Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190

Contents lists available at SciVerse ScienceDirect

Renewable and Sustainable Energy Reviews

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

Determination of the optimum hybrid renewable power generating systems

for Kavakli campus of Kirklareli University, Turkey
Bahtiyar Dursun n
Department of Electrical & Electronic Engineering, Faculty of Technology, Kirklareli University, Kavakli Kampusu 39040, Kirklareli, Turkey

a r t i c l e i n f o abstract

Article history: This study is to search for possibilities of supplying the load demand of Kavakli campus of Kirklareli
Received 8 February 2012 University with solar energy and the fuel cell power generating system (electrolyzer/hydrogen tank/
Received in revised form fuel cell) by using the HOMER software due to the fact that hybrid power systems with renewables can
2 July 2012
significantly reduce emissions which are caused by utilization of non-renewable power sources. In this
Accepted 15 July 2012
Available online 30 August 2012
study, various hybrid systems will be examined and compared among themselves considering cost of
energy (COE), renewable fraction, total net present cost (NPC) and hydrogen production. Additionally,
Keywords: this study will seek whether a fuel cell can be integrated into the hybrid systems. According to the
PV array study results, the grid connected systems appear cost-effective as expected. Although the grid-
Hybrid power systems
connected photovoltaic (PV) hybrid system has the lowest COE and NPC, the grid-connected PV/fuel
Fuel cell
cell hybrid system with COE, 0.294$/kWh has a slightly higher cost than the optimum one. It is strongly
Renewable energy
Optimization believed that this system may be chosen because it is a cleaner system and its emissions are fairly low.
& 2012 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6183
2. Description of Kavakli campus of Kirklareli University . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
2.1. Location and population of the campus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
2.2. Load profiles of the campus. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
2.3. Solar energy potential of the campus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
3. Three major parameters in the economic analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
3.1. Calculation of the annual real interest rate for Turkey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6184
3.2. Levelized cost of energy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6185
3.3. Net present cost (NPC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6185
4. Hybrid power generating system components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6185
4.1. PV–diesel power generating system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6185
4.2. The grid connected PV power generating system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6186
4.3. The stand-alone PV–hydrogen power generating system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6186
4.4. The grid connected PV–hydrogen power generating system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6186
5. Results and discussions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6187
5.1. The stand-alone PV–diesel hybrid system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6187
5.2. The grid connected PV hybrid system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6188
5.3. The standalone PV–fuel cell hybrid system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6188
5.4. The grid-connected PV–fuel cell hybrid system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6189
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6189
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6190
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6190

1. Introduction
n
Tel.: þ90 288 615 3303; fax: þ90 288 615 3750.
E-mail addresses: [email protected], Energy is a fundamental constituent of economical growth
[email protected] and socio-economical development. Energy needs of countries

1364-0321/$ - see front matter & 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.rser.2012.07.017
6184 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190

