You are on page 1of 6

ACTEA 2009

July 15-17, 2009 Zouk Mosbeh, Lebanon

MATLAB Based Modeling Tool for Designing, Predicting and

Analyzing Grid Tied Photovoltaic Systems
A. Assi, IEEE Senior Member, M. Jama, and K. Al Kathairi, IEEE Members

Abstract Planning a photovoltaic (PV) system is complex.

Many different factors must be taken into account such as site
information, module type and inclination of the modules, level
of irradiation, choice of the most suitable inverter, etc. To ease
the planning process, computer based tools must be utilized.
In this paper, a MATLAB-based modeling tool to design,
predict and analyze grid tied PV systems is presented. This
software has a user friendly interface which requires a
minimum of data in order to present your optimum system
and examine its performance in a clear graphical form. The
user can explore different combinations and scenarios to meet
the proposed design requirements. Simplicity, accuracy and
ability to adapt the designed tool to any updates in both site
information and system components would probably make it
an excellent tool for educational and research purposes.


hese are prosperous times for the Middle East and the
Gulf region, as the demand for oil continues to rise. But
like the rest of the world, the region is not immune to
the negative impact of fluctuating oil prices, particularly as
it consumes the black gold for its own air-conditioning,
electricity and drinking water. There is an alternative
energy solution, however, and it is directly overheadthe
sun. We believe that in the coming years, both governments
and industries in the Middle East will face the pressure of
environmental worries and power shortages related to oilbased economic growth. As alternative energy projects
succeed, they will attract more investors and expand in size
and scope [1].
The environmental impact of this domestic dependence
on oil is large. According to the United Nations, Qatar,
Kuwait and the UAE have the three highest per-capita rates
of greenhouse gas emissions, roughly 15 tons higher than
that of the United States. Power and fuel shortages are
beginning to strain the regions economies. As
demonstrated in Bahrain, the consumption of fossil fuels
for domestic needs eats into potential revenues. The real
cost of producing electricity is nearly seven times higher
than the selling price of electricity, which is 3 cents per
Manuscript received May 31, 2009. This work was financially
supported by the UAE University under the contract no. 07-04-7-11/09.
A. Assi is associate professor with the Electrical Engineering
Department, UAE University, 71555 Al Ain, UAE, phone: 971-3-713
3609; fax: 971-3-762-3156; e-mail:
M. Jama is a research assistant in the Electrical Engineering
Department, UAE University, 71555 Al Ain, UAE, e-mail:
K. Al Kathairi is a research assistant in the Electrical Engineering
Department, UAE University, 71555 Al Ain, UAE, e-mail:

978-1-4244-3834-1/09/$25.00 2009 IEEE


kilowatt. In the UAE, the worlds third largest per-capita

consumer of water, drinking water is derived from an
expensive and energy-intensive desalination process that
relies heavily on gas-powered turbines. If the country were
to pursue a renewable energy source to produce drinking
water, it could better meet its projected demandexpected
to increase another 44 percent by 2025with limited
pressure on electricity production, less impact on the
environment and a return of revenues from the sale of
carbon credits on the market. As a result of these issues and
as governments around the globe consider cheaper energy
and sustainability as key issues, the Middle East should
meet changing needs and tap into the growing alternative
energy market [1].
The Middle East, and in particular the nations of the Gulf
region, have extensive experience drawing foreign
investment into the area. UAE has strongly emerged into
PV market, by launching its outstanding MASDAR
initiative. The AED 15 billion initiative would establish a
strong renewable energy society in the country. This
initiative is directed to investments, manufacturing future
energy solution, education and R&D, carbon management
and sustainable development and planning.
PV systems can be connected to the public electricity
grid via a suitable inverter in what is known as grid-tied PV
systems. Energy storage is not necessary in this case. On
sunny days, the solar generator provides power for the
electrical appliances in the house and excess energy is
supplied to the public grid. During night and overcast days,
the house draws its power from the grid [2]. PV systems
operating parallel to the grid have a great technological
potential especially in the UAE, which is subjected to an
average of 3500 sun hours of irradiance annually [3]. Since
the UAE has a high quality modern power network with
low electricity tariffs, it would be not easy to come up with
a solar system that is competitive economically to the
current power resource in the country. On the other hand,
the country is witnessing a tremendous revolution in all
fields and sectors, which in turn will increase the reliance
on fossil fuel resources, so the country is really in need to
reduce the pressure on conventional energy resources by
introducing new inexhaustible sources of power. However,
without subsidies from the government or utilities, PV
solutions are not yet financially competitive [2, 4].
In this work, a MATLAB-based modeling tool to design
and accurately predict the technical performance of grid
tied PV systems is presented.


