Available online at www.sciencedirect.

com

ScienceDirect
Solar Energy 122 (2015) 12791295
www.elsevier.com/locate/solener

Development and analysis of a solar and wind energy

based multigeneration system
Sinan Ozlu , Ibrahim Dincer
Department of Mechanical Engineering, University of Ontario Institute of Technology, 2000 Simcoe Street North, L1H 7K4 Oshawa, Ontario, Canada
Received 11 April 2015; received in revised form 6 October 2015; accepted 13 October 2015
Available online 19 November 2015
Communicated by: Associate Editor Mukund Patel

Abstract
This paper concerns development and analysis of a solar-wind hybrid multigeneration system. Energy, exergy, exergoeconomic and
exergoenvironmental analyses are performed. The analysis studies are undertaken by developing and constructing the codes in Engineering Equation Solver software. The eects of various input conditions on the system performance are investigated through both energy
and exergy eciencies, and an optimization study is undertaken of system eciency and power output are obtained. The average number
of Toronto suites that the system can supply is calculated. As a result, it is seen that energy and exergy eciencies are higher than equivalent single energy systems. The system has 43% maximum energy eciency and 65% maximum exergy eciency. Maximum turbine
output is 48 kW, while cooling eect is 28 kW and heating eect is 298.5 kW. There is resultant savings of 1614 tons of CO2 per year
by the system. This multigeneration system is capable of supplying at least at a minimum 49 suites.
2015 Elsevier Ltd. All rights reserved.

Keywords: Solar energy; Wind energy; Hybrid; Absorption chiller; Rankine cycle; Renewable energy

1. Introduction
Renewable energy sources (solar, wind, etc.) are attracting more attention as alternative energy sources to conventional fossil fuel energy sources. This is not only due to the
diminishing fuel sources, but also due to environmental
pollution and global warming problems (El-Shatter et al.,
2002). Renewable energy sources are the only clean and
continuous energy solution to satisfy current and future
requirements. A system that can utilize more than three
sources is called a multigeneration energy system. The eciency of multigeneration systems are higher than the combined eciency of separate units. Multigeneration systems
that use renewable sources combine the power of clean
Corresponding author.

E-mail address: Sinan.Ozlu@uoit.ca (S. Ozlu).

http://dx.doi.org/10.1016/j.solener.2015.10.035
0038-092X/ 2015 Elsevier Ltd. All rights reserved.

energy with high eciency. Moreover, they help to supply

dierent needs of a public unit such as a multi-suite building or a neighborhood.
One of the main challenges in this subject is the lack of
study, especially a totally renewable based multigeneration
energy system to produce electricity, cooling, heating, hot
water and hydrogen simultaneously. This has made the
proposed system quite interesting. Eorts to develop more
ecient multienergy systems, attracts many researchers.
Another challenge is the scheduling and modeling of the
sources, solar and wind in this case. Solar energy is not
constant, the intensity changes during the day and
throughout the year. There is no sun at all at night. Moreover, neither wind presence nor wind speed is a factor that
can be guaranteed. The problem can be simplied by making some assumptions such as taking the average solar
intensity and wind speed.

1280

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Nomenclature
A
c
C
D
E_ x
F
F0
g
G
h
k
L
m
m_
P
Q
Q_
s
S
S_
t
T
U
V
W
W_
z

area, m2
specic heat, kJ/kgK
power coecient
diameter, m
exergy rate, kW
view factor
collector eciency factor
gravitational acceleration, 9.81 m/s2
solar radiation rate, W/m2
specic enthalpy, kJ/kg; heat transfer coecient, W/m2 K
conductivity of the absorber tube material, W/
mK
length, m
mass, kg
mass ow rate, kg/s
pressure, kPa
heat, kJ
heat transfer rate, kW
specic entropy, kJ/kg
absorbed solar radiation per unit aperture area,
W/m2
entropy rate, kW
time, s
temperature, K
heat transfer coecient, W/m2 K
velocity, m/s
work, kJ; width, m
work rate, kW
elevation, m

Subscripts
abs
absorption
ap
aperture
con
condenser
cs
cold storage
cv
control volume
d
destructed
dwh
domestic water heater

The amount of solar energy that reaches earths upper

atmosphere is about 1350 W/m2. The atmosphere reects,
scatters and absorbs some of the energy. In Canada,
depending on sky conditions, peak solar intensity varies
from about 900 W/m2 to 1050 W/m2. Peak solar intensity
is at solar noon, when the sun is due south (Cengel et al.,
2011). The amount of the suns energy reaching the surface
of the earth also depends on cloud cover, air pollution,
location and the time of year. An active solar system uses
mechanical equipment to collect, store and distribute heat
from the sun. Active systems consist of solar collectors, a
storage medium and a distribution system. Active solar

e
en
eva
ex
exp
gen
hs
hx
i
L
p
Q
r
rc
sc
t
ts
u
w
wt
0
1. . .70

exit stream
energy
evaporator
exergy
expander
generator
hot storage
heat exchanger
inlet
overall
pump; pressure
heat
receiver; Rankine
ammoniawater Rankine cycle
solar cycle; solar collector
tube
thermal storage
unshaded; useful
wind; water
wind turbine
ambient or reference condition
state numbers

Greek letters
a
absorptivity
g
eciency
n
insulation thickness, m
q
density, kg/m3; specular reectance of the concentrator
s
eect of angle of incidence
Acronyms
COP
coecient of performance
EES
Engineering Equation Solver
ORC Organic Rankine Cycle
PR
pump pressure ratio
PV
photovoltaic

systems are commonly used for; water heating, space conditioning, producing electricity, processing heat and solar
mechanical energy.
When higher temperatures are required, concentrated
solar collectors are used. Solar energy falling on a large
reective surface is reected onto a smaller area before it
is converted into heat. This is done so that the surface
absorbing the concentrated energy is smaller than the surface capturing the energy and as a result can attain higher
temperatures before heat loss due to radiation and the convection wastes of the energy that has been collected (Pavlas
et al., 2006).

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Ahmadi et al. (2014) designed and optimized a novel

solar-based multigeneration energy system. They used
ocean thermal energy conversion which uses a heat engine
for harvesting energy from the ocean. In order to produce
electricity a heat engine operates between the relatively
warm ocean surface, which is exposed to the sun, and the
colder (about 5 C) water deeper in the ocean. They found
that the best optimized point from the multi-objective optimization has an exergy eciency of 60%, while the total
cost rate of the system at this point is 154 $/h.
Ratlamwala et al. (2012) proposed an integrated system,
consisting of a heliostat eld, a steam cycle, an ORC and
an electrolyzer for hydrogen production. The results
showed that the power and rate of hydrogen production
increased with an increase in the heliostat eld area and
solar ux. The optimization study yielded maximum
energy and exergy eciencies and the rate of hydrogen production was 18.74%, 39.6% and 1.57 m3/s respectively.
Ozturk and Dincer (2013) worked on a renewable based
multigeneration energy production system producing a
number of outputs, such as power, heating, cooling, hot
water, hydrogen and oxygen. The solar based multigeneration system with an exergy eciency of 57.4% was
obtained. This was higher than using the sub-systems separately. The parabolic dish collectors had the highest
exergy destruction rate among constituent parts of the
solar based multigeneration system, due to a high temperature dierence between the working uid and collector
receivers. Xu et al. (2011) made an energy and exergy analysis of the solar power tower system using molten salt as
the heat transfer uid. They evaluated both the energy
and exergy losses in each component and in the overall system to identify the causes and locations of the thermodynamic imperfections. Wang et al. (2009) proposed a new
combined cooling, heating and power system driven by
solar energy. The system combines a Rankine cycle and
an ejector refrigeration cycle, which could produce cooling
output, heating output and power output simultaneously.
Since the oil crises of the early 1970, utilization of solar
and wind power has become increasingly signicant,
attractive and cost-eective. However, a common drawback with solar and wind energy is their unpredictable nature. Standalone photovoltaic (PV) or wind energy systems
do not produce usable energy for a signicant portion of
time during the year. In the former case this is mainly
due to dependence on sunshine hours, which are variable
and on relatively high cut-in wind speeds, which range
from 3.5 to 4.5 m/s, in the latter case, resulting in an
underutilization of capacity. In general, the variations of
solar and wind energy do not match with the time distribution of demand. The independent use of the systems results
in considerable over-sizing for system reliability, which in
turn makes the design costly. As the advantages of solar
and wind energy systems became widely known, system
designers have started looking toward their integration
(Desmukh and Desmukh, 2008). The term hybrid

