The 10th International Renewable Energy Congress (IREC 2019)

Feasibility of Replacing Traditional Streets Lighting

in Al-Bayda City with Solar Cells Technology
Aiman NOUH, Aboubakr SAED, and Ashraf ELMURTAGA
Electrical & Electronic Engineering Department
University of Omar Al-Mukhtar
Al-Byada, Libya
[email protected]

Abstract—The general electricity company of Libya The main focus of this paper is to determine the options
(GECOL) suffers from frequent interruptions in recent years, that are available to replace grid-powered street lamps system
sometimes reaching a general blackout. This is the result of with a stand-alone PV system. Also, knowing if the utilization
several factors like increasing of demand, lack of maintenance, of stand-alone PV system for long-term will saving the
the attack on company's infrastructure, political and economical electricity and warrants the conversion to new lamps built off
instability in the country. The paper comes within the framework the power grid.
of estimating the demand resulting from the lighting of eight
main streets in Al Bayda city as well as the feasibility of covering The work presented in this paper is exposed in two parts.
this demand using stand-alone photovoltaic solar panel In the first, the region of study and selected streets will be
technology powered light emitting diodes. The software presented. Then, the data collected will be analyzed and
Matlab/Simulink is used to implement and simulate our studied compared for three studied systems. These last are the actual
system. The simulation results will be exposed and discussed. street lighting system with HSP lamps, the actual street
lighting but the HSP lamps are replaced by LED lamps, and
Keywords—street lighting; photovoltaic solar cell; storage the stand-alone PV street lighting system. The second part is
battery; DC/DC converter; charger controller; LED lamp devoted to clarify a Matlab/Simulink implementation of an
I. INTRODUCTION overall model that is PV solar street lights system . Finally, it
finished by exposing the simulation results and discussion.
The Renewable Energy Authority of Libya (REAOL)
predicts that the electricity demand in the country is estimated II. CASE STUDY
to be growing by 7% per year and forecasting at around of 8 The city of Al-Bayda was selected as a study case. It is
GW in 2020 [1]. The classic street lighting system is one of located at the east-north of Libya. Fig.1, shows the main
many loads constituent this demand like residential, selected streets in the city.
commercial, common public transportation, industrial,... etc. It
is consuming a percentage of generated power sharing by an By surveying the selected streets in the city accompanied
electricity grid. This percentage has an approximately by a senior engineer from the distribution department of the
estimated value of about 20% [2][3]. Due to the energy GECOL, the currently status of streets is identified. The
intensive and the high cost of actual Libyan street lighting surveying of streets has taken several weeks until getting the
system based on high sodium pressure (HSP) lams, this data listed in the Table I.
system is inefficient and non sustainable. In the past years,
using stand-alone photovoltaic (PV) was popular in remote
and rural areas to powering small loads like water pumping,
communication field, and cathodic protection station for oil
pipeline. So, to investigate more efficient light sources such as
photovoltaic powered Light Emitting Diodes (LEDs), a system
of PV panels, storage batteries, charge controller and LED
lamp must be adapted. This type of systems is called new
green light sources [4]. Today, Energy-efficient lighting is an
important factor for sustainable development and energy
saving strategies. Indeed, renewable energy utilization permits
the reduction of CO2 emissions and contributes to the
decreasing of fossil energy dependency [5][6][7]. Also, there
are secondary reasons to convert toward an independently
street lighting system. They are the transformation to urban
cities, business boom as well as the sensing of security and
safety at night. Fig. 1. Selected streets in Al-Bayda city.