PV systems may be connected to the public grid. This
requires an inverter for the transformation of the PVgenerated DC electricity to the grid AC electricity at the
level of the grid voltage. National and even regional
regulations differ widely with respect to the policy of
interconnection requirements and reimbursing for PVgenerated electricity fed into the grid. In order to support
the production of PV-generated power, some utilities offer
a better price for the kWh fed into their grid than they
charge for the kWh from the grid. In other locations a oneto-one ratio is applied which means the same kWh-price for
both flow directions. The third version is to pay less for the
generated PV power fed into the grid than for that sold to
the consumer [1]. Comparing the rates, the fixed rates for
the power connection have to be considered also.
Depending on the kind of tariffs, one or two electricity
meters have to be used at the point of utility connection.
Fig.1 shows a block diagram of a grid-connected PV
system suitable for building integration. In grid-connected
applications, PV systems must compete against the cost of
the conventional energy source used to supply the grid. PV
systems are particularly cost-effective when the utility load
and solar resource profiles are well matched [5]. This is, for
example, the case in areas with high air-conditioning loads
(such as the UAE) that have their peaks during the peak
sunshine hours of the summer day.

Fig.1: Grid connected PV system schematic [6].


The MATLAB based modeling tool consists of a number
of consecutive procedures, that can be categorized into four
main procedures, which are the solar irradiance study,
building load profile, PV array installation, and PV system
design as shown in Fig.2. Moreover, the software offers an
analytical tool examining the electrical performance of the
designed PV system.
A. Solar Irradiance
The first step in designing the grid tied PV system is to
specify the site information such as the city, the tilt angle
and the reflectivity factor, these information are enough to
calculate the daily average global irradiance [5], which is a
very important parameter in the PV system as shown in


Fig.2: Modeling tool architecture.

Fig.3: Site information section.

Long series (many years) of daily weather data are

sometimes required for particular purposes, for example,
when studying the long term reliability of PV systems.
However, long series of historical data are hard to obtain. A
typical meteorological year (TMY) is a collation of selected
weather data for a specific location, generated from a data
bank much longer than a year in duration. It is specially
selected so that it 'showcases' the range of weather
phenomena for the location in question [7]. The software
was equipped with TMY data for number of cities in the
UAE, such as Abdu Dhabi, Al Ain, Dubai, etc. Each TMY
was generated using 13 year daily weather data that was
obtained from the National Center of Meteorology and
Seismology in Abu Dhabi [8]. This database includes the
daily mean global radiation on horizontal, maximum
temperature, sun hours, wind speed and humidity. The daily
solar radiation on horizontal and daily sun hours in Al Ain
city is demonstrated in Fig.4.
The average daily solar hours in Al Ain was estimated
to be around 9.8 hours, while the corresponding daily
global solar insolation on horizontal was approximately
6.26 KWh/m2. Al Ain is subjected to an average annual
solar radiation of 2285 KWh/m2, which is obviously a very
significant amount of irradiance [3].

examples include residential, commercial and industrial),

temperature and holiday seasons.
In the load profile section, the user will be able to select
the load profile model from the pop-up menu shown in
Fig.6. The menu includes three load profile models for
residential buildings (villa), each one of these models
represent a level of power demand behavior, for example,
Model A represents the average power consumption, while
Model B and C represent the maximum and minimum
power consumption, respectively [9].