1281

renewable energy system is used to describe any energy system with more than one type of energy source (compared
to systems which use only one source of alternative energy)
to lower costs and increase reliability.
Celik (2003) addressed the sizing and techno-economic
optimization of an autonomous PVwind hybrid energy
system with battery storage. The level of autonomy, i.e.
the fraction of time for which the specied load can be
met, and the cost of the system, were his design parameters.
He showed that the worst month scenarios resulted in too
costly results, so he suggested an alternative solution to
incorporate a third energy source into the system. Notton
et al. (2011) showed that a precise study of renewable
energy potential is indispensable before implementing a
renewable energy system. The solar and wind energy potential is presented for ve sites distributed in a Mediterranean
island and the temporal complementary of these two
energy resources is discussed. Caliskan et al. (2013) performed exergoeconomic and environmental impact analyses, through energy, exergy and sustainability assessment
methods to investigate a hybrid windsolar based hydrogen
and electricity production system.
Kaabeche et al. (2011) proposed an integrated PVwind
hybrid system optimization model, which utilizes the iterative optimization technique following the deciency and
power supply probability, the relative excess power generated, the total net present cost, the total annualized cost
and break-even distance analysis for power reliability and
system costs. Syed et al. (2009) studied the eect that the
integration of the hybrid photovoltaic/wind turbine generation can have on conservation of energy and reduction of
greenhouse gases. They calculated base-case energy
demands using building energy simulation software. Tina
and Gagliano (2011) reported, as a method of evaluating,
a procedure for the probabilistic treatment of solar irradiance and wind speed data. Their results informed the
design of a pre-processing stage for the input of an algorithm that probabilistically optimizes the design of hybrid
solar wind power systems.
The previous studies on solar and hybrid systems discussed in this section help building the model for the proposed system. The objective of studying the proposed
system is to create a hybrid system that can use multiple
sources as substitutes which have multiple outcomes hence
maximizing the eciency.
Solar-wind hybrid systems are studied as these two
sources complement each other in that if one is decient
or inactive the other system comes into eect. Optimization
of the system, by studying the behavior of solar and wind
energy, has been the main focus of the researchers in this
subject. Focus of attention has increased in recent times
on solar energy based multigeneration systems and as a
result more research is being conducted on these
systems every year. This should result in nding new
techniques, increase in the eciencies and a decrease in
operating costs.

1282

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Fig. 1. Conguration of the system.

The multigeneration system studied is a hybrid system

using both solar and wind energy. It utilizes both energy
sources eciently because the extra energy is saved in the
hot storage tank and hydrogen tank. During nighttime or
when there is not strong enough wind, this extra energy
can be used. The system conguration is shown in Fig. 1.
The thermodynamic analysis is based on some assumptions and input data. The assumptions for the system are as
follows:

 Electrolyzer operates with 60% eciency (Ahmadi,

2013).
 There are no pressure changes except through ow
restrictors and pump.
 System is in steady state.
 Wind turbine operates with 90% eciency (Due and
Beckman, 2006).
 Wind turbine operates with the average Toronto wind
speed.
 States at points 1, 4, 8 are saturated liquid.
 State at point 10 is saturated vapor.

 Heat losses from the system boundaries are negligible.

 Possible sources of data noise, e.g. sudden changes in
solar irradiance and electric power demand, are not considered in the analyses (i.e. average hourly values are
used).
 The solar collector operates during the day and stores
heat in the thermal storage units, then uses this energy
during the night.
 The turbine operate with 70% eciency (Ahmadi, 2013).
 The pumps operate with 85% eciency (Ahmadi, 2013).

Input parameters for modeling the system are shown in

Table 1. These parameters have to be set at commencement
in order to perform the other calculations.
Main objective of the multigeneration system studied is
to produce domestic heating, cooling, electricity and
hydrogen for a multi-unit building. The system is designed
to meet the energy needs of a multi-suite facility. A secondary objective is to see the interaction of dierent systems with each other, such as solar energy, wind energy,
Rankine cycle, Absorption chiller, electrolyzer etc. and

2. System description

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Table 1
Input parameters used to model the system.
Collector
Width
Length
Absorber diameter
Transparent envelope outer diameter
Tube material
Receiver eciency
Solar system working uid

2m
2m
25 mm
40 mm
Stainless steel
75%
Therminol 66

Thermal storage
Insulation thickness
Insulation material
Total surface area

30 cm
Polyurethane
6 m2

Rankine cycle
Working uid

Ammoniawater

Wind turbine
Diameter
Average wind speed
Power coecient

34 m
4.2 m/s
60%

Absorption chiller
Evaporator temperature
Condenser temperature
Absorber temperature
Generator temperature

7 C
35 C
40 C
80 C

suggest the optimum conditions in which they should be

combined. The conguration of the system lets the Rankine system makes optimum use of the energy coming from
the solar collectors. The waste heat from the cycle is used
for obtaining hot water and cooling is by means of the
absorption chiller. In this system, the solar energy is collected by a parabolic solar collector. Working uid is Therminol 66. Hot working uid (35) is transferred to the hot
storage tank to be used when there is not enough sun.
From the storage tank it passes through to the boiler
(2930) to heat the ammoniawater mixture in the Rankine
cycle (2021). The evaporated mixture is expanded (2122)
to produce work. Waste heat from the expander is used to
heat domestic water (2627) in the condenser. The pump is
used to pressurize the mixture (1920). The mechanical
energy produced is converted to electrical energy by means
of a turbine and a generator. The rest of the solar energy is
initially used in domestic heating (3017). Subsequently, it
is used in the generator of an absorption chiller for cooling
purposes. The inlet temperature of the generator (17)
should be at least 120 C in order to run the absorption
cooling system. The absorption chiller uses heat instead
of mechanical energy, to provide cooling. The mechanical
vapor compressor is replaced by a thermal compressor that
consists of an absorber, a generator, a pump, and a throttling device. Refrigerant is ammonia in the cycle (789
10). Absorbent is water in the cycle (123456). The
ammonia vapor from the evaporator (10) is absorbed by
the absorber water (61). This solution is then pumped to
the generator where the refrigerant is revaporized (34)
using the remaining solar energy heat source. The ammonia depleted solution is then returned to the absorber via
a throttling device (56). The electricity produced by the

