A fuzzy inference model for short-term load forecasting

Rustum Mamlook
a,
, Omar Badran
b
, Emad Abdulhadi
b
a
Middle East University for Graduate Studies, Faculty of Information Technology, Amman 11942, Jordan
b
Al-Balqa Applied University, Faculty of Engineering Technology, Amman, Jordan
a r t i c l e i n f o
Article history:
Received 20 March 2008
Accepted 30 October 2008
Available online 7 January 2009
Keywords:
Fuzzy sets
Power generation
Short-term load forecasting
a b s t r a c t
This paper is concerned with the short-term load forecasting (STLF) in power system operations. It
provides load prediction for generation scheduling and unit commitment decisions, and therefore
precise load forecasting plays an important role in reducing the generation cost and the spinning
reserve capacity. Short-term electricity demand forecasting (i.e., the prediction of hourly loads
(demand)) is one of the most important tools by which an electric utility/company plans, dispatches the
loading of generating units in order to meet system demand. The accuracy of the dispatching system,
which is derived from the accuracy of the forecasting algorithm used, will determine the economics of
the operation of the power system. The inaccuracy or large error in the forecast simply means that load
matching is not optimized and consequently the generation and transmission systems are not being
operated in an efcient manner. In the present study, a proposed methodology has been introduced to
decrease the forecasted error and the processing time by using fuzzy logic controller on an hourly base.
Therefore, it predicts the effect of different conditional parameters (i.e., weather, time, historical data,
and random disturbances) on load forecasting in terms of fuzzy sets during the generation process.
These parameters are chosen with respect to their priority and importance. The forecasted values
obtained by fuzzy method were compared with the conventionally forecasted ones. The results showed
that the STLF of the fuzzy implementation have more accuracy and better outcomes.
& 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Load forecasting is an important element for economically
efcient operation and for effective control of power systems. The
purpose of the short-term load forecasting (STLF) is to forecast in
advance the system load, that represented by the sum of all the
consumers load at the same time. Also, precise load forecasting is
required to avoid high generation cost and the spinning reserve
capacity. Under prediction of STLF leads to insufcient reserve
capacity preparation and threaten the systems stability, on the
other hand, over prediction of STLF leads to the unnecessarily
large reserve that leads to high cost preparation.
The following factors specify the pattern of the electricity
consumption variation: (a) weather; (b) time; (c) historical data;
and (D) random disturbances. The nature of parameters that affect
the load forecasting includes many uncertainties. Fuzzy logic (FL)
is characterized by generalizing classical two-valued logic for
reasoning under nonlinear and uncertain conditions; it is there-
fore the most appropriate method in describing the human
knowledge that contains vague concepts and huge amount of
data. Load forecasting using fuzzy implementation is considered
to be faster and more accurate than the conventional forecasting
methods which deal with rigid data and have long processing
time. Regression analysis is considered as a conventional way
in power demand prediction. In statistics, regression analysis is
the process used to estimate the parameter values of a function,
in which the function predicts the value of a response variable
in terms of the value of other variables (Tong, 2007); more-
over, the method that based on similarity principle which
use information of the day being similar to the weather condi-
tion is considered as a conventional method (Pandian et al.,
2006).
Conventional load-forecasting systems in Jordan are normally
based on statistical modeling techniques; they have limited
chances in predicting accurate loads for abnormal days when
some irregular events occur due to abnormal weather conditions
or sudden temperature changes. They also need precise values in
the forecasting process, while the fuzzy method uses interval of
values like high or low temperature. Such changeable conditions
lead to the application of fuzzy implementations to be used
instead of statistical schemes in the forecasting systems. And even
though in the US, and most developed countries, the standard
load-forecasting system uses a neural network. Fuzzy logic was
used by Pandian et al. (2006) to achieve much closer results to the
actual consumption than other conventional methods uses
the similarity principle. And Li-Chih and Pan (2007) found that
ARTICLE IN PRESS
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/enpol
Energy Policy
0301-4215/$ - see front matter & 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.enpol.2008.10.051

Corresponding author.