Fig 4: (a) Daily solar radiation in Al Ain (b) Daily solar hours in Al Ain.
Fig.6: load profile section

B. PV Array Installation
The software prompts the user to enter the dimensions of
the array installation area (width and height) as shown in

These three load profile models were developed by

analyzing and studying a 3 year annual load profile for
tens of villas in Al Ain. These large number of load
profiles were obtained from Al Ain Distribution Company
(AADC), to develop a realistic and accurate power demand
profile representing the environment and the life-style in Al
Ain. After selecting the proper load profile model, the user
should specify the coverage factor as shown in Fig.6, which
is the percentage the PV system is expected to cover from
the annual home power demand. By this way, the program
will give an indication message showing to which degree
the designed PV system covers the annual demand.

Fig.5: Array installation parameters.

The program will fit the module on the suggested area

without omitting the separation between modules in each
string and between strings as well. The user will be able to
control the accuracy of installation area filling by adjusting
the area mismatch variable, which is the allowable error
margin between the suggested installation area and the
actual PV array area. The user also can specify the modules
alignment whether it is horizontal or vertical as shown in
Fig.5. This feature has been added to make sure that the
modules alignment matches with the shape of the
installation area, hence ensuring a good appearance of the
PV array as part of the building.
C. Building Load Profile
Before designing a PV system for a building, whether it
is a residential, commercial or public building, a full load
assessment should be performed. The load assessment is
described by a graph called the load profile. This graph
shows the variation in the electrical load versus time [7]. A
load profile will vary according to customer type (typical


D. PV System Design
The next step is to design the grid connected PV system
starting with the DC circuitry of the system, which
concerns the PV array configuration. The software database
includes various types of modules each has different
electrical and mechanical characteristic. To make the
design procedure simpler and faster, the software has some
routines that sort out the most appropriate modules, in
terms of fitting the installation area, so instead of going
through all modules, only the modules that would fit the
suggested installation area will be displayed for the user as
shown in Fig.7.
After the installation area dimensions are confirmed, the
user will have the full chance to select between groups of
modules based on the required electrical behavior not the
modules installation issues, which has been already
resolved. Once the user selects one of the listed modules,
the software determines all possible PV array
configurations which are number of parallel strings (group
of modules in series) and number of modules per string.

back and start from scratch. The duplicated PV arrays are

then connected to the same inverter, so the partitioning in
the DC side is in the PV array level. In the AC side the PV
system can be divided into independent sub-systems, each
subsystem has its own inverter. Then all the subsystems are
combined together via a combiner enclosure to connect it to
the home distribution panel as shown in Fig.9.

Fig.7: The DC analysis section.

The software was designed to show all the possible

configurations, where the number of strings is less than the
number of modules per string as shown in Fig.7, this is to
limit the number of choices presented for the user and
because of the fact that if the number of strings is greater
than the number of modules per string, high overall current
and low overall voltage will result, which is an unfavorable
output from the technical point of view [10].
After the module and the PV generator configuration are
selected, the program searches the inverters that would
work properly with the designed PV array. The inverter
searching procedure is done by comparing the output
voltage, current and power of the DC generator with the
DC input ratings of the inverter [11]. The program lists all
the inverters that can operate properly from the technical
point of view as shown in Fig.8. All selected inverters
would probably work in all conditions such as maximum
irradiance days (maximum current) and maximum ambient
temperature days (minimum voltage).

Fig.8: Inverter selection section.

The user will be free to select the suitable inverter for his
PV installation based on his budget and level of system
reliability he prefers. The software gives a nice clue for the
designer to select the best inverter, based on the inverter
sizing factor range. The higher the sizing factor range, the
better the design would become [5]. The tool offers the
opportunity of designing a partitioned grid tied PV system.
The partitioning feature is available in both DC and AC
circuits of the PV system. In the DC side of the system, if
the system is composed of several PV generators located at
different locations, while they all have the same design
parameters and specifications, the user could design one PV
generator and then specify number of identical PV
generators as shown in Fig.7. This feature is only to
duplicate the system, preventing the designer from going


Fig.9: Schematic of an AC partitioned grid tied PV system.