1283

ammoniawater Rankine cycle turbine can be used in residences when there is extra energy or to run the electrolyzer
to produce hydrogen. Wind turbine also produces electricity when there is enough wind. This electricity is also used
in the building or, if there is extra energy, in an electrolyzer
to produce hydrogen.
The modeling of the system is performed using in the
Engineering Equation Solver software. The system is a
hybrid, stand-alone, renewable, multigeneration system.
3. Analysis
This section outlines the model development and analyses carried out in this paper. In the beginning, thermodynamic and exergy analyses are introduced. Then the
environmental analysis is performed. The optimization
approach is explained next. Finally it is shown how electrical and heat loads are calculated.
3.1. Thermodynamic analyses
In order to analyze a control volume, four things need to
be considered; mass balance, energy balance, entropy balance and exergy balance. By writing these equations for
each system and subsystem, equations can be solved
correctly.
3.1.1. Mass balance equation
According to the conservation of mass principle, the net
mass transfer to or from a control volume during time
interval Dt is equal to the mass entering the control volume
minus mass exiting the control volume (Cengel et al., 2011)
as shown in the equation below:
m_ in  m_ out

dmcv
dt

where m_ in and m_ out are the mass ow rate of inlet and

outlet.
3.1.2. Energy balance equation
According to the First Law of Thermodynamics, energy
is conserved. When this principle is considered for a steady
ow system and a control volume, the following equation is
obtained:

X 
V2
Q_ in W_ in
gz
m_ h
2
in


X
V2
_
_
Qout W out
gz
2
m_ h
2
out
where Q_ is heat transfer rate, W_ is work rate, h is specic
enthalpy, m_ is velocity, g is gravitational acceleration,
and z is the elevation.
3.1.3. Entropy balance equation
There is an increase in the sum of the entropies of the
participating systems according to the Second Law of

1284

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Thermodynamics. Entropy balance equation applied to a

control volume can be expressed as
S_ gen

m_ e se 

m_ i si 

X Q_ k

dS CV

Tk
dt

where e denotes exit, i denotes inlet, Q_ is heat transfer rate,

s is the entropy. Positive direction of heat transfer is to the
system.
3.1.4. Exergy balance equation
When an exergy analysis is performed, the thermodynamic imperfections can be quantied as exergy destructions, which represent losses in energy quality or
usefulness (Dincer and Rosen, 2012). The exergy of a substance is often in 4 dierent forms: physical, chemical,
kinetic and potential energy. The last 2 forms are assumed
to be negligible as elevation changes are small and speeds
are low. In addition, chemical energy is not considered as
solar energy is used as a source.
According to second law of Thermodynamics, exergy
balance equation can be written as
X
X
_ Q
_ W Ex
_ D
Ex
4
m_ i exi
m_ e exe Ex
i

where subscripts i and e denote the control volume inlet

_ D is the exergy destrucand outlet ow respectively and Ex
tion rate. Other terms are explained as:
_ W is the exergy transfer associated with mechanical
Ex
_ W W_ .
_ W is shown as Ex
work. The equation of Ex
_ExQ is the exergy transfer associated with heat transfer
and it depends on the temperature at which it occurs in relation to the temperature of the environment (Caliskan et al.,
_ Q is shown as Ex
_ Q 1  T 0 Q_ i .
2013). The equation of Ex
Ti
where T0 is the ambient temperature, Ti is the temperature
of the heat transferred to the boundary of the control
volume and Q_ i is the heat transfer to the control volume.
After introducing general equations, the equations for
each component in the system are shown. In order to solve
the unknown parameters of the system, each component
should be solved on a one by one basis, before proceeding
to the next one in order to arrive at a solution.
3.1.5. Thermodynamic analyses of system components
 Parabolic solar collector
A parabolic collector can accept both beam and diuse
radiation because of its large acceptance angle. Actual useful energy gain in the collector is expressed as (Due and
Beckman, 2006):


Ar
Qu F R Au S  U L T 34  T 0
5
Ao
where Au is the unshaded area of the concentrator aperture
(m2), Ar is the area of the receiver (m2), S is the absorbed
solar radiation per unit of aperture area (W/m2) and can
be found from:

S I ap sqa

where I ap is the eective incident radiation measured on the

plane of the aperture (W/m2), s is the eect of angle of incidence, q is the specular reectance of the concentrator.
F R can be found from the following equation:



_ p
Ar U L F 0
mc
1  exp
FR
7
_ p
Ar U L
mC
where m_ is the mass ow rate of the heating uid (kg/s), cp
is the specic heat of the heating uid (kJ/kg K), U L is the
overall heat transfer coecient (W/m2 K).
F 0 can be found from the following equation:
F0

1
UL
1
UL

hDt Do i

Do
2k

ln DDoi

where ht is the heat transfer coecient inside the tube (W/

m2 K), Di is the absorber inside diameter (m), k is the conductivity of the absorber tube material (W/m K).
 Hot storage
If an insulation of thickness n (m) and thermal conductivity k (W/m K) is used, the coecient of heat transfer U
(W/m2 K) between the working uid and air is given by
1 1 n

U h k

where h is the coecient of heat transfer from working

uid to air (W/m2 K).
The corresponding hot storage heat loss Q_ hs (W/m2) per
unit area of the surface of the tank is given by
Q_ hs U T 35  T 0 .
where T 35 is the temperature of the uid entering the hot
thermal storage (K) and T 0 is the atmospheric temperature.
The recommended type of insulation is 20 cm mineral wool
insulation (Desmukh and Desmukh, 2008).
If all the parameters are the same, cold storage heat loss
can be found by inserting T 18 instead of T 35 .
where T 18 is the temperature of the uid entering the
cold thermal storage (K).
 Domestic water heater
The hot gases from the ammoniawater Rankine cycle
boiler enter the water heater to warm domestic hot water
to 60 C. Water enters this heater at atmospheric pressure
and ambient temperature. The energy balance for this component is given as follows:
m_ sc h30  h17 m_ w h32  h31

10

 Absorption chiller
The rate of heat to the generator of an absorption system is provided using solar energy and calculated using
the following equation:
Q_ gen m_ sc h17  h18

11

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

where m_ sc is the mass ow rate of the working uid in the

solar cycle (kg/s). In order to obtain the outlet conditions
of the generator, the following equation is used:
m_ 3 h3 Q_ gen m_ 4 h4 m_ 7 h7

12

The mass balance equations for the heat exchanger are

given as: m_ 2 m_ 3 and m_ 4 m_ 5 . The energy balance equation for heat exchanges is given below:
m_ 2 h3  h2 m_ 4 h4  h5

13

The mass balance equation for the condenser is given as

m_ 7 m_ 8 . The energy balance equation for the condenser is
given below:
m_ 8 h8 Q_ con m_ 7 h7

14

The equation for mass balances for the evaporator is

given as m_ 9 m_ 10 . The equation for energy balance for
the evaporator can be written as follows:
m_ 9 h9 Q_ eva m_ 10 h10

15

The energy balance equation that is used to calculate the

heat rejected from the absorber is as follows:
m_ 6 h6 m_ 10 h10 m_ 1 h1 Q_ abs

16

The work done by the pump is calculated using the

equation given below:
W_ p m_ 1 h2  h1

17

The energetic COP is found using the following

equation:
COP en

Q_ eva
_Qgen W_ p

18

The exergetic COP can be expressed as follows:

COP ex

_ eva
Ex
_ gen W_ p
Ex

19

 Rankine cycle
The power that can be obtained from the cycle is dened
as:
W_ t;rc m_ rc h21  h22

20

gex

W_ net;rc
_ boiler;rc
Ex

21

The rate of heat rejected by the condenser is dened as:

Q_ con;rc m_ rc h22  h19

22

The energy eciency of ammoniawater Rankine cycle

is dened as:
gen

W_ net;rc
Q_ boiler;rc

23

The exergy eciency of ammoniawater Rankine cycle

can be expressed as follows:

24

 Wind turbine
Average power obtained from the wind turbine is
expressed as follows (Due and Beckman, 2006):
1
P w gwt qair Awt C p V 3
2

25

where gwt is the wind turbine eciency, qair is the air density
(kg/m3), Awt is the wind turbine area (m2), C p is the turbine
power coecient, V is the average velocity of the wind (m2/
s).
Exergy eciency of the wind turbine is:
gex

W_ wt
_ wt
Ex

26

_ wt is the exergy of the wind turbine and calculated

where Ex
as:
_ wt 1 qair Awt V 3
Ex
2

27

3.2. Exergy analysis

For exergy analysis, exergy destruction in each component of the system, energy and exergy eciency equations
are introduced here.
3.2.1. Exergy balance equations
The exergy destructions for basic components in the system are listed in Table 2. The expressions are based on state
points shown in Fig. 1. Exergy of each state point can be
calculated using EES software based on state pressure
and temperature.
3.2.2. Energy eciency
Energy eciency is dened as the ratio of useful energy
produced to the input energy supplied to the system. In the
system, energy eciencies of ammoniawater Rankine
cycle can be calculated by the following formula:

The power consumed by pump is expressed as:

W_ p;rc m_ rc h20  h19

1285

gen;rc

W_ exp  W_ p;rc
Q_ boiler

28

System eciency equation can be expressed as follows:

gen;system

W_ exp W_ wind  W_ p;solar  W_ p;rc

Q_ solar W_ wind

29

Also energy coecient of performance for the absorption chiller can be calculated as follows:
COP en;chiller

Q_ eva
Q_ gen W_ p

30

1286

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Table 2
Exergy destruction rates of components in the system.
Component

Exergy destruction rate expression

Parabolic solar collector

Hot storage
Cold storage
Domestic water heater

_ 34  Ex
_ 35
_ d;sc Q_ solar 1  T 0 Ex
Ex
T sun
_ d;hs Ex
_ 35  Ex
_ 29  Ex
_ Q
Ex
_ d;cs Ex
_ 18  Ex
_ 33  Ex
_ Q
Ex
_ d;dwh Ex
_ 30  Ex
_ 17 Ex
_ 31  Ex
_
Ex


 32 








_Exd;sc Q_ solar 1  T 0  Q_ dwh 1  T 0  Q_ boiler 1  T 0  Q_ gen 1  T 0  Q_ hs 1  T 0  Q_ cs 1  T 0  W_ p;solar
T sun
T dwh
T boiler
T gen
T hs
T cs

Solar cycle
Absorption
Absorption
Absorption
Absorption
Absorption
Absorption
Absorption
Absorption

_ d;gen Ex
_ 3  Ex
_ 4  Ex
_ 7 Ex
_ 17  Ex
_ 18
Ex
_ d;con Ex
_ 7  Ex
_ 8 Ex
_ 13  Ex
_ 14
Ex
_ d;valve Ex
_ 8  Ex
_ 9
Ex
_ d;eva Ex
_ 9  Ex
_ 10 Ex
_ 11  Ex
_ 12
Ex
_ d;abs Ex
_ 10 Ex
_ 6  Ex
_ 1 Ex
_ 15  Ex
_ 16
Ex
_ d;p Ex
_ 1  Ex
_ 2 W_ p;abs
Ex
_ d;throttle Ex
_ 5  Ex
_ 6
Ex
_ d;hx Ex
_ 2  Ex
_ 3 Ex
_ 4  Ex
_ 5
Ex




T
0
_ d;abs Q_ gen 1 
 Q_ eva 1  T 0  W_ p;abs
Ex

generator
condenser
expansion valve
evaporator
absorber
pump
throttling valve
heat exchanger

Absorption cycle

T gen

Boiler
Ammoniawater Rankine pump
Ammoniawater Rankine condenser
Expander
Ammoniawater Rankine cycle
Wind turbine

T eva

_ d;boiler Ex
_ 20  Ex
_ 21 Ex
_ 29  Ex
_ 30
Ex
_ d;p rc Ex
_ 19  Ex
_ 20 W_ p;rc
Ex
_ d;rc con Ex
_ 22  Ex
_ 19 Ex
_ 26  Ex
_ 27
Ex
_ d;exp Ex
_ 21  Ex
_ 22
Ex




_ d;rc Q_ boiler 1  T 0  Q_ con 1  T 0 W_ exp  W_ p;rc
Ex
T boiler
T con


_ d;wt 1  1  W_ wt
Ex
C p;wt

Table 3
Average emissions from US power plants. Source: United States
Environmental Protection Agency Clear Energy (2015).

Natural gas
Coal
Oil

CO2 (kg/MW h)

SO2 (kg/MW h)

NOx (kg/MW h)

515.00
1020.00
758.00

0.04
6.00
5.00

0.77
3.00
2.00

Exergy coecient of performance for the absorption

chiller can be calculated as follows:


Q_ eva 1  TTeva0


34
COP ex;chiller
Q_ gen 1  TTgen0 W_ p
3.3. Environmental analysis

3.2.3. Exergy eciency

Exergy eciency is the product exergy output divided by
the exergy input. Solar system exergy eciency formula
can be written as follows:






T0
0
Qboiler 1  T boiler
Qgen 1  TTgen0 Qdwh 1  TTdwh


gex;sc
Q_ solar 1  TTsun0
31
Exergy eciency of the Rankine cycle can be expressed
as follows:


W exp  W p;rc Qcon 1  TTcon0
gex;rc
32
Qboiler
System exergy eciency is dened as follows:

gex;system

Air polluting emissions such as nitrogen oxides (NOx),

carbon dioxide (CO2), sulfur dioxide (SO2), methane
(CH4), and mercury (Hg) compounds associated with generating electricity, heat and hydrogen from solar technologies are negligible because no fuels are combusted.
As there is no pollution caused by these systems, air
emissions from fossil fuel red power plants with an equivalent power output can be calculated and saved emissions
can be determined (Dincer et al., 2013).
According to the United States Environmental Protection Agency, average emissions emitted by power plants
with respect to the fuel are shown in Table 3. As natural
gas is widely used in North America, calculations are based
on the usage of natural gas.
Environmental analysis is made based on the emissions
produced using fossil fuels to achieve the same results.







0
W_ exp Q_ con 1  TTcon0 Q_ eva 1  TTeva0 Q_ dwh 1  TTdwh
W_ wt  W_ p;sc  W_ p;rc



3
q A V
Q_ solar 1  TTsun0 air wt2 wind

33

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

1287

3.4. Optimization

Table 4
Building characteristics of units considered. Source: Binkley (2012).

Optimization is maximizing or minimizing an objective

function by manipulating with the independent variables
considering the constraints and the boundaries. In this section, the main parameters of optimization such as objective
function, decision variables, bounds and trial and error values are introduced.