E-mail address: rstmamlk@hotmail.com (R. Mamlook).
Energy Policy 37 (2009) 12391248
the adaptive network-based fuzzy inference gave better results
than articial neural-network model.
Many attempts by researchers (Jingfei, 2006; Chowet al., 2005;
Garca, 2006; Gabbi and Zanotti, 2003; Negnevitsky, 2005) have
been made to improve load-forecasting process in many world-
wide regions. Khan and Abraham (2003) used a hybrid of articial
neural network (ANN) and fuzzy logic to forecast the load in Czech
Republic. They found that hybrid fuzzy neural network and radial
basis function networks are the best candidates for the analysis of
the load in Czech Republic.
Another study has been carried out for power systems in India
by Pandian et al. (2006). They used fuzzy approach for short-term
load forecasting using the temperature and time parameters only
as inputs for the system. They showed that with the aid of fuzzy
logic, they can achieve much closer results to the actual
consumption than the conventional methods which use the
similarity principle.
Mamlook (2006) used fuzzy set methodology for evaluating
energy production alternatives to compare between different
power systems. He showed that the fuzzy method has faster
learning technique than the neuro-fuzzy method. Fuzzy logic
methodology has been utilized for wide range of evaluation
applications, i.e., evaluation and comparison between different
solar systems (Mamlook et al., 2001), evaluation of factors
affecting solar still production (Mamlook and Badran, 2007),
evaluation of parameters that affect leakage in infrastructure
systems (Mamlook and Al-Jayyousi, 2003).
Li-Chih and Pan (2007) applied the adaptive network-based
fuzzy inference system model to forecast the regional electricity
loads in Taiwan, and demonstrated the forecasting performance of
their model. They found that the adaptive network-based fuzzy
inference gave better results than articial neural-network model.
An integrated evolving fuzzy neural network and simulated
annealing for load-forecasting method is presented by Liao and
Tsao (2004) to reduce the error of conventional load forecasting.
Their load-forecasting scheme was tested using data obtained
from a sample study includes 1 year, 1 month, and 24h time
periods. They showed that their method gave good accuracy.
Another study by Kodogiannis and Anagnostakis (1999)
discussed the development of improved neural-network-based
short-term electric load-forecasting models for the power system
of the Greek island of Crete. The performance was evaluated
through a simulation study, using metered data provided by the
Greek Public Power Corporation. Their results indicated that the
load-forecasting models developed provided more accurate fore-
casts than the conventional methods.
Maia and Gonc-alves (2008) proposed an approach for next day
peak load forecasting for electrical companies. A nonlinear model
for the peak load is proposed taking into account the historical
load and the temperature, each of which was estimated using an
on-line recursive algorithm.
Hobbs and Nelson (1992) used a bilevel programming in the
electric utility industry. The model is nonlinear and is used to
analyze various economic issues that affect electric utility
planning. The electric utility at the upper level of the model
seeks to minimize costs or maximize benets while controlling
electric rates and subsidizing energy conservation programs.
Customers at the lower level attempt to maximize their net
benet by consuming electricity and investing in conservation.
Gribik et al. (2007) suggested that the electricity market
models require energy prices for balancing, spot and short-term
forward transactions. For the simplest version of the core
economic dispatch problem, the formulation produces a well-
dened solution to the pricing problem in the usual intersection
of the supply marginal cost and the demand bids. This pricing
supports the equilibrium solution and satises a no arbitrage
condition. In the more general economic unit commitment and
dispatch models, there may be no corresponding uniform energy
price vector that supports the solution. This introduces a need
both dene the appropriate energy prices and determine the
ARTICLE IN PRESS
Fig. 1. Typical consumption curve pattern for a normal day.