Picking up the appropriate components for the PV

system is very important to reserve the systems high level
of reliability and safety [12]. The software determines the
cables sizes at every spot of the system. In electrical power
systems, most conductors are restricted from operating on a
continuous basis at more than 80% of their rated ampacity
[13]. This principle also applies for over-current protection
devices. However, the current carrying conductors used in
PV systems are further de-rated by a factor of 80% due to
the manner in which PV modules generate power in
response to the sun light and because the noon-time
intensity of the sunlight may exceed the standard test
conditions [14]. The ampacity of conductors and the sizing
of over-current devices is an area that requires careful
attention by the PV system designer/installer. Temperatures
and wiring methods must be addressed for each site [14].
Calculations start with the 125% of Isc value to comply
with the UL 1703 requirements, and additional 125% must
then be used for code compliance. Finally, the cable
ampacity is adjusted for the site operating temperature
using the cable de-rating tables [15]. The designer is
requested to specify the ambient temperature of operation
and the cable insulation type at a specific portion of the PV
system, and then the software will determine the adequate
cable size and protection device rating for that part of the
system as shown in Fig.10.

Fig. 10: Cabling and protection section.

The program is designed to determine the cable sizes for

the PV module interconnection cabling, the DC combiner
Inverter cabling, the Inverter AC combiner cabling (if

more than one inverter is used), the equipment grounding

cable, and the DC grounding cable.

Global Radiation vs Inclination Angle through a Year


Grounding all PV system components is an important step

to make the system immune to all kinds of hazards caused
by electrical faults. After determining the cable size, the
software finds the proper protection devices ratings to
protect each cable and all attached equipment from overcurrent that may occur due to any type of faults [13].


Glabal Radiation (kW)/m2)



-0.1 0









Fig. 12: Monthly mean global irradiance for different inclination angles

The developed tool was used to study the PV systems in

the UAE environment using various scenarios and case
studies. The first task was to exactly identify the optimum
panel inclination angle in the UAE. The optimum
inclination angle for maximum irradiance in Al Ain UAE
was found to be approximately 20. However, if the PV
panels are tilted at angles between 10 - 40, it would be
subjected to a permissible amount of irradiance. At angle
20, the average irradiation values in all months are pretty
close to each other unlike when the panels are tilted at
bigger angles like 40, 50, etc [5] as shown in Fig.11.

In order to showcase the full capabilities of this

simulation tool, a case study to design grid tied PV system
was conducted. The PV system was divided into three subsystems; each has its own full circuitry in both the DC and
AC sides. Table 1 demonstrates the main input parameters
of the case study.
Table 1: Main input parameters of the case study.
Al Ain UAE
Geographical information


24 12' North

Longitude: 55 46' East

# of PV subsystems

Tilt angle

15 (1st array) - 30 (2nd & 3rd arrays)

Installation area

26.5 x 1.7 m (1st array) - 6.25 x 3 m

(2nd & 3rd arrays)

Reflectivity factor
Module model
Array Configuration
Fig. 11: Average global irradiance in Al Ain at an inclination angle of 20

In real life applications, mounting the PV modules at

angles less than 15 is mainly not an appropriate solution,
since it would be subjected to more dust accumulation,
which will reduce the life span of the PV system [16].
Fig.12 shows the monthly average global irradiance for
different inclination angles. It can be noticed that very low
irradiance values are obtained in the summer when the
panel is tilted at an angle of 90, this result is reasonable
since the declination angle of the sun is at its maximum in
the summer months, hence minimum direct irradiance is
received by the PV panel at bigger inclination angles.
The three subsystems are all combined together using a
protected combiner enclosure. After designing the system,
the electrical performance of the system was examined
using the analytical tool of the simulation software. Fig. 13
& Fig.14 show some of the most important electrical
parameters of the designed PV system.


Inverter model

BP SX 3200 (1st array) - BP SX 160B

(2nd & 3rd arrays)
1 x 31 (1st array) 2 x 12 (2nd & 3rd
SMC 7000 HV (1st array) - SB4200TL
HC (2nd &3rd arrays)

Fig. 13: The maximum DC output power of the PV generator in subsystem 1.

