FEASIBILITY ANALYSIS OF A GRID-CONNECTED PV SYSTEM FOR HOME APPLICATION

The 3rd Bali International Seminar on Science and Technology (BISSTECH)
October 15-17th, 2015
Grand Inna Kuta, Bali, Indonesia
B5.5-1
FEASIBILITY ANALYSIS OF A GRID-CONNECTED PV SYSTEM
FOR HOME APPLICATION
Wayan G. Santika1
, Putu Wijaya Sunu1
, I Made Arsawan1
1Bali State Polytechnic, Bukit Jimbaran Campus, Bali-Indonesia
E-Mail: wayan.santika@pnb.ac.id
ABSTRACT
The objective of the present study is to provide technical and economical analyses of a grid-connected PV system
for a small house located in Bukit Jimbaran, Bali. The peak load of the house during observation was 390 watt and the
daily electricity consumption is about 4.7 kWh. HOMER, a renewable energy system software developed by National
Renewable Energy Laboratory (NREL), was utilized for simulation and optimization. The house will be installed with a
grid-connected PV system which includes PV arrays, converters, and batteries (optional). The investment cost of the PV
arrays is 3000 USD/kW and their lifetime, derating factor, and ground reflectance are 20 years, 90%, and 20%,
respectively. The PV sizes to consider are 0.5, 1, 1.5, and 2 kilowatts. The grid applies a flat rate of about 0.1 USD/kWh.
The surplus energy of the PV system will be fed into the grid with a net metering system in which the meter run backward
when the excess energy is being fed into the grid. However, the sellback price is zero if energy sales exceed purchases. The
converter costs 1000 USD per kilowatt. The economic inputs required by HOMER are the annual real interest rate and the
lifetime of the project, which are 7% and 20 years, respectively. Results show that the proposed grid-connected PV system
is technically viable. However, the grid-only system is still the most cost effective choice based on the net present cost
(NPC) with the current price of 0.1 USD per kWh. The cheapest choice for the grid-connected PV system is when the PV
and converter sizes are both 0.5 kW. The NPC of the PV system is 3,823 USD and its related cost of electricity (COE) is
0.209 USD/kWh. The renewable fraction of the system is 38%. Sensitivity analysis were also conducted with some
scenarios such as reduction in PV prices, electricity price increases, and CO2 penalties.
Keywords: Renewable energy; grid-connected PV system; HOMER; feasibility analysis; home application.
INTRODUCTION
According to Antara (the news agency of
Indonesia), 26,800 new customers of PLN (the state
electricity company) Bali was on the waiting list due to
power shortage [1]. With the installed capacity and the
highest peak load of 850 MW and 781 MW, respectively,
PLN Bali cannot serve new installments. To solve the
problem, some measures have been taken, such as building
a new power plant at Celukan Bawang and reducing
demand by encouraging costumers to turn off appliances
during peak hours.
Another important measure is to encourage
costumers to apply renewable energy systems, which are
not so popular in Indonesia. Amid cheap electricity and
fuel prices, renewable energy systems are not the most
cost effective choices to power houses.
The objective of the present study is to provide
technical and economical analyses of a grid-connected PV
system for a small house located in Bukit Jimbaran, Bali.
The house is a typical two bedrooms house with a small
kitchen and a bathroom. Four people live in the house
which peak load during a 24-hour observation was 390
watt and the daily electricity consumption is about 4.7
kWh.
HOMER, a renewable energy system (RES)
software developed by National Renewable Energy
Laboratory (NREL), was utilized for simulation and
optimization. HOMER has been widely used by renewable
energy experts in different contexts, such as houses,
schools, hotels, and villages [2]. HOMER compares many
different RES based on their technical and economical
attributes [3].
LITERATURE REVIEW
There are different ways of reducing energy load
and protecting the environment, such as behavioral
changes [4,5], demand side management [6,7], and
renewable energy application.
Energy conservation through behavioral changes
can be done, for example, by encouraging hotel guests to
reuse linens and towels [8] or by changing light bulbs with
energy efficient ones. Demand side management is usually
applied by utilities when demand shifts from peak hours to
off peak hours are expected.
HOMER SOFTWARE
HOMER can answer questions that come up
when installing renewable energy systems, e.g.: is it cost
effective to add PV panels to the grid-connected house,
can the new system serve if the load is growing, or what
should be the electricity price for the PV system to be cost
effective? HOMER has been used as a tool to calculate
technical feasibility and economical viability of renewable
energy system in different fields, for examples, in large
and small hotels [9,10,11], a university building [12], and
remote area and stand alone systems [13,14].
For a system to be technically feasible, the hourly
energy production (from generation and grid purchases)
should be able to satisfy the hourly load and constraints
determined by the user [2,3]. Loads and energy production
over a one-year period is calculated and if there is energy
surplus or deficit, HOMER decides what to do with it. The
surplus can be thermally/electrically stored or sold to the

