Journal of Software Engineering and Applications, 2012, 5, 46-49
doi:10.4236/jsea.2012.512b010 Published Online December 2012 (
Copyright © 2012 SciRes. JSEA
Peak Load Modeling for Kingdom of Bahrain
Isa S Qam ber
Deanship of Scientific Research, University of Bahrain, P. O. Box 32038, Isa Town, Kingdom of Bahrain.
Received 2012
Deriving some models to estimate the electrical demand for future for the Kingdom of Bahrain is carried out in the
present study. The ambient temperature is taken into the account as well as the time factor (Year). The model was de-
veloped in away describing the electric power demand during a summer period. The estimated values of the maximum
electrical load is obtained and evaluated on actual peak load data of the Kingdom of Bahrain.
Keywords: Peak Load; Curve Fitting; Maximum Annual Load
1. Introduction
Electricity systems models are software tools used to
manage electricity demand and the electricity systems
and for generation e xpansion planning purposes. Various
portfolios and scenarios mentioned later in the literature
review are modeled in order to compare the effects of
decision making in policy and on business development
plans in electricity systems. The modeling techniques
developed to study vertically integrated state monopolies
are now applied in liberalized markets where the issues
and constraints are more complex. This pa p er r eviews t he
modeling response to key developments. The paper pro-
vides an overview of electricity systems modeling King-
dom of Bahrain.
2. Literatu re Revi ew
Through the literature review there are several articles
discussing load modeling by using different techniques.
Linear regression models have been widely used.
Annual maximum power load models are software
tools used to manage electrical load demand. These mod-
els are used to help the engineers and managers for gen-
eration expansion planning purposes. A number of
scenarios are tested and modeled in order to be compared
and tested to reach a suitable decision.
Peik-Herfeh etl al in their article [1] which will be
published in the year 1213 present the improvements in
renewable energy technologies which are used in the re-
sources. A number of factors are considered where por-
tions of electrical demand have been met. A virtual pow-
er plane is decentralized energy management system
tasked to aggregate the capacity of some Distributed
Generations, storage facilities, and Dispatchable for the
purpose of energy trading and/or providing system sup-
port services. Due to the stochastic behavior of the prime
sources of some Distributed Generations, such as wind
speed and temperature, the steady state analysis of the
systems with integration of such Distributed Generation
units requires a probabilistic approach.
Orlando et al in their paper [2] studied the electric
energy prices in Brazil. The electric energy prices are
higher during 3 h of peak consumption on working days.
However, consumers could use a generator system for
partially reducing the electricity costs from utility com-
pany during peak hours. T he pr oce s s of sizing a gene rator
system begins developing the so called load capacity
curve. The acquired building electricity consu mption data
was used. This authors in their paper developed two me-
thodologies to estimate the curve for every month of the
year. The first method use the frequency of occurrence of
the measured values of the load power as supplied b y the
local utility monthl y bills. Fina lly a generator s ystem was
sized, minimizing the e lec tric al ener gy cost supplie d b y it,
when the generator meets partially the building load.
Kaldellis et al i n their paper [3] considering on the ba-
sis of probability distribution of the load demand of a
representative Greek island and the corresponding data
related to the available wind potential. In their study es-
timates the maximum - acceptable by the local grid -
wind energy contribution. For that reason, an integrated
computational algorithm has been developed from first
principles, based on a stochastic analysis. Based on the
results obtained, it becomes evident that with the current
wind turbine technology, wind energy cannot play a key
role in coping with the electrification problems encoun-
tered in many Greek island regions, excluding however
the case of introducing bulk energy storage systems that
Peak Load Modeling for Kingdom of Bahrain
Copyright © 2012 SciRes. JSEA
may provide considerable recovery of the remarkable
wind energy rejections expected.
Niknam et al [4] in their paper they used the green
energies and rising concerns about high cost of energy
from fossil fuels, renewable energy sources appears to be
a promising approach for producing local, clean, and in-
exhaustible ener gy. T his motivate s the i mplementatio n of
microgrids introduced as a cluster of electrical and/or
thermal loads and different, renewable energy sources.
Due to different uncertainties linked to electricity supply
in renewable microgrids, probabilistic energy manage-
ment techniques are going to be necessary to analyze the
system. The authors in their paper proposes a probabilis-
tic approach for the energy and operation management of
renewable Microgrids under uncertain environment. The
authors in their paper considered uncertainties in load
demand, market prices and the available electrical power
of wind farms and photovoltaic systems.