Fig. 2. Monthly variation of electricity consumption in year 2006.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1240
associated uplift makewhole payments needed to support the
solution. Different approximations of the optimal value function
yield different price and uplift results.
The manual forecasting previously performed by the system
dispatchers has been replaced by fuzzy logic STLF model
incorporated with meteorological effects to improve load predic-
tions of the power systems in Jordan.
The input fuzzy sets have been modeled to represent the
temperature, historical data for different days, the forecasted
weather and the day duty Holiday or normal workday, while the
fuzzy output represents forecasted load. The proposed fuzzy
STLF methodology was applied to a case study of the National
Electric Power Company (NEPCO) in Jordan that based on actual
hourly load data for the last 7 years (January 2000June 2007).
Thus, efforts aimed at the implementation of accurate and
effective STFL are highly worthwhile.
2. Load attribute in Jordan
Good understanding of the system load characteristics helps in
designing a reasonable forecasting model to achieve accurate
results at variable conditions.
The system load represents the sum of all the consumers load
at the same time (Jingfei, 2006; Chow et al., 2005).
A typical load curve pattern in Jordan for normal day of 24h is
shown in Fig. 1.
Fig. 2 shows that the highest consumption rates occurred in
the summer season, when the energy demand for cooling and air
conditioning is high (Garca, 2006), while the high consumption in
winter is due to the usage of heating systems (NEPCO, 2006).
The load pattern curve undergoes different changes due to
unstable conditional factors; these factors that should be
considered during the load-forecasting process (Jingfei, 2006)
are as follows:
weather,
time,
economy, and
random disturbance.
Each parameter will be explained to show its inuence on the
consumption variation:
2.1. Weather
Weather factors include temperature, humidity, wind speed,
cloud cover, light intensity, etc. The Jordanian National Electric
Power company (NEPCO, 2006) use limited weather inputs; they
use temperature at the peak load only, while the fuzzy logic
considers the temperature at each hour to track the accurate load
variation to achieve better output.
Other factors like humidity and wind speed should be included
in the forecasting process to achieve high accuracy of prediction
(Khan and Abraham, 2003; Gabbi and Zanotti, 2003).
The pattern of the load consumption in Jordan that varies due
to the change in temperature is illustrated in Fig. 3a and b. Fig. 3a
and b shows the behavior of the power consumption variation
during morning and evening peak load at low, normal, and high
temperatures. As can be inferred from Fig. 3a and b, the higher
Megawatt (MW) consumption takes place at low and high
temperatures, while it starts to decrease when the temperature
gets near to the normal condition. In other words, the quantity of
energy required for heating or cooling is strongly dependent on
the weather condition; below normal temperatures in winter
create higher demand for heating; above normal temperatures in
summer create higher demand for energy to meet air conditioning
needs (Khan and Abraham, 2003; Gabbi and Zanotti, 2003).
2.2. Time
Time plays an important role in inuencing the load con-
sumption, which includes the time during the day or at night as
shown in Fig. 4. The consumption will consider whether it is a
normal workday or a holiday, and the season. It can be seen from
Fig. 4 that the consumption rate increased during the mid-day
ARTICLE IN PRESS
3%
2%
15%
16%
29%
35% Household
Industry
Commercial
Water pumping
Street lights
Others
Fig. 4. Sectorial distribution of electricity consumption for the year 2006 (NEPCO,
2006).
Fig. 3. Consumption variation versus temperature (20002007): (a) Temperature
at mid-day peak load. (b) Temperature at evening peak load.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1241
which corresponds to the working hours in the industrial, private,
and governmental sectors; therefore in the evening, the con-
sumption rate will be increasing, specially in the night time due to
the high usage of lights and household utilities which represent
335% of the total sectorial consumption (Fig. 4).
2.3. Economy
The economic situation also affects the utilization of electri-
city. The degree of industrialization is an explicit example
categorized under the economy factor. Since the industrial
sector consumption in Jordan represents 29% of the total
electricity consumption as shown in Fig. 5, it is important to
consider the production/consumption layout of the main factories
in Jordan, during the year, to improve the load forecasting and to
get more accurate results.