B5.5-2
grid. The deficit can be resolved by purchasing energy
from the grid or discharging the stored energy.
For a system to be economically viable, HOMER
estimates the life-cycle costs of the system by calculating
its net present value (NPC). The net present value can be
defined as the present value of the total cost and revenue
incurred over the lifetime of the project. HOMER uses the
equation below to calculate NPC:
CRF
TAC
NPC 
(1)
where TAC is the total annualized cost including capital
costs, replacement costs, operation and maintenance costs,
fuel costs, electricity purchased, and revenues from selling
excess electricity and the salvage value of the components
and CRF is the capital recovery factor:
1)11(
)11(


 N
N
i
i
CRF (2)
where N is the project lifetime and i is the annual real
interest rate, given by the following equation:
f
fi
i



1
'
(3)
where i’ is the nominal interest rate and f is the annual
inflation rate.
HOMER calculates the levelized cost of energy
(COE) using the following equation:
salesgriddefDCprimACprim EEEE
TAC
COE
,,, 

(4)
where TAC is the total annualized cost, Eprim,AC is the total
amount of AC primary load served per year, Eprim,DC is the
total amount of DC primary load served per year, Edef is
the total amount of deferrable load served per year, and
Egrid,sales is the total grid sales per year.
METHODS
Before HOMER simulates and optimizes the
system, we are required to input data. Those inputs are the
electric load, equipment to consider, resources (solar
resource in our case), economics, system control,
emissions (if applicable), and constraints.
Load
Load inputs are collected from hourly load
observation of the house over a 24-hour period. Data were
taken in May 2015. Figure-1 shows the hourly load profile
of the house in a day. The peak load of the house during
observation was less then 400 watt and the daily electricity
consumption is about 4.7 kWh.
HOMER synthesizes the data to estimate the
hourly load profile over a year. To do so, HOMER asks
for day-to-day and time-step-to-time-step random
variability, which are 15% and 20%, respectively. Figure-
2 shows HOMER estimation of the seasonal load profile
of the house. The estimated peak load is now 658 watt.
Figure-1. Hourly load profile of the house
Equipment to consider
The house will be installed with a grid-connected
PV system which includes PV arrays, converters, and
batteries (optional). The grid provides alternating current
(AC) and serves the load directly. PV panels and batteries
are connected to direct current (DC) bus and converted to
AC by a converter.
Figure-3. The proposed grid-connected PV system
The grid sells electricity at a flat rate of 0.1
USD/kWh. The house will use net metering with the net
purchases calculated monthly. The surplus energy of the
PV system will be fed into the grid with a net metering
system in which the meter run backward when the excess
energy is being fed into the grid. However, the sellback
price is zero if the energy sales exceed the purchases.

B5.5-3
Figure-2. Estimation of the seasonal load profile
The chosen PV panels, which have no tracking
system, are expected to operate for 20 years with derating
factor, slope, azimuth, and ground reflectance of 90%, 8o
,
180o
, and 20%, respectively. The investment and
replacement costs of the system are the same: 3000
USD/kW [15]. The sizes to consider are 0.5 kW, 1 kW, 2
kW, and 3 kW.
The batteries are Trojan T-105 with the nominal
voltage of 6 volt, the nominal capacity of 225 Ah (1.35
kWh), and the lifetime throughput of 845 kWh. The
investment cost and replacement cost are estimated to be
125 USD and its related O/M cost is 5 USD. The sizes of
the batteries to consider are 0, 1, 2, and 3 batteries.
The converter costs 1000 USD/kW. Its lifetime is
expected to be 15 years and its efficiency is 90%. When it
converts AC to DC, the efficiency is estimated to be 85%.
We consider converters of 0.5 kW, 1 kW, 2, kW, and 3
kW.
Resources Inputs
Since we propose a PV system, HOMER requires
solar resource input to calculate hourly PV power
production over the year. The data were collected from
NASA. Figure-4 shows Global horizontal solar radiation
near the site.
Figure-4. Global horizontal solar radiation
Economic Inputs
The economic inputs required by HOMER are the
annual real interest rate and the lifetime of the project,
which are 7% [16] and 20 years, respectively. The annual
real interest rate is difference between the nominal interest
rate and the inflation rate [17].
Constraints
We allow the maximal capacity shortage of 5%.
We also set that the operating reserve should be at least
10% of the hourly load, 25% of solar power output, an
50% of wind power output.
RESULTS
When all inputs are provided, HOMER is ready
for simulation an optimization of the system
configurations that are specified previously. HOMER
calculates load and the available resources and discards
any configurations that cannot satisfy the load given
constraints that were specified previously. This infeasible
configurations are not shown in optimization and
sensitivity results.
Figure-5 shows the optimization results of each
configuration. The grid-only system is still the cheapest
choice based on the net present cost (NPC) with the
current electricity price of 0.1 USD per kWh. The grid-
only system NPC is 1,829 USD. The total net present cost
of the grid-connected PV system is 3,823 USD and its cost
of electricity (COE) is 0.209 USD/kWh. Its renewable
fraction is 38%. Grid with 0.5 kW PV panels and 0.5 kW
converter is the optimal configuration for the grid-
connected PV system. Bigger capacities of PV panels or
converter or adding batteries lead to higher NPC and
capital cost. With the current prices of electricity and PV
system, grid-connected PV system is not the most cost
effective choice.
Results in Figure-5 also mean that all the possible
configuration is technically feasible. Technical
characteristics of the grid-connected PV system are shown
in Table-1 to Table-3. Table-1 shows that 62% of the load
is estimated to be served by the grid and only 38% by the
PV arrays. Table-2 shows that most of the electricity
(86%) is to serve the load and only 14% is sold to the grid.
Table-1. Electricity production of the Grid/PV system
Production kWh/yr %
PV array 885 38
Grid purchases 1,420 62
Total 2,305 100
0.0
0.2
0.4
0.6
0.8
1.0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
0
1
2
3
4
5
6
7
DailyRadiation(kWh/m²/d)
Global Horizontal Radiation
ClearnessIndex
Daily Radiation Clearness Index