978-1-7281-0140-8/19/$31.00 ©2019 IEEE

 TABLE I. STREETS COLLECTED DATA

Poles Towers
Street Name Length
Single Double N° Lamps
Northern Ring - 170 - - 7425.37 m
AL-Aorouba 196 166 8 6 8066.85 m
Hospital 37 63 - - 2302.35 m
South Ring - 76 3 4 2915.66 m
Cobree AL-Wesata - 40 - - 1276.94 m
-
New university - 56 - 1585.12 m

AL-Naseem - 62 - - 2980.99 m
Mol Jwahr - 60 - - 1954.36 m
Fig. 3. Total electricity consumption

To estimate the real electricity consumption and the cost, a The total cost is compared for the three proposed systems
comparison, between the actual street lighting system with as illustrated in Fig. 4. Here, it can be observed that the street
HSP lams, the actual street lighting system with LED lamps, lighting system based on LED lamps realizes an economy
and the PV street lighting system powered LED lamps, is percentage of 66% from the budget allocated to the currently
evaluated. The evaluation of consuming energy is based on the HSP lighting system. Although, the initial cost of the PV street
calculation of daily light hours, the monthly average daily lighting system is higher than the initial cost of conventional
light hours, and the monthly average night hours. While the HSP system.
evaluation of cost is taken into account the initial cost of the
components that compose the street lighting system like pole, However, it is an inexpensive solution for long term
foundation of pole, HSP lams, LED lamp, PV panel, storage compared to the HSP lamps system. The stand-alone PV
batteries, charge controller, lamp arm, cables, distribution box system saves money and achieves a percent of 41% over 20
and timer. In addition to that, the cost of electricity years compared to HSP system. Indeed, it saves fuel because it
consumption from gas station only, CO2 emission cost, and the is an independent fuel system. It employs PV panels to collect
maintenance cost are considered. the sun radiation then converting it to electricity.
Fig.2, shows the electricity consumption by street for one
year. It also shows that the street of AL-Aorouba consumes
more energy than the street of Cobree AL-Wesata and that is III. PERFORMING THE SIMULATION
because of the its length. Indeed, the HSP lams consumes To understand the dynamic behavior of a PV street lights
more electricity than LED lamps. While the PV system system, the software of Matlab/Simulink (R2014a version) is
consumes zero electricity from the utility grid. used. The studied system consists of one PV panel, one charge
controller, two storage batteries, and one LED lamp.
The total electricity consumption is shown in Fig. 3. It is
clear that the lighting system using LED saves about 75% of In order to implement and simulate our system, we make
the consumption that is needed by the lighting system using the following assumptions:
HPS, while PV Street lights System saves 100%.
• A one pole lighting unit is considered.
• Day light hours are eleven.
• The irradiance profile has been assumed.
• The simulation time is twenty four hours.

Fig. 5, shows an overall diagram of Simulink program for

PV panel feeding a street LED lamp. The output of PV
module is boosted by a DC/DC boost converter. The batteries
receive this energy and store it during the day. Then, they
supply the LED lamp during the night. The only input to the
system is the sun irradiance profile. Two switches are used to
control the charging and discharging. They cannot be closed
together, if one of them is closed the other must be opened.
Fig. 2. Electricity consumption by street.
 = − − (1)
+ +
= − −1 − (2)

With,

=( , + ∆ ) (3)

= (4)

Where, IPV, VPV are the current and voltage of cell, Iph, Iph,n
are photo-generated current and its value at standard test
condition, αI is the current/temperature variation coefficient, I0
Fig. 4. Total cost. is the reverse saturation current, VT is the thermal voltage, a is
the ideality factor of the diode, K is the Boltzmann's constant,
q is the electron charge, T, Tn are the temperature in Kelvin
and the nominal temperature, S, Sn are solar radiation and the
nominal solar radiation, RS is the equivalent series resistance,
RSH is the equivalent parallel resistance.
The I-V curve and the P-V curve that are considered as the
key to understand the behavior of the PV cell, module or
array, strongly depend on the internal components (RS, RSH)
and on the external conditions such as solar radiation (S) and
ambient temperature (T) [14][15]. The graph represents the
Fig. 5. Overall studied model
PV module characteristics (I-V curve), and (P-V curve) for
different irradiance values is illustrated in Fig. 7.

