain Energy Rev [51] Huide F, Xuxin Z, Lei M, Tao Z, Qixing W, Hongyuan S. A comparative study on
2017;76:728–42. three types of solar utilization technologies for buildings: photovoltaic, solar
[35] Yuan W, Ji J, Li Z, Zhou F, Ren X, Zhao X, et al. Comparison study of the perfor- thermal and hybrid photovoltaic/thermal systems. Energy Convers Manage
mance of two kinds of photovoltaic/thermal (PV/T) systems and a PV module at 2017;140:1–13.
high ambient temperature. Energy 2018;148:1153–61. [52] Lu Z. Technical engineering design, thermal experimental and economic simulation
[36] Calise F, d’Accadia MD, Vanoli L. Design and dynamic simulation of a novel solar analysis of absorption cooling/heating systems in China. Energy Convers Manage
trigeneration system based on hybrid photovoltaic/thermal collectors (PVT). 2018;172:155–63.
Energy Convers Manage 2012;60:214–25. [53] Li Z, Liu L. Economic and environmental study of solar absorption-subcooled
[37] Noro M, Lazzarin RM. Solar cooling between thermal and photovoltaic: an energy compression hybrid cooling system. Int J Sustain Energy 2018;38:123–40.
and economic comparative study in the Mediterranean conditions. Energy [54] Su B, Qu W, Han W, Jin H. Feasibility of a hybrid photovoltaic/thermal and liquid
2014;73:453–64. desiccant system for deep dehumidification. Energy Convers Manage
[38] Shirazi A, Taylor RA, White SD, Morrison GL. Transient simulation and parametric 2018;163:457–67.
study of solar-assisted heating and cooling absorption systems: an energetic, eco- [55] National Institute of Standards and Technology. REFPROP; 2019. URL < https://
nomic and environmental (3E) assessment. Renew Energy 2016;86:955–71. www.nist.gov/srd/refprop > .
[39] Buonomano A, Calise F, Palombo A. Solar heating and cooling systems by absorp- [56] Zhang J, Li Z, Jing Y, Xu Y. Performance of solar absorption-subcooled compression
tion and adsorption chillers driven by stationary and concentrating photovoltaic/ hybrid cooling system for different flow rates of hot water. Appl Sci 2020;10:1–18.
thermal solar collectors: modelling and simulation. Renew Sustain Energy Rev [57] Li Z, Ye X, Liu J. Performance analysis of solar air cooled double effect LiBr/H2O
2018;81:1112–46. absorption cooling system in subtropical city. Energy Convers Manage
[40] Li Z, Liu L, Jing Y. Exergoeconomic analysis of solar absorption-subcooled com- 2014;85:302–12.
pression hybrid cooling system. Energy Convers Manage 2017;144:205–16. [58] Fong KF, Lee CK. Dynamic performances of trigeneration systems using different
[41] Yu J, Li Z, Chen E, Xu Y, Chen H, Wang L. Experimental assessment of solar ab- prime movers for high-rise building application: a comparative study. Build Simul
sorption-subcooled compression hybrid cooling system. Sol Energy 2017;10:509–23.
2019;185:245–54. [59] DeST Group. DeST; 2018. URL < http://www.dest.tsinghua.edu.cn > .
[42] Jing Y, Li Z, Liu L, Lu S. Exergoeconomic assessment of solar absorption and ab- [60] Mittelman G, Kribus A, Dayan A. Solar cooling with concentrating photovoltaic/
sorption–compression hybrid refrigeration in building cooling. Entropy thermal (CPVT) systems. Energy Convers Manage 2007;48:2481–90.
2018;20:1–27. [61] Hocine HBCE, Touafek K, Kerrour F. Theoretical and experimental studies of a new
[43] Li Z, Chen E, Jing Y, Lv S. Thermodynamic relationship of subcooling power and configuration of photovoltaic–thermal collector. J Sol Energy Eng
increase of cooling output in vapour compression chiller. Energy Convers Manage 2017;139:021012.
2017;149:254–62. [62] Chen H, Li Z, Xu Y. Evaluation and comparison of solar trigeneration systems based
[44] Chen E, Li Z, Yu J, Xu Y, Yu Y. Experimental research of increased cooling output by on photovoltaic thermal collectors for subtropical climates. Energy Convers Manage
dedicated subcooling. Appl Therm Eng 2019;154:9–17. 2019;199:111959.
[45] Novaes Pires Leite Gd, Weschenfelder F, Araújo AM, Villa Ochoa ÁA, Franca [63] Li Z, Yu J, Chen E, Jing Y. Off-design modeling and simulation of solar absorption-
Prestrelo Neto Nd, Kraj A. An economic analysis of the integration between air- subcooled compression hybrid cooling system. Appl Sci 2018;8..

14