2.4. Random disturbances
Random disturbances such as wars and conicts will lead to
the increase in the number of electricity consumers due to
immigrations and the use of frequent emergency services. In spite
of the prevailing circumstances, Jordans stability and security
forms a good environment for investment, consequently, the rate
of growth of investments during 20022006 has a steep growth
rate as shown in Fig. 5 (Jordan Investment Board, 2006). Such
steep growth rate has increased the electricity consumption
in Jordan.
3. Fuzzy sets proposed methodology
Over the past two decades, there has been a tremendous
growth in the use of fuzzy logic controllers in power systems
applications. Recent series of tutorials in the IEE Power Engineer-
ing focused entirely on the applications of fuzzy logic in power
systems can be an evidence of the importance of the fuzzy models
in the energy eld (Cirstea et al., 2002).
Since precise load forecasting remains a great challenge, the
main objective of the fuzzy methodology in the current study is to
develop a practical model that can achieve an accurate forecasting
result. Fuzzy logic has the ability to absorb the human experience,
contain a large amount of data and infer the desired actions
represented with an accurate forecasting.
The advantages of using fuzzy logic in load-forecasting
applications include the following:
Fuzzy method uses fuzzy sets that enabled us to condense
large amount of data into smaller set of variable rule.
Fuzzy logic controllers are based on heuristics and therefore
able to incorporate human intuition and experience (Cirstea
et al., 2002).
ARTICLE IN PRESS
240
200
160
120
80
40
0
-40
Foreign Local
year
-35.9%
4.7%
14.3%
80.9%
47.2%
188.5%
205.1%
108.2%
2005 - 2006 2004- 2005 2003 - 2004 2002 - 2003
[
%
]
Fig. 5. Percentage growth of local and foreign investment (Jordan Investment Board, 2006).
Fig. 6. Fuzzy input/output combination.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1242
The fuzzy inputs used are listed as following; they showed good
response in tracking the actual load consumption:
Last day consumption (MWh).
Last week consumption (MWh).
Last day temperature (1C).
Forecasted temperature (1C).
Weather.
Day (binary sets determine whether it is a normal work day or
holiday).
The classical four steps for fuzzy implementation are shown
below:
3.1. Determining the linguistic variables and the fuzzy sets
The fuzzy input/output combination is shown in Fig. 6. Each
input and output is divided into number of fuzzy sets as shown
in Figs. 711.
The output is divided into seven fuzzy sets; it shows the degree
of consumption for the forecasted load with respect to the inputs
situation, the fuzzy output sets are as follows:
Very very low consumption (VVL).
Very low consumption (VL).
Low consumption (L).
ARTICLE IN PRESS
Fig. 8. Membership functions for the historical data input in MW.
Fig. 9. Membership functions for temperature inputs in degree Celsius (1C).
Fig. 10. Membership function for weather input.
Fig. 11. Membership function for Day input.
Fig. 7. Membership functions for the forecasted load in Megawatt (MW).
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1243
Normal consumption (N).
High consumption (H).
Very high consumption (VH).
Very very high consumption (VVH).
The output consumption intervals were divided into Gaussian
sets as shown in Fig. 7.
The values of the output consumption shown in Fig. 7 are
within the range from 900 to 2000MW which represents the
minimum and the maximum consumption limit, respectively.
The intervals for the Last day consumption and the Last
week consumption inputs have been divided into seven member-
ship functions as follows:
Very very low consumption (VVL).
Very low consumption (VL).
Low consumption (L).
Normal consumption (N).
High consumption (H).
Very high consumption (VH).
Very very high consumption (VVH).
The membership functions for the historical data input are
shown in Fig. 8 as follows:
Last day consumption and Last week consumption inputs
are included in the fuzzy implementation to determine the
behavior of the consumption during the last day and the last
week, respectively.