B5.5-4
Table-2. Electricity consumption of the Grid/PV system
Consumption kWh/yr %
AC primary load 1,726 86
Grid sales 289 14
Total 2,016 100
Figure-5. Optimization results
Figure-6. Hourly profiles of load and PV electricity production on the first week of May
Table-3 shows the performance of PV arrays. Its
mean output is predicted to be 0.1 kW or 2.42 kWh/day. It
has 20% capacity factor and 51% penetration. Total
electricity production is 885 kWh/year. The system
operates 4,384 hour/year to produce electricity.
Table-3. PV arrays performance
Quantity Value Units
Rated capacity 0.5 kW
Mean output 0.10 kW
Mean output 2.42 kWh/d
Maximum output 0.54 kW
Capacity factor 20.2 %
PV penetration 51.3 %
Total production 885 kWh/yr
Hours of operation 4,384 hr/yr
The comparison of hourly profiles of load and PV
electricity production is shown in Figure-6. The figure
shows that PV electricity production during the day can
satisfy load. During the night, however, the grid should
serve the load.
Sensitivity analysis
The present study performs sensitivity analysis
with the following scenarios: PV panel price reduction to
50%, 25%, and 10% of the current price, electricity price
increase from 0.1 USD to 0.15, 0.2, and 0.25 USD, and
CO2 penalties of 10 USD/ton CO2, 25 USD/ton CO2, and
50 USD/ton CO2. The main purpose of sensitivity analysis
are to find out in which scenarios the grid-connected PV
system is more cost effective than the grid-only system.
Figure-7 shows optimal systems for different PV
capitals and electricity prices. CO2 penalty is not
applicable. The figure shows that, with the current prices,
the grid-only system is still the most cost effective choice.
Even when the PV capital is reduced to half its current
price and the electricity price increases by 50% (0.15
USD/kWh), the grid-only system is still superior to the

B5.5-5
other options. If the PV capital drops to 20% its current
price, the Grid/PV system is the best choice.
Figure-7. Optimal systems for different PV capitals and
electricity prices. No CO2 penalty.
Figure-8. Optimal systems for different PV capitals and
electricity prices. CO2 penalty is 50 USD/ton CO2
In a more extreme case in which the CO2 penalty
is 50 USD/ton CO2, similar patterns exist (see Figure-8).
Only when the PV capital is reduced to half its current
price and the electricity price doubled to 0.2 USD/kWh do
we have the Grid/PV/Battery system to be more cost
effective than the grid-only system.
In another extreme case, in which the PV capital
is 25% its current price, the Grid/PV system is the most
cost effective when the electricity price increases by 50%.
Figure-9 shows the scenario.
Figure-9. Optimal systems for different electricity prices
and CO2 penalties.
CONCLUSIONS
The present study provides technical and
economical analyses of a grid-connected PV system for a
small house located in Bukit Jimbaran, Bali. The present
study is supposed to answer two main questions: is it cost
effective to add PV panels to the grid-connected house? or
what should be the electricity price for the PV system to
be cost effective?
Results shows that each configuration is
technically feasible. They can satisfy load and constraints
set by the user. However, HOMER shows that, at the
current electricity and PV panel prices, the grid-only
system is much more cost effective than the grid-
connected PV system. The NPC and COE of the grid-
connected PV system are 3,823 USD and 0.209 USD/kWh,
respectively, which are much higher than those of grid-
only system (NPC = 1,829 USD and COE = 0.1
USD/kWh). They are about twice as much as those of
grid-only system.
Sensitivity analysis shows that a CO2 penalty
policy alone does not have strong impact on promoting the
grid-connected PV system to be more cost effective than
the grid-only system at the current electricity price. The
same conclusion is true for the scenario of electricity price
increase only or PV capital reduction only. Only when all
scenarios are applied simultaneously are the grid-
connected PV system more cost effective than the grid-
only system.
REFERENCES
[1] Rhismawati, N.L. Pelanggan Masuk Daftar Tunggu
PLN Bali. Antara News. 25th of February, 2015.
http://bali.antaranews.com/berita/68606/26800-
pelanggan-masuk-daftar-tunggu-pln-bali. Accessed
08/13/2015.
[2] Santika, W.G., Sudirman and Suamir, I.N.
Feasibility Analyses of Grid/Wind/PV Hybrid
Systems for Industrial Application. ARPN Journal of
Engineering and Applied Sciences. In press.
[3] Lambert, T., Gilman, P., and Lilienthal, P. 2006.
Micropower system modelling with HOMER. In F.A.
Farret, M.G. Simoes (Eds.). Integration of alternative