increase with the rising population. Turkey, having population 2. Description of Kavakli campus of Kirklareli University
more than 70 million, has 1.31% increase each year in population
which is foreseen to reach 83.4 million in 2022 [1]. Furthermore, 2.1. Location and population of the campus
Turkey has many rich energy sources, especially renewable
energy sources such as geothermal, wind, hydraulic, wave, bio- Kavakli campus of Kirklareli University is located in the
mass and solar energy. However, it cannot utilize them very Northern Marmara region. Its coordinates are 41,650 latitude
efficiently. Solar energy is one of the most popular renewable and 27,166 longitudes. Area of the campus is 50,000 m2. Besides,
energy sources and is non-depletable, site-dependent, non-pol- since it is a newly formed university and it was established in
luting, and potential source of alternative energy as well [2,3]. 2007, it contains only two institutions, six faculties, seven
However, it has an unstable and variable characteristic. Due to research and development centers and its student population is
intermittent nature of the solar energy, only solar power cannot about 6000 [16].
meet the continuous energy demand. Therefore, solar power is
used together with hybrid systems which are combined with one 2.2. Load profiles of the campus
or more of the renewable energy resources like solar and wind
energy. Generally, there are many studies about the hybrid wind/ Energy requirement of the campus is currently supplied by
photovoltaic (PV) system, PV/diesel or wind/diesel. Most of them electricity grid. Load data used in this study was obtained from
deal with cost and economic optimization, system design and TEIAS. According to the load data, the average daily energy
optimization in hybrid energy systems [4]. Some of them are demand of the campus is about 485 kWh [17].
shortly described as following. Dufo-Lopez and Bernal-Austin HOMER simulates the operation of a system by making energy
optimized a PV–diesel system using two different ways. One of balance calculations for each hour in a year [18]. The hourly load
the ways is applied to a PV–diesel system using the HOGA profiles are not available for a whole year, so HOMER is used to
software program. The other way includes a classical design synthesize the load profiles (with randomness) by entering the
method which is based on the available energy under worst-case values for a typical day. Either day-to-day randomness or time-
conditions [5]. Similarly, Muselli et al. developed a methodology step-to-time-step randomness is taken as 5% in the study.
for calculating the correct size of a PV-hybrid system and for According to the load data, the minimum load demand occurs
optimizing its management [6]. Moreover, Ashari and Nayar between 00:00 and 06:00 . Load profile of the Kavakli campus is
presented dispatch strategies for the operation of a solar PV– shown in Fig. 1.
diesel–battery hybrid power system using ‘‘set points’’. They In summer, load demand is higher because air conditioning
determined the optimum values of set points for starting and systems in offices are used more. For summer, maximum value of
stopping of the diesel generator to minimize the overall system the load demand is 36 kWh which occurs between 10:00 and 11:00.
costs [7]. Similarly, Nfah et al. modeled solar–diesel–battery
hybrid power systems for the electrification of typical rural 2.3. Solar energy potential of the campus
households and schools [8]. Additionally, apart from the studies
on renewable energy systems excluding fuel cell and electrolyzer Solar radiation data of the region where Kavakli campus is
components, there are also many studies on the PV hybrid system located was obtained from Turkish State Meteorological Service
combined with fuel cell power generating unit as well. One of (TSMS) in the year 2010 [19]. Monthly average solar energy
them was carried out by Perez [9] and Lehman et al. [10]. Since density values are shown in Fig. 2. Annual average solar energy
then, several systems have been developed and evaluated in density value is calculated as 4.864 kWh m " 2 d " 1. HOMER
recent years [11,12]. For Zoulias et al. [13] the use of FCs in synthesizes solar radiation values for each hour of a year by using
stand-alone power systems may prove to be a large market niche. the Graham algorithm. This algorithm produces realistic hourly
Thus, FCs may compete, in the mid-term, with currently available data, and it is easy to use because it requires only the latitudes
commercial technologies, such as diesel generators and battery and the monthly averages [18].
banks. Hollmuller and Joubert studied the performance of a
privately owned PV–hydrogen production and storage installation
in a one-family house at Zollbruck i.e. in Switzerland [11]. More- 3. Three major parameters in the economic analysis
over, Ghosh et al. [12] introduced the PHOEBUS demonstration
plant that supplied energy to part of the Central Library in 3.1. Calculation of the annual real interest rate for Turkey
Forschungszentrum Julich, Germany. Furthermore, Hwang et al.
developed a mathematical model for a stand-alone renewable The annual real interest rate is one of the HOMER’s inputs.
power system, referred as the ‘‘PV–fuel cell (PVFC) hybrid system’’, The annual real interest rate is related to the nominal interest rate
which maximizes the use of a renewable energy source [14].
Finally, Mohd presented a renewable energy/hydrogen based
power system model to provide electricity to a coastal residential
area in east coast area (Kuala Terengganu) of Malaysia [15].
This study is to search for the possibilities of supplying the
load demand of Kavakli campus of Kirklareli University which is
currently supplied by the electricity grid, with the solar energy
and hydrogen fuel cell power generating system (electrolyzer/
hydrogen tank/fuel cell) since renewable energy based hybrid
power systems can significantly reduce the amount of emission
gases from the non-renewable power system (conventional
power systems). In this study, the National Renewable Energy
Laboratory’s (NREL) optimization tool ‘‘HOMER’’ will be used in
identifying probable hybrid configurations and their applicability
among various hybrid systems, considering some parameters,
COE, renewable penetration rate, NPC, and hydrogen production. Fig. 1. Load profile of Kavakli Campus of Kirklareli University.
 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190 6185

Fig. 2. Monthly average solar energy density values of Kırklareli, Turkey.