Average
Average
Average
Average
Average

3.4.1. Objective function

Objective function is the variable to be minimized or
maximized depending on the targets of the decision maker.
In multigeneration systems objective function can be as
follows:
 Eciency (energy, exergy etc.).
 Cost (investment, annualized costs, cost of exergy
destruction etc.).
 Emitted pollutants (CO2, SO2, NOx etc.).
If more than one objective is chosen, it is called multiobjective optimization (Dincer and Rosen, 2012).
3.4.2. Decision variables
Decision variables are the variables that aect the objective function. It is necessary to select as many independent
variables as there are degrees of freedom. However only the
most important variables with a major eect should be chosen. Examples for the decision variables are solar radiation,
ambient temperature, number of solar units, condenser
outlet temperature, pump pressure ratio etc.
3.4.3. Bounds
Each independent variable requires lower and upper
bounds. These bounds are specied for the following
properties:
 Dimensions or weight of the system.
 Highest temperature that uids and the components are
used having regard to safety.
 Highest pressure allowed by the uids and the
components.
 Maximum ow rate of the working uids.
 Minimum temperature that the components or working
uids can operate.

3.4.4. Selecting value(s)

Choosing (by trial and error) the correct value(s)
improve the likelihood of nding an optimum. Incorrect
selection may result in it taking too long or even be impossible, for the program to converge into a solution.
3.5. Estimation of heating, cooling, electricity loads
When designing multigeneration systems, it is important
to determine the target of the output. The systems designed

number of storeys per building

number of suites per building
date of construction
gross oor area per building (m2)
attributed suite area (m2)

13
188
1984
18,400
104

in this paper are for a Toronto multi-unit residential building. The building characteristics considered are in Table 4.
The average energy intensities per suite and per building
are shown in Table 5. Energy intensity is a buildings
annual energy consumption per unit of gross oor area.
The table shows that the annual energy needed to heat a
suite in Toronto in equivalent kW h. Natural gas consumption in cubic meters is used to nd the energy intensity. The
conversion from cubic meters of natural gas supplied to
equivalent kilowatt-hours of energy was based on a factor
of 37.08 MJ/m3 or 10.3 kW h/m3 (Binkley, 2012).
According to Binkley (2012), the average end-use distribution for Toronto buildings is 38% electricity, 37% space
heating (30% electricity and 70% natural gas), and 25%
domestic hot water. The annual energy intensity is based
on the total annual energy consumed from both electric
and natural gas sources divided by the buildings gross
oor area. The ratio of electricity is 38% and natural gas
is 62%.
In order to calculate the percentage of electricity for
cooling, historical data related to degree days for Toronto
was obtained from Toronto Hydro (2015).
The average annual degree days for heating for Toronto
between 2001 and 2013 were 3638, whereas average annual
degree days for cooling was 380. As a result the cooling
load is 10% of heating load. By sharing the energy intensity
to loads based on the percentages, the loads in equivalent
annual kW h in Table 6 are found.
4. Results and discussion
In this section system results are analyzed. This is done
by laying out comparison graphs and optimizing the system parameters.
4.1. Thermodynamic modeling results
The results obtained from the system are tabulated in
Table 7. These outputs are subject to change depending
on the parameter that is under consideration. For maximum eciency, eciency may increase while other parameters would drop. This is discussed in optimization section.
The exergy destruction for the main components of the
system are shown in Fig. 2. The last column shows the total
exergy destruction in the system.
Highest exergy destruction occurs in the solar system
while the lowest is in the absorption chiller. This means
that due to irreversibilities, 251 kW of energy is lost in

1288

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Table 5
Annual natural gas intensity of a suite in Toronto. Source: Binkley (2012).
Annual value
per suite (ekW h)

295

25,100

Electricity (cooling + the rest)

Space heating
Domestic hot water

Annual value
per suite (ekW h)

112 (11 + 101)

109
74

9538 (929 + 8609)

9287
6275

298.5
28
48
43
65
1.96
0.26
0.80
0.31
1613
485

600

Exergy Destrucon Rate (kW)

ex (N=4)

0.3
0.25

0.15
200

500
400
300
200
100

Rankine
cycle

Wind
Turbine

Domesc
Water

System

Fig. 2. Exergy destructions in the system.

the wind turbine and 485 kW is lost in the system. An

exergy destruction graph is a useful to tool to focus on
the sources of irreversibilities.
It is shown in Fig. 3 that the eect of solar radiation on a
number of solar units and corresponding the eciency
levels. When there are 2 solar units installed, both system
energy and exergy eciencies drop radically, they stabilize
after around 800 W/m2. The reason for this is that solar
radiation increases but the output does not increase at
the same rate as the solar radiation. When there are 4 solar
units installed, both energy and exergy eciencies are lower
compared to 2 solar units for low solar radiation. Eciencies for 4 units are greater than that for 2 units after 600

300

400

500

600

700

800

900

1000

Solar radiaon (W/m2)

Fig. 3. System energy and exergy eciencies vs. solar radiation.

0.34
0.32

en

ex

0.3

Eciency

Q_ heating (kW)
Q_ cooling;absorption (kW)
Maximum W_ turbine (kW)
Maximum gmulti (%)
Maximum wmulti (%)
m_ H2 (kg/h)
m_ dwh (kg/s)
Absorption chiller COP en
Absorption chiller COP ex
CO2 emissions saved (tons/year)
Total exergy destruction rate (kW)

Solar system Absorpon

Chiller

en (N=4)

0.2

Annual value
per m2 (ekW h/m2)

Table 7
Parameter values resulting from energy and exergy analyses of the system.

ex (N=2)

0.35

Table 6
Design loads of the system.
Load

en (N=2)

0.4

Eciency

Natural gas energy intensity

Annual value
per m2 (ekW h/m2)

0.45

0.28
0.26
0.24
0.22
0.2

10

15

20

25

30

35

40

45

50

T0 (C)

Fig. 4. System energy and exergy eciencies vs. ambient temperature.

and 700 W/m2 for exergy and energy eciencies respectively. The reason for this phenomena is that when there
are 4 solar units installed, for higher solar radiation the
increase in output is much higher than 2 solar units compared to the increase in solar heat input.
Ambient temperature (T0) aects system energy and
exergy eciencies indirectly in Fig. 4. Energy eciency
drops because the increase in the output is lower than the
increase in solar energy. Exergy eciency drops at a faster
rate because it is more aected by the increase in ambient
temperature. As reference temperature increases, the temperature dierence of the process temperature drops. If
the process temperature does not increase at the same rate
as the reference temperature, the exergy value drops.
As it is shown in Fig. 5, the pressure ratio of the Rankine cycle pump aects the Rankine cycle and system eciencies. System eciencies increase at the same rate with
increasing pressure ratio. However, while Rankine cycle
exergy eciency increases at a faster rate, energy eciency
increases even more so, then stabilizes at a value of 10%
after a pressure ratio of 20. The reason for the fast rate
of increase in the Rankine cycle energy eciency at low
pressure ratios is that by increasing pressure ratio, the
increase in output drops and stabilizes. The reason for
the slower rate of increase in exergy eciency is because
heat output from the Rankine cycle drops. However, as
the work output increase surpasses heat output increase,
the eciency increases.
According to Fig. 6, by increasing boiler outlet temperature all the eciencies drop. The drop is more signicant

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

1289

ex,rankine

en,system

Eciency

0.4

0.3

0.2

0.1

32

ex,system

Rankine Cycle Heat Output (kW)

en,rankine

12

17

Qout

270

30
250

28

230

26

210

24

190

22

170

20

150
100

22

Wout

120

140

160

Rankine Cycle Work Output (kW)

290
0.5

18
200

180

Boiler Outlet Temperature (C)

Pressure Rao

Fig. 5. System and Rankine cycle eciencies vs. Rankine cycle pump
pressure ratio.