Tawalbeh et al [5] in their paper presented a nonlinear
approach to estimate the consumed energy in distribution
feeders. The proposed method uses the statistical solution
algorithm to analyze the activ e energy monthly consump-
tion, which enables one to estimate the energy consump-
tion during any period of the year. The energy readings
and the normalized accumulated energy profile are used
to estimate the hourly consumed active power, whic h ca n
be used for future planning and sizing the equipment of
the electrical system. The effectiveness of the proposed
method is demonstrated by comparing the simulated re-
sults with that of real measured data.
Qader and Qamber in their paper [6] developed a mod-
el for load estimation of Kingdom of Bahrain. The calcu-
lation method involves a Monte Carlo technique for the
simulation of the load. The model enables the predication
of the load against the time during years, where each year
is divided into 52 weeks. The forecasting model, com-
putes minimum mean square error (MMSE) forecasts of
the conditional mean of reserve power and conditional
stand ard de viation o f the i nnovati ons in e ach per iod ove r
a user specified forecast possibility. To do this, it views
the conditional mean and variance models from a linear
filtering perspective, and applies iterated conditional ex-
pectations to the recursive equations, one forecast period
at a time. The results are obtained and discussed.
Three scenarios have been carried out by Qamber [7]
to calculate the predicted maximum annual load for the
kingdom of Bahrain with the objective of formulating an
expa nsion plan fo r a f uture generatin g sys tem. The results
of the three scenarios were obtained and compared using
a comprehensive analysis. The maximum annual load
was calculated at average rates of 6.79% in the more
reasonable scenarios using the MATLAB package fol-
lowing the curve -fitti ng pol ynomial tec hnique.
3. Results and Discussions
It is well known that the electrical power demand is
linked to several weather variables, mainly the air tem-
perature. The present work concentrates on the effect of
the annual maximum temperature plus the population to
the electric load demand in Bahrain. A number of models
are derived which allowed a better characterization of the
observed modeling relationships. The models which are
designed to forecast the load’s behavior should be able to
show the effect of the considered factors. Temperature
has a gre a t e f fec t o n the hour ly ene rg y c on su mp ti on .This
effect first of all causes the annual maximum change in
the consumption load curve in a way that the curve is
higher for a hot summer day than a colder day and
For electric power load forecasting, the accurate mod-
els are essential to operation and planning of utility. Fur-
thermore, a particular network is possible to predict the
next year maximum load. It is well known that it is im-
possible to pre dict the next ye ar maximu m load with high
accuracy as the next day load. The next day load can be
predicted with an accuracy of 1-3%.
Table 1 reproduced from the previous study carried
out by Qamber [7], the only difference that the popula-
tion of Kingdom of Bahrain is added to the Table.
In Figure 1 the relation between the maximum annual
actual load versus years in Bahrain, as it shows that the
load increased in the years 2009, 2010 and 2011. The
load decreased in the year 2011, because the maximum
temperature during year 2011 was less than the previous
year increasing. The formula for such figure is as fol-
LoadAcP (χ) =132.5*x 263743 (1)
where: LoadAcP (χ) is the higher actual peak load (MW)
χ is the year
The operating maximum peak load for Kingdom of
Bahrain can be found by substituti ng the year into Equa-
tion (1) to find out the estimated maximum peak load for
the substituted value of the year. The data for the present
study are taken from a study done by Qamber [7].
Table 1. Hist or ical data f or king dom of bahrain.
Year Maximum
Temperature (Co) Actual Peak
Load (MW) Population
2009 45.7 2437 1178.415
2010 47.4 2708 1228.543
2011 45.9 2500 1224.571
2012 45.7 2948 1248.348
Figure 2 illustrates the variation combination of the
maximum temperature (˚C) and actual peak load (MW)
Peak Load Modeling for Kingdom of Bahrain
Copyright © 2012 SciRes. JSEA
versus the years from 2009 and 2012 through the histo-
Figure 3 shows the relation between the population in
Bahrain versus years. It shows that the population in-
creased year after year. The model obtained for such a
type of data is as follows:
Populatio n (χ) = 21582.7*x 42169549 (2)
where: Population (χ) is the Populatio n in Bahrain
χ is the year
Figure 4 illustrates the histogram of the maximum
temperature (˚C) versus years from 2009 and 2012.
Figure 1. Relationship bet ween Peak L oad and Ye ar.
Figure 2. The Histogram showing the combination of Maxi-
mum Temperature a nd actual Peak Load versus Years.
Figure 3. Relationship be tween Population a nd Year.
Figure 5 shows the relation between the actual peak
load and the population in Bahrain through the years
2009 and 2012. It shows that the population increased
year after year; where the annual load increased in three
years 2009, 2010 and 2012. The model obtained for such
a type of data is as follows:
(Population) = 0.0055953443*Population – 4191.8864
where: LoadAcP (Population) is the higher actual peak
load (MW)
Populatio n is the Population in B a hra in
Table 2 illustrates the variation of actual peak loads
(2009-2012) plus estimated values (2013-2015) versus
Figure 6 shows the relation between the actual peak
loads for the years (2009-2012) plus the estimated high-
est values annual maximum load for the years (2013-
2015) obtained by Qamber [7] versus years in Bahrain.