by the following equation:

i0 "f
i¼ ð1Þ
1 þf
where i is the real interest rate, i0 is the nominal interest rate (the
rate at which you could get a loan), and f is the annual
inflation rate.
For Turkey, i0 ¼9% (interest rate was taken from the Turkish
Central Bank in August 13, 2011) and f¼6.4% (annual inflation Fig. 3. All components of PV–diesel hybrid power generating systems.
rate was taken from the Turkish Central Bank at the end of 2010)
are used. With these values, using Eq. (2), real interest rate is
found to be 2.45% as calculated below [20,21]: where Cann,tot is the total annualized cost ($/yr), CRF is the capital
0:09"0:064 recovery factor, i is the real interest rate (%), and Rproj is the
i¼ ¼ 0:0245 ð2Þ project lifetime (yr). The capital recovery factor is a ratio used to
1 þ 0:064
calculate the present value of an annuity (a series of equal annual
In HOMER simulations, 2.45% is used for real interest rate. cash flows). The equation for the capital recovery factor is
ið1 þ iÞN
3.2. Levelized cost of energy CRFði,NÞ ¼ ð5Þ
ð1 þ iÞN "1
HOMER defines the levelized COE as the average cost/kWh of where i is real interest rate (%) and N is number of years.
useful electrical energy produced by the system. The equation for
the COE is as follows:
4. Hybrid power generating system components
C ann,tot
COE ¼ ð3Þ
Eprim,AC þEprim,DC þ Egrid,sales
Various stand-alone and grid connected hybrid systems
where Cann,tot is total annualized cost ($/yr), Eprim,AC is AC primary discussed for Kavakli campus mainly comprise PV panels, fuel
load served (kWh yr " 1), Eprim,DC is DC primary load served cell, hydrogen tank, electrolyzer, battery, converter, diesel gen-
(kWh yr " 1), and Egrid,sales is total grid sales (kWh yr " 1). The total erator and grid. Main components in each system discussed in
annualized cost is the sum of the annualized costs of each system this study are shortly summarized below.
component, plus the other annualized cost. It is an important
value because HOMER uses it to calculate both the levelized COE 4.1. PV–diesel power generating system
and total NPC [21,22].
The schematic diagram of PV–diesel power system compo-
3.3. Net present cost (NPC) nents are presented in Fig. 3.
The hybrid power generating system consists of a diesel
The present value of the cost of installing and operating the generator, PV arrays, batteries and power converters. Some inputs
system over lifetime of the project is also referred as lifecycle cost. for each component of the hybrid system such as cost, number of
Project lifetime in this study is considered as 20 yr. The total NPC units, capacity, operating hours and other specifications are
is HOMER’s main economic output. All systems are ranked needed to run the simulation using the HOMER software. The
according to NPC, and all other economic outputs are calculated details of the system components were obtained from manufac-
for the purpose of finding the NPC. The NPC is calculated turers of the equipments and previous studies [23–26]. The
according to Eq. (4) [21,22]: descriptions of these components are given below.
Diesel generator: The cost of a commercially available diesel
C ann,tot generator may vary from $250 to $500 kW " 1 [23]. For larger units
C NPC ¼ ð4Þ
CRFði,Rproj Þ per kW cost is lower and for smaller units cost is more. A 50 kW
6186 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190