The physical complexity of the PV cell is usually treated Fig. 8, shows the used DC/DC converter model that has an
based on a representation by an equivalent circuit model. unregulated input voltage (supplied by the PV panel). The
There are three electrical equivalent circuits that represent the inductor (L) is charged from the input voltage during switch
behavior of PV cell. They are known as simple circuit, single conduction. At this period, the diode is reverse biased. When
diode circuit and two diodes circuit [8][9][10][11][12]. Fig. 6, the switch is OFF the inductor discharges current to the DC
shows the most popular (single diode equivalent circuit). The bus capacitor because the diode is forward biased during this
current source (Iph) represents charge carrier generation in the period. At the output terminals of boost converter, a sum of
semiconductor layer of the PV cell caused by incident the source voltage and the inductor voltage is seen. The
radiation. The shunt diode represents the recombination of altering between the diode and switch (IGBT device) is based
charge carriers at a forward-bias voltage (VPV + IPVRS). The on switch frequency [16][17][18].
PV current (IPV) at the output terminals is equal to the light- The converter uses a MPPT system based on incremental
generated current (Iph) less the diode current (ID) and the conductance method which automatically varies the duty cycle
shunt-leakage current (ISH). The series resistance (RS) in order to generate the required voltage then extracting
represents the internal resistance to the current flow, and maximum power. Firstly, this method measures the output
depends on the pn-junction depth, impurities, and contact current and voltage of the PV system in order to calculate the
resistance. The shunt resistance (RSH) resists the leakage conductance (G) and the incremental conductance (∆G).
current.

Fig. 6. PV equivalent circuit model

Referring to the equivalent circuit model, the equations of

Fig. 7. I-V and P-V curves of PV module
currents are [13]:
 The SimPower System library of Simulink suggests a
dynamic battery model parameterized to represent most
popular types of rechargeable batteries [19]. In this work, we
select two identically batteries with nominal voltage of 12V
and rated capacity of 120Ah each. They are connected in
series to supply a 24V load voltage as shown in Fig.3 10. They
are of Lead Acid battery type.
Finally, As a constant load consumes 100 W from 24V DC
supply, the LED lamp is modeled by a resistor of 5.76Ω in
order to simplify the simulation and reduce the simulation
Fig. 8. Circuit of boost converter
time.
The parameters of components that compose the studied
Secondly, by comparing the conductance and the overall model are given in Table II.
incremental conductance, the decision to raise or to decrease
the PV voltage is made until reaching the MPP at which the
derivative of the power is equal to zero.
The output power of PV array is given as:
= .
( . )
= = + .
(5)
1
. = +

By definition, the conductance and incremental

conductance are,

=
(6)
∆ =−
Fig. 9. Power derivative on P-V curve
Substituting in equation (5), yields:

1
. = −∆ (7)

From equation (7), it is clear that, the maximum power

point is only occurred if the difference between the
conductance and the incremental conductance is equal to zero.

=0 ⟹ =− ,∆ = ,

(8) Fig. 10. Storage batteries

>0 ⟹ >− ,∆ < ,
TABLE II. SIMULATION PARAMETERS
<0 ⟹ <− ,∆ > ,
PV Panel Boost Converter
If this difference has positive or negative value that is VOC 22.0999 V L 5 mH
means there is a deviation of power from MPP. So, the
ISC 8.36955 A CDCBUS 100 μF
algorithm goal is to track the voltage operating point in a
manner that conductance is equal to incremental conductance. VMAXP 17.7 V Switch frequency 5000 Hz