Fig. 7. Rankine cycle heat and work outputs vs. boiler outlet temperature.

ex,system

en,rankine

0.6
0.5
0.4

0.2
0.1
0

Rankine cycle mass ow rate (kg/s)

Fig. 8. System and Rankine cycle eciencies vs. Rankine cycle mass ow
rate.

has to work harder to pump high mass ow rate uid.

The increase in work output is not as high as work input
as a result Rankine cycle energy and exergy eciencies
drop.
In Fig. 9 the system and solar system energy and exergy
eciencies versus solar system mass ow rate are analyzed.
All the eciencies drop with an increasing mass ow rate
of the solar system. The most signicant drop is in solar

ex,rankine

en,system

ex,system

ex,solar

0.3
0.25

Eciency

0.7

Eciency

ex,rankine

0.3

0.8
0.6
0.5
0.4

0.2
0.15
0.1

0.3

0.05

0.2
0.1
0
100

en,rankine

0.7

0.35
en,system

ex,system

0.8

1
0.9

en,system

0.9

Eciency

in Rankine cycle exergy eciency, then in system exergy

eciency. System energy and exergy eciencies are aected
negatively because heat output from the Rankine system
drops. Although heat output to the domestic water heater
increases, this increase is not enough to increase the eciency. The reason for the drop in Rankine cycle eciencies
is due to the decrease in work and heat outputs from the
Rankine system.
Both work output and heat output from the Rankine
cycle are negatively aected by an increase in boiler outlet
temperature as seen in Fig. 7. When the boiler outlet temperature increases, less heat is transferred to the Rankine
cycle, hence reducing the heat and work outputs from the
system. The reason for the irregularity in the work output
graph is because the amount of data is not sucient to
make the curve smooth. If larger amount of data is taken
to plot the graph, it would be a smooth curve.
It is shown in Fig. 8 that when the mass ow rate of the
Rankine cycle is increased, the energy and exergy eciency
of Rankine cycle drops while system energy and exergy eciencies increase. The rate of increase is the same in system
eciencies while rate of drop is higher in Rankine cycle
exergy eciency than energy eciency. The reason for
the drop in Rankine cycle eciencies is pump work input
increases when the mass ow rate increases. The pump

0
0.5
120

140

160

180

200

1.5

2.5

3.5

4.5

Solar System Mass Flow Rate(kg/s)

Boiler Outlet Temperature (C)

Fig. 6. System and Rankine cycle eciencies vs. boiler outlet temperature.

Fig. 9. System and solar system eciencies vs. solar system mass ow
rate.

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

1
COP_en

0.9

COP_ex

en,system

ex,system

0.8
0.7

COP/Eciencies

system and system exergy eciencies, followed by solar

system energy eciency. All the graphs are parabolic with
the rate of drop decreasing. Eciencies decrease because
the pump in the solar system has to work much harder
to pump working uid. As a result, work input in the system increases but the output heat and work do not increase
at the same rate as the work input. The reason that solar
system exergy eciency drops faster than system eciencies is that solar system exergy increases at a much higher
rate than the system eciency parameters.
Cold storage outlet temperature aects the eciencies.
All the eciencies increase with increasing outlet temperature as shown in Fig. 10. The eect is more signicant in
the solar system exergy eciency. The rate of increase of
solar system eciency is much higher than system eciencies because the heat input to the Rankine cycle increases at
a high rate (which is an output for the solar cycle). The rate
of increase is the same in system eciencies. System eciencies increase because of the increase of work and heat
output from Rankine cycle.
It is shown in Fig. 11 that the energy coecient of performance (COP) of absorption chiller increases with
increasing evaporator temperature while exergy COP
drops. Energy COP straight-lines after 18 C. System
energy eciency stays constant while system exergy eciency decreases slightly. The reason for the increase of
energy COP is due to the increase in evaporator heat input
with increasing evaporator temperature. Exergy COP
decreases because evaporator temperature is an eective
factor for exergy COP calculation. Numerator is
Q_ eva 1  TTeva0 for COP exergy calculation. Because of the
increase in evaporator temperature, heat output from the
evaporator increases, and exergy COP decreases.
All the COPs and eciencies drop with increasing
absorption chiller condenser temperature as Fig. 12 shows.
The drop is more signicant for the COPs while eciencies
are almost constant, dropping slightly. The highest drop is
in the energy COP. Because of the output of the absorption
chiller, evaporator heat drops at a faster rate. The drop for
the exergy COP is slower that energy COP because, during
the calculation of exergy COP, evaporator heat is
multiplied by a factor which reduces the decrease rate.

0.6
0.5
0.4
0.3
0.2
0.1
0

11

13

15

17

19

Absorpon Chiller Evaporator Temperature(C)

Fig. 11. Absorption chiller COPs and system eciencies vs. absorption
chiller evaporator temperature.

1
COP_en

0.9

COP_ex

en,system

ex,system

0.8

COP/Eciencies

1290

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

25

27

29

31

33

35

37

39

Absorpon Chiller Condenser Temperature(C)

Fig. 12. Absorption chiller COPs and system eciencies vs. absorption
chiller condenser temperature.

All the COPs and eciencies drop by increasing absorption chiller absorber temperature as shown in Fig. 13. The
graphs are exactly the same as absorption chiller condenser
temperature graphs with a shift of 5 C. The drop is more
signicant for the COPs while eciencies are almost constant, dropping slightly. The highest drop is in the energy
COP because evaporator heat (which is the output of the
absorption chiller), drops at a faster rate. The drop for
the exergy COP is slower than energy COP because during

0.34

en,system

ex,system

0.32

COP_ex

en,system

ex,system

0.8

COP/Eciencies

0.3

Eciency

COP_en

0.9

ex,solar

0.28
0.26
0.24

0.7
0.6
0.5
0.4
0.3
0.2

0.22
0.2

0.1

10

15

20

25

30

35

40

45

50

Cold Storage Outlet Temperature(C)

Fig. 10. System and solar cycle eciencies vs. cold storage outlet
temperature.

0
30

32

34

36

38

40

42

44

Absorpon Chiller Absorber Temperature(C)

Fig. 13. Absorption chiller COPs and system eciencies vs. absorption
chiller absorber temperature.