The intervals for the (last day temperature) and (forecasted
temperature) have been divided into three membership functions
(Fig. 9):
Low temperature (L).
Normal temperature (N).
High temperature (H).
The average temperature values in Fig. 9 are occurring between
0 and 401C which represent the low and the high average
temperatures, respectively.
The weather interval inputs have been divided into four
membership functions (snow, rain, cloud, and clear) and shown
in Fig. 10.
The weather inputs are used in the fuzzy implementation to
determine which weather condition is dominant (i.e., snow, rain,
cloud, or clear) as shown in Fig. 10.
(Day) is an input with binary boundaries which determines if
the day is a normal workday or holiday; it is illustrated in Fig. 11.
The Day input is used to determine the day duty (holiday or
workday) within 01 range.
3.2. Constructing fuzzy rules
In the current study, 37 fuzzy rules have been used to predict
the load under the effect of different inputs shown in Fig. 6. The
rules processed are shown in Fig. 12.
3.3. Performing fuzzy inference into the system
The fuzzy inference is the process of formulating the mapping
from a given input to output using fuzzy logic. The mapping
then provides a basis from which decisions can be made or
patterns discerned. The process of fuzzy inference involves
membership functions, fuzzy logic operators, and if-then rules
(The MathWorks, 19842007).
This procedure is used to compute the mapping from the input
values to the output values, and it consists of three sub-processes,
fuzzication, aggregation, and defuzzication (Fig. 13).
For example, rule 6 shown in Fig. 13 shows that the Last day
consumption was 1650MW and day equal to 1 is a Workday,
the degree of activation for the output set at rule six carries the
ARTICLE IN PRESS
Fig. 12. Forecasting process using fuzzy logic.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1244
same minimum input set Last day consumption degree because
the operator is and. The activated sets due to fuzzication sub-
process will be aggregated in the next step to form the combined
shape shown in Fig. 13, after that, it will be defuzzied to get a
crisp number (forecasted load 1680MW); the rules in the
MATLAB toolbox are shown in Fig. 14.
If the (day) is holiday, then the (last day consumption) will be
ignored and Last week consumption value should be considered
as a last holiday.
4. Case study
The data for this case study have been taken from the
National Electric Power Company in Jordan. NEPCOs capa-
city portfolio includes fuel oil-red steam generating units, gas-
red combustion turbines, diesel-red combustion turbines,
and diesel engines. NEPCO used to provide all the bulk power
to the national grid, except for that supplied by interconnected
industrial companies (about 100MW). NEPCO also operates
ARTICLE IN PRESS
Fig. 14. Fuzzy rules.
Fig. 13. Fuzzy implementation sequence.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1245
the bulk power network in Jordan; it operates the main
substations to provide services to the various distribution
networks, which have an aggregate capacity of about 2101MW
(NEPCO, 2006).
The rules in the fuzzy logic were made based on the historical
system load data sheet of the years (20002007). The fuzzy sets
were tuned to get the right response based on the variation of
different parameters, the following Tables 14, compare between
the conventional forecasting and the fuzzy forecasting. The mean
absolute percentage error (MAPE) is used to measure the error for
both methods.
The error caused by the fuzzy method is due to the varied
nature of the load consumption and due to the limited input data
that have been discussed earlier. Whenever new input data are
ARTICLE IN PRESS
Table 1
Comparison between conventional and fuzzy forecasted load for 24h 23/5/2007.