Referring to expression (8), the algorithm determines the IMAXP 7.62959 A Storage Battery
side at which the operating point is located. Then a PMAX 135.044 W Nominal Voltage 12 V
perturbation must injected to shift the operating point toward
LED Lamps Rated Capacity 120 Ah
the MPP and the perturbation is repeated until (dP/dV = 0).
This Process is illustrated in Fig. 9. Rated Power 100 W Initial SOC 80%
 IV. SIMULATION RESULTS AND DISCUSSION The battery current has the shape illustrated in Fig. 15.
Due to the complexity of the overall model and the long During the period of battery charging, this current is negative.
time of simulation, the simulation runs for about five days. and has the shape of solar radiation. Its value becomes
The only input to the studied system is vary as shown in constant positive during the battery discharging or supplying
Fig.11. This input represents the solar irradiance that starts the LED lamp.
with low values as the sun rises, then reaching its peak at the Fig. 16, shows the variation of load voltage. During the
midday and gradually decreases as the sun sets. Finally, it has solar radiation period, there is no load voltage because the
no value during the darkness. switch of load is opened and the PV panel converts the solar
The variation of PV voltage with respect to simulation irradiance to electricity and stocks it in the batteries. Through
time (or data points) is given in Fig. 12. It is clear that the the night period the load switch becomes closed while the
panel output voltage oscillates about a mean value of ten volt battery switch opens in order to disconnect the PV panel and
during the day hours, and that is because the settling of initial supply the LED lamps. Here, the value of load voltage is
duty cycle at about 0.6. This oscillation can be attenuated by doubled due to the series coupling of batteries.
acting on the input capacitor value of boost converter. During Finally, in order to knowing that the battery will be fully
the night hours, there is not electricity produce by PV panel. charged or not if its initial SOC is resettled at the remaining
Fig. 13, represents the behavior of PV current. The PV value in the battery after thirteen hours of discharging of the
current has the shape as solar irradiance and that is because the first night, we adjusted the SOC at about 50% then the
direct relationship between them. It has an existence during simulation is restarted. Referring to Fig. 17, it is clear that the
the period of solar radiation only. battery is only gained about 23.6% and not reached its fully
SOC. This means that the battery needs more time long than
The state of charge of battery is illustrated in Fig. 14. the day hours for completely charging. This time may exceeds
Because we assumed that the initial state of charge (SOC) of three consecutively days without loading. So, it is important
the battery is settled at 80%, the charging process starts from that the batteries used for PV street lighting application have a
this value. The battery needs more than eleven hours for fully minimum state of charge level of about 80%.
charging from the PV panel. In our simulation a battery of 120
Ah capacity reached a SOC of about 95.3% over the charging
period. During the night, the battery discharge in order to
supply the LED lamp. Through this period, the SOC decreases
from about 95.3% and reach a low level of about 51.6%. The
batteries used in our system have an autonomy of more than
13 hours.

Fig. 13. PV current variation.

Fig. 11. Variation of solar radiation.

Fig. 14. State of charge of battery.

Fig. 12. PV voltage variation.