diesel generator is used in this power system as the peak power

demand is less than 50 kW. Regarding the study carried by
Dalton, capital, replacement and operational costs are taken as
$450, $400 and $0.150 h " 1, respectively. Meanwhile, the lifetime
is 15,000 h. In this study no diesel generator (0 kW) or a 50 kW
unit was used for simulation by HOMER.
PV-array: The installation cost of PV arrays may vary from
$6.00–$10.00 W " 1. A 1 kW solar energy system installation and
replacement costs are taken as $7000 and $6000, respectively [23].
Various sizes of PV array are considered, ranging from 0 to 150 kW
in this study. The lifetime of the PV array is taken as 20 yr and
there is no tracking system in the PV arrays.
Batteries: Batteries are supposed to have a big share in the total
cost of small-scale stand-alone power systems. A Surrette-6CS25P
model battery with 6 V, 1156 Ah and 9645 kWh is chosen in this
simulation [24]. The estimated lifetime of the battery is 5 yr and Fig. 5. Configuration of the stand-alone PV–hydrogen power generating system.
the cost of one battery is $1250 with a replacement cost of $1100
while the operation and maintenance cost is $0.02 yr " 1. Mean-
while, the number of units for the battery stacks is in the range of
0–200 units. array are the same as those in Section 4.2. The considered sizes of
Power converter: A power electronic converter maintains flow the PV array are in the range of 0–200 kW.
of energy between the ac and dc components. For a 1 kW system Electrolyzer: Currently, the production cost of electrolyzers is
the installation and replacement costs are taken as $800 and between $1500 and $3000 kW " 1. With improvements in polymer
$750, respectively. Four different sizes of converters (0, 15, 30 and technology, control systems and power electronics it is expected
45 kW) are considered for the simulation. Meanwhile, lifetime of that costs would reduce much in 10 yr [23]. In this analysis,
the power converter is considered to be 15 yr with an efficiency of various sizes of electrolyzer (0–50 kW) are considered. A 1 kW
90% [15,25]. system has a capital cost of $2000, a replacement cost of $1500
and an operational and maintenance cost of $20. Meanwhile,
lifetime is considered as 25 yr with efficiency 75% [15].
4.2. The grid connected PV power generating system Power converter: Power electronic converter in this hybrid
system has similar properties with one in the previous system.
The grid connected PV power system is powered by the grid For a 1 kW system, the installation and replacement costs are
and it does not include diesel generator unlike the stand-alone taken as $800 and $750, respectively. Three different sizes of the
PV–diesel hybrid power system. All components of the hybrid converter with power capacity between 10 and 60 kW are taken
system are the same as those in the standalone PV–diesel hybrid in the model.
system mentioned in Section 4. All components of the hybrid Fuel cell system: The cost of fuel cell varies greatly depending
system are indicated in Fig. 4. on the type of technology, reformer, auxiliary equipments and
power converters. At present, a fuel cell cost varies from $3000 to
$6000 kW " 1 [23]. Here, the capital, replacement and operational
4.3. The stand-alone PV–hydrogen power generating system costs are taken as $3000, $2500 and $0.020 h " 1 for a 1 kW
system, respectively. Four different sizes of fuel cells are taken
The conventional hybrid power system can be easily upgraded in the simulation process: 0 (no fuel cell used), 5, 10 and 20 kW.
to the stand-alone PV–hydrogen hybrid power system that is Meanwhile, lifetime and efficiency of the fuel cell are taken as
schematically designed as in Fig. 5. 40,000 h and 50%, respectively.
All the meteorological data used here is the same as the Hydrogen tank: Cost of a tank with 1 kg hydrogen capacity is
previous simulation. The equipments of this hybrid system are assumed to be $1300. The replacement and operational costs are
PV array, battery, fuel cell, hydrogen tank, electrolyzer, and power taken as $1200 and $15 yr " 1, respectively. Five different sizes
electronic converter. In this hybrid energy system, the type of (0, 20, 40, 60 and 80 kg) are included to widen the search space
battery has the same properties with one used in the previous for a cost effective configuration, and the lifetime is also con-
system, which is Surrette 6CS25P. In order to determine optimum sidered as 25 yr.
combination of equipment dimensions, different sizes of the
battery are selected. Stand-alone PV–hydrogen system compo- 4.4. The grid connected PV–hydrogen power generating system
nents are described more detail below.
PV-array: For this hybrid system, the PV capital, replacement, In this grid connected hybrid system, the grid supplies no
and operation and maintenance costs, as well as lifetime of the PV power to the electrolyzer device in order to produce hydrogen in
the case of the deficiency of power supplied by the PV system.
Only excess electricity, which was produced in PV panels but was
no longer used in meeting the demand, serves to activate the
electrolyzer. The schematic appearance of the hybrid system is
indicated in Fig. 6. Such hybrid system includes the similar
equipments as those in the stand-alone PV–hydrogen hybrid
system described in Section 4.3.
Unlike the standalone PV–hydrogen hybrid system, it is a
grid connected system. The single rate that refers to the fix
power price, sellback rate and demand rate is set for the case
Fig. 4. Configuration of grid gonnected PV power generating systems. of residential consumers. The fix power price is 0.1$ kWh " 1
 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190 6187

while the sellback rate and demand rate are 0.05$ kWh " 1 and Kirklareli. The optimum (best) configurations obtained for these
0.00$ kWh " 1 month " 1, respectively. The grid system has two hybrid systems are presented in detail in Table 1.
working conditions. When the renewable energy system produces Furthermore, all outputs about the economic and environ-
more power than the demand, the excess power is fed back into the mental parameters of these hybrid systems that was simulated by
grid. Otherwise, when the system does not produce enough power, means of the HOMER software, are detailed in Table 2.
then the required power can be provided from the grid [15]. In this section, considerable outputs concerned with each
optimum stand-alone and grid connected hybrid systems, which
were obtained in the study employing the HOMER software, will
be presented item by item in the following sub-sections.
5. Results and discussions