B5.5-6
sources of energy. John Wiley & Son, Inc. pp. 379-
418.
[4] Steg, L. and Vlek, C. 2009. Encouraging pro-
environmental behaviour: An integrative review and
research agenda. Journal of Environmental
Psychology. 29(3): 309-317.
[5] Midden, C.J.H., Kaiser, F.G., McCalley, L.T. 2007.
Technology’s four roles in understanding individuals’
conservation of natural resources. Journal of Social
Issues. 63(1): 155-174.
[6] Gellings, C. W. 1985. The concept of demand-side
management for electric utilities. Proceedings of the
IEEE. 7310: 1468-1470.
[7] Fels, M.F. and Keating, K.M. 1993. Measurement of
energy savings from demand-side management
programs in us electric utilities. Annual Review of
Energy and the Environment. 18(1): 57-88.
[8] Santika, W.G., Antara, D.M.S. and Harmini, A.A.N.
2013. Memotivasi perilaku hemat energy dan ramah
lingkungan di sebuah hotel. Bumi Lestari Journal of
Environment, 13(2): 374-383.
[9] Dalton, G.J., Lockington, D.A. and Baldock, T.E.
2009. Feasibility analysis of renewable energy
supply options for a grid-connected large
hotel. Renewable Energy, 34(4), 955-964.
2008. Feasibility analysis of stand-alone renewable
energy supply options for a large hotel. Renewable
Energy. 33(7): 1475-1490.
2009. Case study feasibility analysis of renewable
energy supply options for small to medium-sized
tourist accommodations. Renewable Energy. 34(4):
1134-1144.
[12] Ngan, M.S. and Tan, C.W. 2012. Assessment of
economic viability for PV/wind/diesel hybrid energy
system in southern Peninsular Malaysia. Renewable
and Sustainable Energy Reviews. 16(1): 634-647.
[13] Rohani, A., Mazlumi, K. and Kord, H. 2010.
Modeling of a hybrid power system for economic
analysis and environmental impact in HOMER. In
Proceeding of 18th Iranian Conference on Electrical
Engineering (ICEE), 2010, pp. 818-822.
[14] Zoulias, E.I. and Lymberopoulos, N. 2007. Techno-
economic analysis of the integration of hydrogen
energy technologies in renewable energy-based
stand-alone power systems. Renewable
Energy. 32(4): 680-696.
[15] US Department of Energy. 2014. Photovoltaic
System Pricing Trends: Historical, Recent, and Near-
Term Projections. Information on
http://www.nrel.gov/docs/fy14osti/62558.pdf.
(accessed on 05.07.2015).
[16] The World Bank. Information on
http://data.worldbank.org/indicator/FR.INR.RINR
(accessed on 05.07.2015).
[17] T. Givler, P. Lilienthal. 2005. Using HOMER®
software, NREL’s micropower optimization model,
to explore the role of gen-sets in small solar power
systems. Case Study: Sri Lanka. National Renewable
Energy Laboratory, Golden, Colorado.

FEASIBILITY ANALYSIS OF A GRID-CONNECTED PV SYSTEM FOR HOME APPLICATION

More Related Content

What's hot (18)

Viewers also liked (6)

Similar to FEASIBILITY ANALYSIS OF A GRID-CONNECTED PV SYSTEM FOR HOME APPLICATION (20)

More from Wayan Santika (8)

Recently uploaded (20)

FEASIBILITY ANALYSIS OF A GRID-CONNECTED PV SYSTEM FOR HOME APPLICATION