Time Conventional
load (MW)
Fuzzy forecasted
load (MW)
Actual
load (MW)
Conventional
error (%)
a
Fuzzy
error (%)
23/5/2007 Wednesday
1 1178 1180 1176 0.2 0.3
2 1130 1160 1129 0.1 2.7
3 1083 1110 1095 1.1 1.4
4 1070 1120 1098 2.6 2
5 1105 1120 1093 1.1 2.5
6 1085 1100 1080 0.5 1.9
7 1152 1200 1195 3.6 0.4
8 1281 1340 1327 3.5 1
9 1424 1480 1509 5.6 1.9
10 1505 1530 1567 4.0 2.4
11 1566 1600 1614 3.0 0.9
12 1635 1620 1640 0.3 1.2
13 1585 1590 1610 1.6 1.2
14 1588 1610 1600 0.8 0.6
15 1586 1590 1591 0.3 0.1
16 1580 1570 1547 2.1 1.5
17 1485 1510 1528 2.8 1.2
18 1483 1470 1482 0.1 0.8
19 1451 1450 1418 2.3 2.3
20 1670 1670 1700 1.8 1.8
21 1604 1620 1633 1.8 0.8
22 1490 1510 1515 1.7 0.3
23 1326 1410 1417 6.4 0.5
24 1310 1270 1293 1.3 1.8
Average error 2 1.3
a
MAPE |((conventional or fuzzy load)(actual load))/(actual load)|.
Table 2
Comparison between conventional and fuzzy forecasted load for 24h in 15/1/2007.
Time Conventional
load (MW)
Fuzzy forecasted
load (MW)
Actual
load (MW)
Conventional
error (%)
Fuzzy
error (%)
15/1/2007 Monday
1 1220 1190 1149 6.2 3.6
2 1120 1130 1082 3.5 4.4
3 1075 1090 1060 1.4 2.8
4 1050 1050 1046 0.4 0.4
5 1121 1120 1128 0.6 0.7
6 1196 1190 1164 2.7 2.2
7 1345 1290 1192 12.8 8.2
8 1367 1390 1373 0.4 1.2
9 1502 1490 1481 1.4 0.6
10 1575 1590 1581 0.4 0.6
11 1589 1580 1601 0.7 1.3
12 1670 1620 1610 3.7 0.6
13 1655 1580 1590 4.1 0.6
14 1686 1590 1575 7.0 1
15 1652 1570 1547 6.8 1.5
16 1699 1620 1501 13.2 7.9
17 1806 1710 1620 11.5 5.6
18 1860 1790 1800 3.3 0.6
19 1798 1740 1723 4.4 1
20 1761 1700 1675 5.1 1.5
21 1770 1680 1667 6.2 0.8
22 1636 1580 1563 4.7 1.1
23 1542 1490 1448 6.5 2.9
24 1333 1310 1298 2.7 0.9
Average error 4.6 2.2
Table 3
Comparison between conventional and fuzzy forecasted load for 24h in 18/1/2007.
Time Conventional
load (MW)
Fuzzy forecasted
load (MW)
Actual
load (MW)
Conventional
error (%)
Fuzzy
error (%)
18/1/2007 Thursday
1 1101 1170 1191 7.6 1.8
2 1033 1130 1108 6.8 2
3 966 1070 1040 7.1 2.9
4 964 1050 1053 8.5 0.3
5 945 1090 1065 11.3 2.3
6 1025 1160 1164 11.9 0.3
7 993 1260 1209 17.9 4.2
8 1103 1370 1335 17.4 2.6
9 1272 1520 1513 15.9 0.5
10 1411 1550 1589 11.2 2.5
11 1527 1590 1352 12.9 17.6
12 1565 1600 1512 3.5 5.8
13 1580 1570 1492 5.9 5.2
14 1544 1540 1483 4.1 3.8
15 1555 1540 1445 7.6 6.6
16 1630 1530 1460 11.6 4.8
17 1720 1630 1550 11.0 5.2
18 1770 1760 1730 2.3 1.7
19 1660 1680 1660 0.0 1.2
20 1620 1660 1620 0.0 2.5
21 1575 1620 1610 2.2 0.6
22 1510 1550 1545 2.3 0.3
23 1422 1430 1415 0.5 1.15
24 1342 1300 1318 1.8 1.4
Average error 7.6 3.2
Table 4
Comparison between conventional and fuzzy forecasted load for 24h in 29/6/2007.