In this study, four stand-alone and grid connected hybrid

5.1. The stand-alone PV–diesel hybrid system
systems are considered and analyzed using the HOMER software
to determine the optimum hybrid systems for the campus in
& The HOMER software completed the simulation within only 9 s.
& The least COE of this system is $0.817 kWh " 1 and results from
the combination of 120 kW PV, 50 kW diesel generator,
150 kW battery and 45 kW converter.
& The renewable energy fraction of this hybrid system is 0.736.
& Diesel fuel consumption in this hybrid system is only 21,025 lt.
& Simulation results show that the consumption of diesel fuel
and amounts of emission gases (CO2, NOx, CO and SOx) per year
are reduced about 75% by the introduction of 120 kW PV
panels into the system when compared with the stand-alone
diesel system using only diesel generator as power supplier.
& The distribution of annualized cost for each component of the
stand-alone hybrid PV energy system is presented in Table 3.
The capital cost, total NPC and COE for this optimal hybrid
system are $1,068,000, $1,849,654 and $0.817 kWh " 1, res-
pectively. The most expensive cost, $954,020 comes from the
PV panels. It is followed by diesel generator, battery and
Fig. 6. Configuration of the grid connected PV–hydrogen power generating system. converter costs.

Table 1
Optimum configurations for the considered hybrid systems.

PV Gen FC Battery Converter Electrolyzer H2 tank Grid Initial capital Operating cost Total COE ($/ Ren. FC
(kW) (kW) (kW) (kW) (kW) (kg) (kW) ($) ($/yr) NPC kWh) frac. (h)

120 50 – 150 45 – – – 1,068,000 61,146 1,849,654 0.817 0.74 –

40 – – – 25 – – 1000 300,000 29,075 838,768 0.256 0.31 –
190 – 5 800 40 5 40 – 2,439,000 54,362 3,446,346 1.051 1.00 24
50 – 10 – 30 10 20 100 450,000 27,759 964,381 0.294 0.38 16

Table 2
All outputs associated with some economical and environmental parameters of all the systems considered in the study.

Systems considered in COE Total cost Renewable Capital Fuel Emissions (kg yr " 1)
the study ($/ ($) fraction cost ($)
kWh) Diesel H2 production CO2 CO Unburned Particulate SO2 NOx
(lt yr " 1) (kg yr " 1) hydrocarbons matter

Diesel 1.023 2,315,640 0 4500 82,719 – 217,826 538 59.6 40.05 437 4798
Standalone PV/diesel 0.817 1,849,654 0.74 1,068,000 21,025 – 55,365 137 151 10.3 111 1219
Standalone PV/fuel 1.051 3,446,346 1 2,439,000 – 75.6 0 0.257 0 0 0 2.3
cell
Grid connected PV 0.256 711,060 0.31 82,000 – – 103,815 0 0 0 450 220
Grid connected 0.294 964,381 0.38 450,000 – 52.6 72,857 0 0 0 316 157
PV/fuel cell

Table 3
Annualized cost for main components of the hybrid PV/diesel system.

Component Capital ($) Replacement ($) O&M ($) Fuel ($) Salvage ($) Total ($)

PV 840,000 224,500 15,340 0 " 125,819 954,020

Diesel generator 4500 3376 27,689 537,537 " 553 572,549
Surette 6CS25P 187,500 122,752 38 0 " 35,241 275,049
Converter 36,000 14,083 575 0 " 2621 48,037
System 1,068,000 364,710 43,642 537,537 " 164,235 1,849,654
6188 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190