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

the calculation of exergy COP, evaporator heat is multiplied by a factor which reduces the drop rate.
As it is shown in Fig. 14, all the parameters except
energy COP increase with the increasing absorption chiller
generator temperature. The changes are not very signicant
for eciencies while COPs change more signicantly.
Energy COP increases because evaporator heat input
increases. Exergy COP drops because as generator temperature increases, the denominator also increases, thereby
decreasing the fraction.
4.2. Optimization results
In this section signicant output parameters are tested
for optimization by nding the appropriate values of the
input parameters. Optimization is done by EES Min/Max
property.
The function to be maximized is the power output from
the Rankine cycle. The independent variables, system
bounds, optimum values and maximum power output are
listed in Table 8. Minimum and maximum values for each
variable are determined based on physical limits.
After running the maximization property of EES, the
result is achieved by solving 330 equations by 199 iterations
in 10.9 s. Maximum number of iterations is set initially to
have the opportunity to abort the optimization if an optimum value is not found in a specic time. The results show
that for a solar system ow rate of 1.38 kg/s, Rankine cycle
pressure ratio of 100, ambient temperature of 50 C, boiler
output temperature of 100 C, solar radiation of 1000 W/
m2, the power output is maximized at 47.75 kW.
System energy and exergy eciency are the parameters
that have to be maximized. Five independent variables that
have an eect on the eciency are chosen as shown in
Table 9. These are; solar radiation, solar cycle mass ow
rate, ambient temperature, Rankine cycle pressure ratio
and boiler exit temperature. These variables are input values of the model created in EES. Minimum and maximum
values of these variables can also be seen in Table 9.
The value of the variables to reach maximum eciency
in the system are in the Opt gen column in Table 9.
The last row shows the result of the energy eciency

1291

Table 8
Independent variables of the system for power output maximization.
Variable
2

Solar radiation, G (W/m )

Ambient temperature, T0 (C)
Rankine cycle pressure ratio, PR
Solar cycle mass ow rate, m_ oil (kg/s)
Boiler exit temperature, T30 (C)
Maximum power output (kW)

Min

Max

Opt

0
0
2
0.50
100

1000
50
100
5.00
150

1000
50
100
1.38
100
47.75

optimization. Maximum energy eciency of the system is

43.2% if the parameters are chosen as shown in Table 9.
In order to check maximum exergy eciency of the system,
another run is performed. The optimum variables are
shown in Table 9 in Opt gex column. The maximum
value of exergy eciency is 65% for the system.
4.3. Environmental impact assessment results
It is shown in Table 3 that natural gas produces 515 kg/
MW h CO2, 0.04 kg/MW h SO2 and 0.77 kg/MW h NOx.
As shown in Table 7, heating load is 298.5 kW, cooling
load is 28 kW and maximum power output from the turbine is 48 kW, also shown in the optimization section.
When these three outputs are added, total output of the
system is found as 375.5 kW. The system saves heat in
the thermal storage and assumed to work 24 h/day and
365 days/year. So the annual output becomes 3289 MW h.
If the same output is obtained from a plant that works
by natural gas, 1694 tons of CO2 would be emitted. It
can be calculate that 131 kg of SO2 and 2.5 tons of NOx
is emitted as well. Natural gas is the source that produces
less emissions than coal or oil. The amount of greenhouse
gases saved instead of using natural gas, coal and oil are
shown in Table 10.
4.4. Loading results
This system is designed to supply a multi-unit building.
The average load for an average Toronto, Ontario suite is
used in the calculations. The cooling load obtained from
the system is 28 kW. This makes 245.3 MW h/year.
According to Table 5, this much load is enough to supply
a 264 suite building. When the power output is considered,
it can be shown by the same method that the system
can supply 49 units. The heating load is used for two
purposes; space heating and domestic hot water. When
these two are combined, it can be seen that there is enough
load for a 168 suite building. The results are shown in
Table 11.
4.5. Comparison with experiment results

Fig. 14. Absorption chiller COPs and system eciencies vs. absorption
chiller generator temperature.

In order to compare theoretical results with experimental data, a trigeneration system setup designed and built by
Tarique (2014) is used. The studied system has outputs the
same as the experimental setup. The calculations of the

1292

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Table 9
Independent variables of the system for eciency maximization.
Variable
2

Solar radiation, G (W/m )

Solar cycle mass ow rate, m_ oil (kg/s)
Ambient temperature, T0 (C)
Rankine cycle pressure ratio, PR
Boiler exit temperature, T30 (C)
Maximum eciency

Min

Max

Opt gen

Opt gex

0
0.500
10
1
100

1000
5.000
35
100
150

100
1.989
10
20
100
43:2%

660
1.989
10
1
100
65%

Table 10
Emissions saved if fossil fuels were used.

Natural gas
Coal
Oil

CO2 (tons/year)

SO2 (tons/year)

NOx (tons/year)

1694.00
3355.00
2493.00

0.13
20.00
16.00

2.50
10.00
7.00

Table 11
Number of suites supplied by the system based on loads.
Load

Number of suites supplied

Cooling
Electricity except cooling
Space heating + domestic hot water

264
49
109

initial and nal conditions in the system are compared with

the experimental results. The results of the experiments are
important and provide a deeper understanding of the

Table 12
Ammoniawater based trigeneration system with Rankine and ejector
cycle integration. Source: Tarique (2014).
Process

Description

12
23
345
5615

Pressurization of liquid (pumping)

Preheating (regenerator)
Vapor generation
Flow splitting of superheated vapor (#5 toward expander,
#15 toward ejector)
67
Vapor expansion
78
Regeneration (heat release from two-phase ow)
8916
Mixing of streams #8 and #16
910
Ammonia resorption and incomplete condensation
1011
Complete condensation and heat rejection
11112 Liquid ow splitting (#1 toward pump, #12 toward throttling
valve)
1314
Evaporation and heat absorption from cooling process
141516 Ejector process (1416 compression, 1516 expansion)
1718
Cold liquid injection for lubrication

Fig. 15. Simplied version of ammonia water based trigeneration system (adapted from Tarique (2014)).

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Table 13
Performance comparison of experimental system with the proposed
system. Source: Tarique (2014).
Quantity

Experimental system

Proposed system

Net generated work

Generated heat
Generated cooling
Heat input
Energy input
Generated heat exergy
Exergy of cooling eect
Energetic COP
Exergetic COP

69.1 kJ/kg (2.6%)

1930 kJ/kg (73%)
226 kJ/kg (11%)
2105 kJ/kg
537.1 kJ/kg
176 kJ/kg (8.4%)
19.13 kJ/kg (1%)
1.06
0.49

21.83 kJ/kg (6.2%)

53.39 kJ/kg (15%)
27.94 kJ/kg (8%)
331 kJ/kg
17.37 kJ/kg
9 kJ/kg (2.7%)
1.8 (0.5%)
0.80
0.31

processes. This also allows for validation of the thermodynamic analysis results.
In order to utilize low-grade heat to generate power,
cooling eect and hot water, a test bench is being built.
The trigeneration system uses ammoniawater as a working uid. The test bench consists of an expander, an air
cooled condenser, a compressor, an evaporator, shell and
tube heat exchanger, and auxiliary components. This integrated system combines power and cooling cycles, where
the source heat is used to generate power through a scroll
expander and a portion of the heat is used in an ejector
cooling system. The residual heat, which is normally

1293

released to the environment in this type of power cycle, is

captured for hot water heating or space heating (Tarique,
2014). A simplied version of the experimental setup is
shown in Fig. 15. Black arrows signify heat uxes, gray
arrow signies cooling and white arrow signies power.
The processes in the system are explained in Table 12.
The comparison of the outputs from the experimental
setup and the system designed is shown in Table 13. As
the quantity of the inputs (heat inputs) are dierent, each
result is given as a percentage of the heat input.
Due to unprocessed data, eciencies, heat losses and
dierent system congurations, the results do not match.
In the proposed system, energy is distributed to system
components in dierent proportions. Heating, cooling,
electricity or other outputs can be favored depending on
the application. The same system can be run to demonstrate dierent outputs. This is one of the reasons that
the output of the system and experimental setup do not
necessarily match. The proposed system and the experimental setup have similar outputs but with some major
changes. For example in the proposed system wind energy
and electrolyzer dier from the experimental setup. In addition, in the experimental setup, there is a quadruple eect
absorption chiller which is dierent than a single eect
absorption chiller. System input parameters other than

Fig. 16. Similar multigeneration system (adapted from Ozturk and Dincer (2013)).