Fig. 15. Battery current.
 [3] Nan Su Mon Aung. Zaw Htet Myint, Design of Stand-Alone Street
lighting System with LED. International Journal of Scientific
Engineering and Technology Research, Vol. 03, Issue 17, August 2014,
pp.3518-3522.
[4] J. A. B. Vieira, A. M. Mota, “A High-Performance Stand-Alone Solar
PV Power System for LED Lighting,” Hindawi Publishing Corporation,
ISRN Renewable Energy, Vol. 2013.
[5] Usenobong F. Akpan, Godwin E. Akpan, "The Contribution of Energy
Consumption to Climate Change: A Feasible Police Direction,"
International Journal of Energy Economics and Policy, Vol. 2, N°. 1,
2012, pp. 21-33
[6] Silpa Baburajan, "Cost Benefits of Solar-Powered LED Street Lighting
Fig. 16. Load voltage. System: Case Study-American University of Sharjah, UAE,"
InternationalResearch Journal of Engineering and Technology, Vol. 04,
Issue 02, Feb.2017.
[7] Ahmad Hemami, "Electricity and Electronics for Renewable Energy
Technology: An Introduction," CRC Press, Taylor & Francis Group,
LLC, 2015.
[8] Chaitanya T., Saibabu Ch., Kumari J.Surya, "Modeling and Simulation
of PV Array and its Performance Enhancement Using MPPT (P&O)
Technique," International Journal of Computer Science &
Communication Networks (IJCSCN), Vol 1(1), 2011, pp. 9-16.
[9] Hatice Okumus, Fatih M. Nuroglu, "Modelling and Simulation of PV
Modules with Voltage Feedback in MATLAB/Simulink," International
Conference on Image Processing, Production and Computer Science
(ICIPCS'2016), London (UK), March 26-27, 2016, pp. 10-16.
Fig. 17. SOC readjusted at about 50%
[10] Jay Patel, Vishal sheth, Gaurang Sharma, "Design & Simulation of
Photovoltaic System Using Incremental MPPT Algorithm," International
Journal of Advanced Research in Electrical, Electronics and
Instrumentation Engineering, Vol. 2, Issue 5, 2013, pp. 1647-1653.
CONCLUSION [11] Rahul Prasad, Suresh Gawre, Nikhil Kumar, "Modelling and Simulation
The work presented in this paper treats the feasibility of of Solar Photovoltaic System and Interfacing with Closed Loop Boost
Converter and Neutral Point Clamped Multilevel Inverter," Electrical
replacing the conventional street lighting system in Al-Byada and Electronics Engineering International Journal (ELELIJ), Vol 4, No
city with a system based LED lamps or stand-alone PV street 4, 2015, pp. 77-97.
lighting system. Here, we can conclude that: [12] Rakesh R., Kannan S. A., Jomy Joy, Kamala Devi V., Jayaraju M.,
"Modelling and Analysis of MPPT Techniques for Grid Connected PV
• The grid powered lighting system using LED saves System," International Journal of Innovative Research in Electrical,
about 75% of the energy consumption that is needed by Electronics, Instrumentation and Control Engineering, Vol. 2, Issue 2,
the lighting system using HPS, while PV street lights 2014, pp.1031-1037.
system saves 100% because it is an independent system. [13] Sangita R. Nandurkar, Mini Rajeev, "Modeling Simulation & Design of
Photovoltaic Array with MPPT Control Techniques," International
• The grid lighting system based LED lamps realizes an Journal of Applied Power Engineering (IJAPE), Vol. 3, No. 1, 2014, pp.
economy percentage of 66% from the budget allocated 41-50.
to the currently HSP lighting system over 20 years. [14] Kumari J. Surya, Babu Ch. Sai, "Mathematical Modeling and Simulation
of Photovoltaic Cell using Matlab-Simulink Environment," International
• Although, the initial cost of solar powered LED street Journal of Electrical and Computer Engineering (IJECE), Vol. 2, No. 1,
lighting system is higher than the initial cost of 2012, pp. 26-34.
conventional HSP system. However, it is an [15] Tarak Salmi, Mounir Bouzguenda, Adel Gastli, Ahmed Masmoudi,
inexpensive solution for long term compared to the HSP "MATLAB/Simulink Based Modelling of Solar Photovoltaic Cell,"
International Journal of Renewable Energy Research, Vol.2,No.2, 2012,
lamps system. The stand-alone PV system saves fuel pp. 213-218.
and money. It achieves an economic feasibility of 41% [16] Almoataz Y. Abdelaziz, Ahmed M. Atallah, Raihan S. Jumaah,
over 20 years compared to HSP system. "Simulation and Modeling of a Three Phase Two-Stage Grid Connected
Photovoltaic System," International Journal on Power Engineering and
• It is important that the batteries used for PV street Energy (IJPEE), Vol. 6 – No. 1, 2015, pp. 512-520.
lighting application have a minimum state of charge [17] Sanjeev Sharma, Sonia, Sushil Kumar, "Simulation Model of Boost
level of about 80%. Converter Used in Photovoltaic System," International Journal of
Advances in Electrical and Electronics Engineering (IJAEEE), Vol. 2,
No. 1, 2012, pp. 218-224.
[18] Shridhar Sholapur, Mohan K. R., Narsimhegowda T. R., "Boost
REFERENCES Converter Topology for PV System with Perturb And Observe MPPT
[1] M. Almakhtar, “Towards an Extensive Exploitation of Solar PV Algorith," IOSR Journal of Electrical and Electronics Engineering
Technology in Libya: Legislative, Economic and Environmental (IOSR-JEEE), Vol. 9, Issue 4, Ver. II, 2014, pp. 50-56.
Considerations,” Libyan Journal of Science & Technology 7:2, 2018, [19] Tremblay, O., Dessaint, L.-A, "Experimental Validation of a Battery
pp. 103-113. Dynamic Model for EV Applications," World Electric Vehicle Journal.
[2] A. Khalil, A. Asheibi, “The chanes and Challenges for Renewable Vol. 3 - ISSN 2032-6653, Norway, May 2009.
Energy in Libya,” 4th International Conference on Renewable Energy
Research and Applications. Palermo, Italy, Nov. 2015.