& From simulation results, the PV–diesel hybrid system has a cost which is equal to $471,644. It is followed by PV and
total annual electrical energy production of 231,320 kWh yr " 1. converter costs.
The biggest contributor is PV panel which produces 74% of the & According to the simulation results, the grid connected PV
total electrical energy production (170,234 kWh yr " 1). hybrid system has a total electrical energy production of
184,008 kWh yr " 1. 69% of it is supplied by grid purchases
and the rest is served by PV panels. Furthermore, operational
5.2. The grid connected PV hybrid system and maintenance cost of the hybrid system makes up to
almost 60% of the total cost.
& The HOMER software completed the simulation within only
16 s.
& The least COE of this system is $0.217 kWh " 1 and results from
the combination of 10 kW PV panel, 15 kW power converter, 5.3. The standalone PV–fuel cell hybrid system
1000 kW grid. But, its renewable fraction is just 0.08.
& When considering that renewable fraction is greater than 25%, & The HOMER software completed the simulation within 2 min
the hybrid power system including a 40 kW PV panel, 25 kW and 6 s.
power converter, and 1000 kW grid has the least COE of & According to the optimization results, the optimum configura-
$0.256 kWh " 1 besides that, its renewable fraction becomes 31%. tion for this hybrid system contains a 190 kW PV panel, 5 kW
& When regarding environmental effects of these two hybrid fuel cell, 40 kW power converter, 800 kW battery bank, 5 kW
systems with least COE, with an increase in the value of electrolyzer and 40 kg hydrogen tank and has the least COE,
renewable fraction from 0.08 to 0.31, the emission rates of $1.051 kWh " 1.
CO2, NOx, CO and SOx can be reduced about 20%. & Renewable fraction of the optimum hybrid system is 1 because
& The distribution of annualized cost for each component of the there is neither electricity production from diesel generator
PV–grid hybrid power system with the renewable fraction of nor electricity purchasing from the grid. Regarding environ-
0.31 is presented in Table 4. mental effects of the hybrid system with the least COE,
The capital cost, total NPC and COE for the optimal hybrid emission rates of CO2, NOx, CO and SOx are nearly zero.
system are $300,000, $838,768 and $0.256 kWh " 1, res- & According to the simulation outputs about how often fuel cell
pectively. For this hybrid system, the grid has the most expensive generator is operated in the system, the fuel cell power
generator operates for only 24 h during the 1 yr period.
It means that there would be no more contribution from the
Table 4 fuel cell power generator to the total electricity production of
Annualized cost for main components of the grid connected PV hybrid system. the hybrid system.
Component Capital Replacement O&M Fuel Salvage Total
& The distribution of annualized cost for each component of the
($) ($) ($) ($) ($) ($) stand-alone PV–fuel cell hybrid system is presented in Table 5.
The capital cost, total NPC and COE for the optimal hybrid
PV 280,00 147,901 7412 0 " 98,282 337,031 system are $2,439,000, $3,446,346 and $1.051 kWh " 1, res-
Grid 0 0 471,644 0 0 471,644 pectively. The most expensive cost, $1,709,984 draws from
Converter 20,000 13,041 463 0 " 3413 30,092
System 300,000 160,943 479,519 0 " 101,694 838,768
the battery which makes up about half of the total cost.
The battery cost is followed by the PV cost with $1,600,900.

Table 5
Annualized cost for main components of the standalone PV/fuel cell hybrid system.

Component Capital ($) Replacement ($) O&M ($) Fuel ($) Salvage ($) Total ($)

PV 1,330,000 702,531 35,207 0 " 466,839 1,600,900

Fuel cell 15,000 0 44 0 " 6552 8492
Surette 6CS25P 1,000,000 1,150,433 0 0 " 440,448 1,709,984
Converter 32,000 20,866 741 0 " 5460 48,147
Electrolyzer 10,000 5217 1853 0 " 1365 15,705
Hydrogen tank 52,000 0 11,118 0 0 63,118
System 2,439,000 1,879,046 48,964 0 " 920,664 3,446,346

Fig. 7. The amount of yearly hydrogen production.