PV system installation. Guidance and helping features were

added so that the user will be able to understand all
technical parameters and terminologies while going through
the design process. All modules and inverters data sheets
were added to help the designer to choose the most
appropriate choice of PV components. Also the software
has the ability to provide a full and comprehensive
summary of the designed system as a final output
documentation mean.
Fig. 14: The total monthly energy yield of sub-system 1.


It is obvious that the monthly energy consumption values

are much larger than the energy produced by the PV
system, which is an expected result. The PV system was
basically designed to cover up to 25% - 30%, while the
actual PV system coverage factor is approximately 28%,
keeping in mind that the overall wiring losses is around 5%
(typical value). The monthly total injected energy was
equal to zero in all months, because all the generated PV
power is consumed by the villa loads. Fig 15 shows the
monthly building energy consumption (green) versus the
monthly PV energy produced by the PV system (blue).The
designed PV system has an overall system efficiency of
15%, which makes sense due to the large losses.

The authors would like to thank engineers Sina Fattahi

and Ibrahim Al Shehhi for their help in building various
databases, the National Center of Meteorology and
Seismology (Abu Dhabi) for providing weather data, and
Al-Ain Distribution Company (AADC) for providing load
profiles. Thanks also go to UAE University for the financial
support under contract no. 07-04-7-11/09.



Fig. 15: Building energy consumption Vs PV energy output.


The developed computer modeling tool has a strong
capability to predict PV systems power generation. The
strongest feature in this modeling tool is the use of daily
weather data and the utilization of simple irradiance models
that mainly are as accurate as some other complex models.
This tool is flexible enough to adapt quickly if some other
locations are investigated. Besides, the tool provides a
variety of PV products such as modules, inverter, etc. The
tool has been designed to help the user to easily build a grid
connected PV system and analyze the main electrical
parameters in a course of a year. To make the design
process simple and quick, some smart mechanisms have
been added to the tool to simplify components selection
procedure and continuously check that none of the design
constraints have been violated. All the procedures in this
tool are in compliance with the standard codes issued for



R. Ziegler, A. Ariatti, M. De Clercq, Tapping the Sun: Solar

Energy in the Middle East Executive agenda, A.T. Kearney.
G. G. Merino , David Jones, laverne e. Stetson, 2001. Performance
of a Grid-connected Photovoltaic system using Actual and kriged
hourly solar radiation, Nebraska, USA.
A. Assi, M. Jama, K. Al Khathairi, I. Al Shehhi, S. Fattahi,
Predicting the Electrical Behavior of Grid-Tied Photovoltaic
Systems in Al Ain UAE / Model and Case Study, UAE, ICSET,
E. Romn, R. Alonso, S. Elorduizapatarietxe, P. Ibez," Economic
Analysis Of Modular PV Systems For Building Integration", 21st
European Photovoltaic Solar Energy Conference, 4-8 September
2006, Dresden, Germany.
Antonio Luque & Steven Hedgus, 2005. Handbook of Photovoltaic
Science and Engineering, 1st ed, England.
Phil Livingston, First Steps in Renewable Energy for your Home,
Home Power Magazine, April & May 2007 edition, pp.68-69.
Ian Woodfeden and Chris Laforge, Getting Started with Renewable
Energy (Porffessional Load Analysis and Site Survey, Home Power
Magazine, August & September 2007 edition, pp. 44-46.
National center of meteorology and seismology, department of
meteorology, climate division, Abu Dhabi, UAE.
Al Ain Distribution Company (AADC).
Ian Woofenden, Small Scale System for an on Grid Island
Getaway, Home Power Magazine, February & March 2007 edition,
SMA Sunny Boy, Installation and Operation Manual (Grid Tied
String Inverter for PV Systems).
Friedrich Sick & Thomas Erge, 1996, The design handbook for
architects and engineers, 1st ed, Freiburg, Germany.
IEEE P929. Recommended practice for utility interface of
photovoltaic (PV) systems. Draft 9.
IEC 1173. Overvoltage protection for PV power generating systems guide 1992-08.
National Electrical Code (NEC). USA 2005.
UL 1703. Standard for flat-plate photovoltaic modules and panels. 20
November 2000.