Time Conventional
load (MW)
Fuzzy forecasted
load (MW)
Actual
load (MW)
Conventional
error (%)
Fuzzy
error (%)
29/6/2006 Thursday
1 1250 1270 1285 3 1.2
2 1160 1190 1210 4 1.7
3 1120 1140 1170 4 2.6
4 1110 1130 1130 2 0
5 1080 1130 1145 6 1.3
6 1035 1060 1080 4 1.9
7 1110 1120 1140 3 1.8
8 1277 1330 1360 6 2.2
9 1447 1460 1485 3 1.7
10 1514 1560 1548 2 0.8
11 1569 1620 1612 3 0.5
12 1600 1620 1640 2 1.2
13 1595 1610 1631 2 1.3
14 1583 1620 1600 1 1.3
15 1548 1590 1605 4 0.9
16 1518 1560 1565 3 0.3
17 1470 1510 1486 1 1.6
18 1425 1470 1420 0.3 3.5
19 1374 1400 1380 0.4 1.4
20 1480 1520 1535 4 1
21 1570 1600 1615 3 0.9
22 1470 1520 1520 3 0
23 1445 1470 1475 2 0.3
24 1334 1370 1370 3 0
Average error 2.8 1.2
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1246
ARTICLE IN PRESS
Fig. 15. Comparison between conventional forecasting load and fuzzy forecasting load in (a) 23/5/2007, (b) 15/1/2008, (c) 18/1/2007, and (d) 29/6/2006.
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1247
existed, the mission for the fuzzy method will be facilitated to
predict the load variation in more accurate way.
During the power plant visits, we noticed that the conventional
method of forecasting uses limited data inputs for the load-
forecasting process; so that it is highly recommended to consider
other parameters like the temperature level at each hour,
humidity, wind speed, temperature, and changes in production
or consumption rate. So that fuzzy set implementation incorpo-
rate the weather database is found to be highly capable to track
the load variation in more accurate way. Hence, the conventional
forecasting previously performed by the system dispatchers has
been replaced by the fuzzy STLF software as a modern energy
management system.
Fig. 15ad shows small deviation (o5%) between the actual
load and the fuzzy set evaluation that is due to the limitedness of
the dispatchers data collected by NEPCO and due to the sudden
variations occur in the load consumption rate under the effect of
different climatical conditions. But as an overall view, the fuzzy
forecasting shows similar trend to the conventional method with
more accurate predictions, because it takes into account the real
variations of weather and rate of consumptions uctuations. Also,
Fig. 15ad of the fuzzy method forecasted accurately the rate of
the power consumption during morning and evening peak load.
5. Conclusions
Load forecasting is the most important factor for estimating
the power needed for consumers; it provides load prediction for
generation scheduling and unit commitment decisions, and
therefore precise load forecasting plays an important role in
reducing the generation cost.
Fuzzy logic has the ability to absorb the human experience,
contain a large amount of data and infer the desired actions
represented with an accurate forecasting. Fuzzy logic controllers
are based on heuristics and therefore able to incorporate human
intuition and experience. The advantages of using fuzzy logic for
load-forecasting applications are it condenses large amount of
data (Tables 14) into smaller set of variable rules, it contains the
operators experience, and it infers the correct actions from
historical data.
Load forecasting using fuzzy implementation is faster and
more accurate than the conventional forecasting methods that
deals with rigid data and have long processing time. Conventional
load-forecasting systems in Jordan are normally based on
statistical modeling techniques. They have limited chances in
predicting the loads for abnormal days when some irregular
events occur, at abnormal weather or changes in temperature, and
the sudden changes in the consumption rate. Hence, the
incorporation of meteorological effects into STLF model will
undoubtedly provide improved load predictions.