 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190 6189

As seen from Table 3, PV cost is followed by hydrogen tank, that there would be no more contribution from the fuel cell
converter, electrolyzer and fuel cell with $63,118, $48,147, power generator to the total electricity production of the hybrid
$15,705 and $8492, respectively. system.
& According to the simulation results, the PV–fuel cell hybrid & The distribution of annualized cost for each component of the
system generates total electrical energy production, 269,538 stand-alone PV–fuel cell hybrid system is presented in Table 6.
kWh yr " 1. Nearly all of this value is generated by the PV & The capital cost, total NPC and COE for the optimal hybrid
panel. system are $450,000, $964,381 and $0.294 kWh " 1, respec-
& The monthly hydrogen production of the 5 kW electrolyzer can tively. The most expensive cost, $427,239 draws from the grid
be seen from Fig. 7. According to Fig. 7, hydrogen is mostly which makes up about half of the total cost. The grid
produced in the months of March, October and September purchasing cost is followed by the PV cost with $421,289.
with 0.58 kg d " 1, 0.48 kg d " 1 and 0.35 kg d " 1, respectively. & The PV cost is followed by the hydrogen tank, converter,
Meanwhile, the amount of yearly hydrogen production is electrolyzer and fuel cell costs.
75.6 kg. & According to the simulation results, the grid connected PV–
fuel cell hybrid system generates total electrical energy with
an amount of 186,370 kWh yr " 1, 70,931 kWh yr " 1 is gener-
ated by PV panel and the rest is served by the grid.
5.4. The grid-connected PV–fuel cell hybrid system
& The monthly hydrogen production of the 5 kW electrolyzer can
be seen from Fig. 8. According to Fig. 8, hydrogen is mostly
& The HOMER software completed the simulation within 1.5 min.
produced in the months of March, February, September and
& According to the optimization results, the best configuration of
December with 0.265 kg d " 1, 0.195 kg d " 1, 0.18 kg d " 1 and
this hybrid system contains a 50 kW PV panel, 10 kW fuel cell,
0.175 kg d " 1 respectively. Additionally, the amount of yearly
30 kW power converter, 10 kW electrolyzer and 20 kg hydro-
hydrogen production is 52.6 kg.
gen tank and has the least COE, $0.294 kWh " 1.
& Regarding the hybrid systems whose renewable fraction is at
least 0.25, the renewable fraction value for the optimum
hybrid system becomes 0.38. 6. Conclusions
& When regarding environmental effects of this optimum hybrid
system, emission rates of some pollutants such as CO2, SOx, Some major outputs which were achieved in the study are
NOx, CO, unburned hydrocarbons and particulate matter given as follows:
are 72,857 kg yr " 1, 316 kg yr " 1, 157 kg yr " 1, 0.338 kg yr " 1,
0.0375 kg yr " 1 and 0.0255 kg yr " 1, respectively. & Fuel consumption and emission rates in the stand-alone
& According to the simulation outputs about how often fuel cell PV–diesel hybrid system is 74% lower than the diesel system
generator is operated in the system, the fuel cell power excluding a power supplier fueled by solar energy. Similarly, of
generator operates for only 16 h during the 1 yr period. It means the stand-alone PV–diesel hybrid system is fairly low com-
pared to the diesel system.
& Regarding the environmental effects of the optimum stand-
alone PV–fuel cell hybrid system, the emission rates are very
Table 6 low and nearly zero because there is neither electricity
Annualized cost for main components of the grid connected PV/fuel cell hybrid production from diesel generator nor electricity purchasing
system.
from the grid in this hybrid system.
Component Capital Replacement O&M Fuel Salvage Total & Fuel cost in the diesel system occupies nearly 100% of the total
($) ($) ($) ($) ($) ($) cost. Only $4500 is the capital cost while total cost and fuel
cost are $2,315,640 and 2,311,140 , respectively.
PV 350,000 184,877 9265 0 " 122,852 421,289
Fuel cell 30,000 0 59 0 " 13,286 16,773
& The stand-alone hybrid systems have the highest COE because
Grid 0 0 427,239 0 0 427,239 they have expensive energy storage equipments (Battery bank
Converter 24,000 15,650 556 0 " 4095 36,110 and power converters) in contrast to the grid connected systems.
Electrolyzer 20,000 10,433 3706 0 " 2730 31,409 The battery and converter costs of the stand-alone PV/diesel
Hydrogen 26,000 0 5559 0 0 31,559
hybrid system cover nearly 20% of the NPC. Meanwhile, since
tank
System 450,000 210,959 446,385 0 " 142,964 964,381
there is no battery used in the grid connected PV/fuel cell hybrid
system, there is no cost inquired due to batteries. It concludes

Fig. 8. The monthly hydrogen production of the grid connected PV–fuel cell hybrid system.
6190 B. Dursun / Renewable and Sustainable Energy Reviews 16 (2012) 6183–6190