1294

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

heat input can be adjusted to achieve the same result as the

experimental results.
4.6. Comparison with similar system
Ozturk and Dincer (2013) studied a system similar to the
one under review. The dierences between the systems are
the lack of wind energy supply and an additional Rankine
cycle. A comparison of the two systems is performed in this
section.
The system reviewed by Ozturk and Dincer (2013) produces a number of outputs, such as power, heating, cooling, hot water, hydrogen and oxygen. The solar-based
multi-generation system consists of four main subsystems: Rankine cycle, organic Rankine cycle, absorption
cooling and heating, and hydrogen production and utilization. They carried out exergy destruction ratios and rates,
power or heat transfer rates, energy and exergy eciencies
of the system components. The system is shown in Fig. 16.
According to the analysis, this systems exergy eciency
varies between 55% and 58%, where the exergy destruction
rate varies between 9500 and 12,000 kW in the ambient
temperature of 1030 C. The exergy eciency of the proposed system at the same temperatures varies between 29%
and 32% which is lower than the system being used for
comparison purposes. The system has a maximum energy
eciency of 52.82% and a maximum exergy eciency of
57.39% whereas the proposed system has a maximum
energy eciency of 43% and a maximum exergy eciency
of 65% as summarized in Table 7.
The results of the two systems are similar, especially the
maximum system eciency values. The reasons for the differences are dierent congurations of the systems,
assumptions and the calculation methods.
5. Conclusions
This study focuses on developing a novel multigeneration energy system using solar and wind energy to
meet all the energy requirements of a multi-unit building.
In order to provide a comparison with deeper detail, this
system is considered for performance assessment. Exergy
and environmental impact analysis of the system is conducted to gain a better insight into this study.
In the system, solar energy is used to produce electricity,
domestic heating water, cooling and hydrogen. There is a
wind turbine to supply the system and measure the eects.
The system utilizes a Rankine cycle, absorption chiller and
electrolyzer. This system has 43% maximum energy eciency and 65% maximum exergy eciency. Maximum turbine output is 48 kW, while cooling eect is 28 kW and
heating eect is 298.5 kW. 1614 tons per year of CO2 is
saved by the system. It is capable of supplying at least 49
suites.
The systems are compared with the outputs of a trigeneration system developed in the lab. Due to the unprocessed data and dierent outputs, the results do not

match 100%. However, it is seen that by using a multigeneration system, eciency is higher than the combined eciency of a system with separate units.
With the assistance of this study, the same or similar system can be built and used to achieve higher eciencies by
using renewable sources to serve multi-unit buildings or
districts. The future energy solutions have to contain
renewable sources as an alternative to fossil fuels. Other
similar hybrid energy generating systems can be analyzed
with a fair degree of accuracy using the proposed technique. The results of this thesis should be used to design
new multigeneration systems or develop these systems to
achieve better results in the future.
References
Ahmadi, P., 2013. Modeling, analysis and optimization of integrated
energy systems for multigeneration purposes. Ph.D. Thesis. Faculty of
Engineering and Applied Science, UOIT, Oshawa, Ontario.
Ahmadi, P., Dincer, P., Rosen, M.A., 2014. Multi-objective optimization
of a novel solar-based multigeneration energy system. Sol. Energy 108,
576591.
Binkley, C., 2012. Energy consumption tends of multi-unit residential
buildings in the city of Toronto. M.Sc. Thesis, Department of Civil
Engineering, University of Toronto, Toronto, Ontario
Caliskan, H., Dincer, I., Hepbasli, A., 2013. Exergoeconomic and
environmental impact analyses of a renewable energy based hydrogen
production system. Int. J. Hydrogen Energy, 18.
Celik, A.N., 2003. Techno-economic analysis of autonomous PVwind
hybrid energy systems using dierent sizing methods. Energy Convers.
Manage. 44, 19511968.
Cengel, Y.A., Boles, M.A., Kanoglu, M., 2011. Thermodynamics: an
engineering approach. McGraw-Hill, New York.
Desmukh, M.K., Desmukh, S.S., 2008. Modeling of hybrid renewable
energy systems. Renew. Sustain. Energy Rev. 12, 235249.
Dincer, I., Rosen, M.A., 2012. Exergy Energy Environment and Sustainable Development. Elsevier Science.
Dincer, I., Colpan, C.O., Kadioglu, F., 2013. Causes Impacts and
Solutions to Global Warming. Springer, New York.
Due, J.A., Beckman, W.A., 2006. Solar Engineering of Thermal
Processes. John Wiley & Sons Inc..
El-Shatter, T.F., Eskandar, M.N., El-Hagry, M.T., 2002. Hybrid PV/fuel
cell system design and simulation. Renew Energy 27, 479485.
Kaabeche, A., Belhamel, M., Ibtiouen, R., 2011. Techno-economic
valuation and optimization of integrated photovoltaic/wind energy
conversion system. Sol. Energy 85, 24072420.
Notton, G., Diaf, S., Stoyanov, L., 2011. Hybrid photovoltaic/wind
energy systems for remote locations. Energy Procedia 6, 666677.
Pavlas, M., Stehlk, P., Oral, J., Sikula, J., 2006. Integrating renewable
sources of energy into an existing combined heat and power system.
Energy 31 (13), 24992511.
Ratlamwala, T.A.H., Dincer, I., Aydin, M., 2012. Energy and exergy
analyses and optimization study of an integrated solar heliostat eld
system for hydrogen production. Int. J. Hydrogen Energy 37, 18704
18712.
Ozturk, M., Dincer, I., 2013. Thermodynamic analysis of a solar-based
multi-generation system with hydrogen production. Appl. Therm. Eng.
51, 12351244.
Syed, A.M., Fung, A.S., Ugursal, V.I., Taherian, H., 2009. Analysis of
PV/wind potential in the Canadian residential sector through
high-resolution building energy simulation. Int. J. Energy Res. 33,
342357.
Tarique, M.A., 2014, Design, analysis and experimental investigation of a
new scroll expander based tri-generation system. Ph.D. Thesis, Faculty
of Engineering and Applied Science, UOIT, Oshawa, Ontario.

S. Ozlu, I. Dincer / Solar Energy 122 (2015) 12791295

Tina, G., Gagliano, S., 2011. Probabilistic analysis of weather data for a
hybrid solar/wind energy system. Int. J. Energy Res. 35, 221232.
Toronto Hydro, Heating Degree days. <http://www.torontohydro.com/
sites/corporate/Newsroom/Documents/DegreeDays_1.pdf>, retrieved
on April 7th, 2015.
United States Environmental Protection Agency Clean Energy web page,
retrieved on February 01, 2015, <http://www.epa.gov/cleanenergy/
energy-and-you/aect/air-emissions.html>.

1295

Wang, J., Dai, Y., Gao, L., Ma, S., 2009. A new combined cooling heating
and power system driven by solar energy. Renewable Energy 34, 2780
2788.
Xu, C., Wang, Z., Li, X., Sun, F., 2011. Energy and exergy analysis of
solar power tower plants. Appl. Therm. Eng. 31, 390413.