Forecasts using fuzzy logic consistently resulted in a forecast-
ing error that was 5% less than conventional statistical forecasting
methods. It can be concluded that the fuzzy forecasted values
provided by fuzzy implementation have more accuracy and better
outcomes than conventionally forecasted results. Also, the present
fuzzy STLF model can save the utility signicant sums of money
by reducing the error in load predictions.
References
Chow, H., Wu, F., Momoh, J., 2005. Applied Mathematics for Restructured Electric
Power Systems. Springer, England.
Cirstea, M.N., Dinu, A., Khor, J.G., McCormick, M., 2002. Neural and Fuzzy Logic
Control of Drives and Power Systems. Newnes, England.
Gabbi, G., Zanotti, G., 2003. Climate variables and weather derivatives: gas
demand, temperature and seasonality effects in the Italian case. Social Science
Research Network.
Garca, S.V., 2006. Energy efciency and energy policy evaluation in the Spanish
hotel sector. Master Thesis, Utrecht University, Utrecht, The Netherlands.
Gribik, P.R., Hogan, W.W., Popeii, S.L., 2007. Market-clearing electricity prices and
energy uplift. Harvard Report, USA.
Hobbs, B.F., Nelson, S.K., 1992. A nonlinear bilevel model for analysis of electric
utility demand-side planning issues. Annals of Operations Research 34,
255274.
Jingfei, Y., 2006. Power system short term load forecasting. Master Thesis,
Technische Universita t, Darmstadt University.
Jordan Investment Board, 2006. Annual Report.
Khan, M.R., Abraham, A., 2003. Short term load forecasting models in Czech
Republic using soft computing paradigms. International Journal of Knowledge-
Based Intelligent Engineering Systems 7 (4), 172179.
Kodogiannis, V.S., Anagnostakis, E.M., 1999. A study of advanced learning
algorithms for short-term load forecasting. Engineering Applications of
Articial Intelligence Journal 44 (9), 159173.
Li-Chih, Y., Pan, M.C., 2007. Using adaptive network based fuzzy inference system
to forecast regional electricity loads. Energy Conversion and Management 49
(2), 205211.
Liao, G.C., Tsao, T.P., 2004. Application of fuzzy neural networks and articial
intelligence for load forecasting. Electric Power Systems Research 70 (3),
237244.
Maia, C.A., Gonc-alves, M.M., 2008. Application of switched adaptive system to load
forecasting. Electric Power Systems Research 78 (4), 721727.
Mamlook, R., 2006. Fuzzy set methodology for evaluating alternatives to compare
between different power production systems. Journal of Applied Science 6 (9),
21172125.
Mamlook, R., Al-Jayyousi, O., 2003. Fuzzy sets analysis for leak detection in
infrastructure systems: a proposed methodology. Journal of Clean Technology
Environmental Policy 6 (1), 2631.
Mamlook, R., Badran, O., 2007. Fuzzy sets implementation for the evaluation of
factors affecting solar still production. Desalination 203, 394402.
Mamlook, R., Akash, B., Nijmeh, S., 2001. Fuzzy sets programming to perform
evaluation of solar systems in Jordan. Energy Conversion and Management 42
(14), 17171726.
Negnevitsky, M., 2005. Articial Intelligence. Pearson Education, England.
NEPCO, 2006. Annual Report. Amman, Jordan.
Pandian, S., Duraiswamy, K., Rajan, C., Kanagaraj, N., 2006. Fuzzy approach for
short term load forecasting. Electric Power Systems Research 67 (67),
541548.
The MathWorks. Fuzzy Logic Toolbox. Fuzzy Inference System. The MathWorks,
Natick, MA, 19842007. See also /http://www.mathworks.com/access/help-
desk/help/toolbox/fuzzyS.
Tong, F., 2007. Capacity demand and climate in Ekero development of tool to
predict capacity demand under uncertainty of climate effects. Master Thesis,
Stockholm, Sweden.
ARTICLE IN PRESS
R. Mamlook et al. / Energy Policy 37 (2009) 12391248 1248