that there is only converter cost in the system, covering only [9] Perez R.. The Schatz PV hydrogen project, Home power#221991, /http://
about 3% of the NPC of the system. www.xlabs.pl/schematy_pliki/magazynh2.pdfS.
[10] Lehman PA, Chamberlin CE, Pauletto G, Rocheleau M. Operating experience
& In general the grid connected systems provide cost-effective with a photovoltaic-hydrogen. Energy system. In: Block DL, Veziroglu TN,
solutions. Editors. Proceedings of the 10th world hydrogen energy conference, Cocoa
& The grid-connected PV hybrid system has the lowest COE Beach, FL; 1994.
[11] Hollmuller P, Joubert J, Lachal B, Yvon K. Evaluation of a 5 kWp photovoltaic
($0.256 kWh " 1) and NPC ($82,000). hydrogen production and storage installation for a residential home in
& On the other hand, the grid-connected PV/fuel cell hybrid Switzerland. International Journal of Hydrogen Energy 2000;25:97–109.
system with COE, $0.294 kWh " 1 that has a slightly higher cost [12] Ghosh PC, Emonts B, Janben H, Mergel J, Stolten D. Ten years of operational
experience with a hydrogen-based renewable energy supply system. Solar
than the grid-connected PV hybrid system. This system may be
Energy 2003;75:469–78.
chosen because its renewable fraction is higher and its gas [13] Zoulias EI, Glockner R, Lymberopoulos N, Tsoutsos T, Vosseler I, Gavalda O,
emission rates are fairly low. Thus, it is cleaner and envir- et al. Integration of hydrogen energy technologies in stand-alone power
onmentally friendlier when compared to the grid-connected systems: analysis of the current potential for applications. Renewable
Suststainable Energy Reviews 2006;10:432–62.
PV hybrid system. [14] Hwang JJ, Wang DY, Shih NC, Lai DY, Chen CK. Development of fuel-
cell-powered electric bicycle. Journal of Power Sources 2004;133:223–8.
[15] Mohd ZI, Roziah Z, Marzuki I, Muzathik AM. Pre-feasibility study of hybrid
Acknowledgments hydrogen based energy systems for coastal residential applications. Energy
Research Journal 2010;1:12–21.
[16] KLU. Information Guide for International Student, Kirklareli University; 2011
I am fairly thankful to the TEIAS and Turkish State Meteor- /http://www.kirklareli.edu.tr/katalog-en/S.
ological Service for providing the load data and solar data of the [17] TEIAS. Turkish Electricity Transmission Company, Load data of the Kavakli
Campus of Kirklareli University; 2010. /http://www.teias.gov.tr/eng/S.
Kavakli campus of Kirklareli University, respectively. Also, I would
[18] HOMER Software Version 2.67, National Renewable Energy Laboratory
like to thank Kirklareli University for supporting this study. (NREL), USA, /http://www.nrel.gov/HomerS.
[19] TSMS. Turkish State Meteorological Service. The solar radiation data of
Kavakli Campus of Kirklareli University; 2011 /http://www.dmi.gov.tr/
References
en-US/forecast-cities.aspxS.
[20] Central Bank of the Republic of Turkey; 2011 /http://www.tcmb.gov.tr/yeni/
[1] SIS. Statistics, Statistical Yearbook of Turkey. Prime Ministry Republic of eng/S.
Turkey; 2003. [21] Demiroren A, Yilmaz U. Analysis of change in electric energy cost with using
[2] Boyle G. Renewable energy: power for a sustainable future. First ed.. Oxford renewable energy sources in Gokceada, Turkey: an island example. Renew-
University Press; 1998 pp. 1–40. able Sustainable Energy Reviews 2010;14:323–33.
[3] Kaya D. Renewable energy policies in Turkey. Renewable and Sustainable [22] HOMER, Help Files, National Renewable Energy Laboratory (NREL), USA,
Energy Reviews 2006;10:152–63. /http://www.nrel.gov/HomerS.
[4] Taskin S, Dursun B, Alboyaci B. Performance assessment of a combined solar [23] Dalton GJ, Lockington DA, Baldock TEJ. Feasibility analysis of renewable
and wind system. Arabian Journal for Science and Engineering 2009;34: energy supply options for a grid-connected large hotel. Renewable Energy
217–27. 2009;34:955–64.
[5] Dufo-Lopez R, Bernal-Agustin J. Design and control strategies of PV–diesel [24] Khan MJ, Iqbal MT. Pre-feasibility study of stand-alone hybrid energy
systems using genetic algorithms. Solar Energy 2005;79:33–46. systems for applications in Newfoundland. Renewable Energy 2005;30:
[6] Muselli M, Notton G, Poggi P, Louche A. PV-hybrid power systems sizing 835–54.
incorporating battery storage: an analysis via simulation calculations. [25] Zoulias EI, Lymberopoulos N. Technoeconomic analysis of the integration of
Renewable Energy 2000;20:1–7. hydrogen energy technologies in renewable energy-based stand alone power
[7] Ashari M, Nayar CV. An optimum dispatch strategy using set points for a systems. Renewable Energy 2007;32:680–96.
photovoltaic (PV)–diesel–battery hybrid power system. Solar Energy 1999;66: [26] Goh WC, Barsoum NN, Balancing cost, operation and performance in
1–9. integrated hydrogen hybrid energy system. In: AUPEC’06, The Australasian
[8] Nfah EM, Ngundam JM, Tchinda R. Modelling of solar/diesel/battery hybrid universities power engineering conference, Melbourne, Australia; December
power systems for far-north Cameroon. Renewable Energy 2007;32:832–44. 10–13, 2006.