The model obtained for such a type of data is as follows:
LoadEHLoad (Year) = 198.07143*Year 395554.43 (4)
where: LoadEHLoad (Year) is the estimated highest
load (MW)
Year is the Year
Table 3 illustrates the relationship between Maximum
Loads versus Years.
Figure 4. The Histogram showing the Maximum Tempera-
ture versus Years.
Figure 6. The Relationship between the Load versus Ye ars.
Figure 7 shows the relation between the actual peak
maximum loads for the years (2009-2012) plus the esti-
Peak Load Modeling for Kingdom of Bahrain
Copyright © 2012 SciRes. JSEA
mated maximum values of the annual load for the years
(2013-1015) - obtained in the present study from the
graph represented by Figure 7 - versus years in Bahrain.
The model obtained for such a type of data is as follows:
LoadEMLoad (Year) = 132.5 * Year 263743 (5)
where: LoadEMLoad (Year) is the estimated maximum lo ad
(MW) Yea is the Year
In this p ap er the linear regression analysis of Bahrain’s
electrical load recognizes that the load pattern is heavily
dependent on ambient temperature, and finds a linear
function between the load and the temperature. The ma-
thematical model of the temperature dependency de-
scribes the increase in annual load when ambient temper-
ature changes.
Table 2. Peak Loa ds versus y ears.
Year Load (MW)
2009 2437
2010 2708
2011 2500
2012 2948
2013 3168
2014 3383
2015 3613
Table 3. Maximum Loads versus Years.
Year Load (MW)
2009 2437
2010 2708
2011 2500
2012 2948
2013 2979.5
2014 3112
2015 3244.5
Figure 7. Relationship of the loads v ers us years.
The curve fitting and modeling techniques are helping
and advising for the best solution and approach to reach
the suitable model. The present paper highlights on the
previous s tud ies with the software used in each s tudy plus
the models built in one of them and tested I the present
study to reac h the be st so lution.
4. Conclusions
In this paper, a simple and accurate models are discussed
earlier for annual maximum load and efficient algorithms.
The coefficients for each model are calculated. The input
parameters of the models are the historical maximum
annual loads, maximum ambient temperatures and popu-
lation in the Kingdom of Bahrain. In the prese nt stud y, it
is clear how much the weather conditions (ambient tem-
perature) influence the load. The models obtained in the
present study considering and developing an algorithm
for the annual peak loads.
5. Acknowledgement s
The author would like to express his thanks to the Uni-
versity of Bahrain for the preparation of the facilities to
make this research possible.
[1] M. Peik-Herfeh, H., Seifi and M.K. Sheikh-El-Esl a mi,
“Decision making of a virtual power plant under uncer-
tainties for bidding in a day-ahead market using point es-
timate method”, International Journal of Electrical Power
and Energy Syste m s , Vol. 44, No. 1, 2013, pp. 88-98.
[2] A. F. Orlando, M. P. Málaga, M.M. Huamani, “Metho-
dology for generating electric load profiles for sizing an
electric en erg y gen erati on syste m”, Energy and Buildings,
Vol. 52, 20 12, pp. 16 1-167.
[3] J. K. Kaldellis, M. Kapsali , and D. Tiligadas, “Presen ta-
tion of a stochastic model estimating the wind energy
contribution in remote island electrical networks”, Ap-
plied Energy, V ol. 97, 2012 , pp. 68-76.
[4] T. Niknam, F. Golestaneh and A. Malekpour, “Probabilis-
tic energy and operation managemen t of a microgrid con-
taining wind/photovoltaic/fuel cell generation and energy
storage devices based on point estimate method and
self-adaptive gravitational search algorithm”, Energy,
Vol. 43, N o. 1, 20 12, pp. 42 7-437.
[5] N. Tawalbeh and R. El-Khazali, “Peak load evaluation
based o n the accumulated annual energ y”, Proceedings of
the 16th IEEE Mediterranean Electrotechnical Confe-
rence, MELECON 2012;Yasmine Hammamet; 25-28
March 2012, Article number6196492, pp. 544-547.
[6] M.R. Qader and I. Qamber, “Enhancement, evaluation
and implementation of a load forecasting method”,
JAAUBAS, Vol . 11 , A pri l 2 01 2, pp. 1-8.
[7] I. Qamber, “Maximum Annual Load Estimation and
Network Strengthening for the Kingdom of Bahrain”,
AGJSR, Vol. 28, No. 4, 20 10, pp. 214 -223.