Electrical Power and Energy Systems 64 (2015) 340346

Contents lists available at ScienceDirect

Electrical Power and Energy Systems

journal homepage: www.elsevier.com/locate/ijepes

Short term load forecasting using wavelet transform combined with

HoltWinters and weighted nearest neighbor models
G. Sudheer , A. Suseelatha 1
Department of Mathematics, GVP College of Engineering for Women, Visakhapatnam 530048, Andhra Pradesh, India

a r t i c l e

i n f o

Article history:
Received 5 August 2013
Received in revised form 26 June 2014
Accepted 7 July 2014

Keywords:
Holt-Winters
Weighted nearest neighbor
Haar wavelet
Load forecasting
Hybrid model

a b s t r a c t
Short term load forecasting (STLF) is an integral part of power system operations as it is essential for
ensuring supply of electrical energy with minimum expenses. This paper proposes a hybrid method based
on wavelet transform, Triple Exponential Smoothing (TES) model and weighted nearest neighbor (WNN)
model for STLF. The original demand series is decomposed, thresholded and reconstructed into deterministic and uctuation series using Haar wavelet lters. The deterministic series that reects the slow
dynamics of load data is modeled using TES model while the uctuation series that reects the faster
dynamics is tted by WNN model. The forecasts of two subseries are composed to obtain the 24 h ahead
load forecast. The performance of the proposed model is evaluated by applying it to forecast the day
ahead load in the electricity markets of California and Spain. The results obtained demonstrate the forecast accuracy of the proposed technique.
2014 Elsevier Ltd. All rights reserved.

Introduction
The accurate forecast of electricity demand (load) for a range of
lead times is important for the effective management of power systems [1]. Short term load forecast (STLF) of power system loads
with lead times ranging from an hour to several days ahead has
a signicant impact on the operational efciency of electrical utilities [2]. The forecasts are needed for a variety of utility attributes
such as generation scheduling, the scheduling of fuel purchase,
maintenance scheduling, security analysis and energy transactions
[3]. This importance of load forecasts on decision making in energy
sector, calls for the development of accurate, fast and simple prediction algorithms. Unfortunately, the load demand is a nonstationary process inuenced by a multitude of factors ranging
from weather conditions over seasonal effects to socio-economic
factors and random effects, which makes load forecasts difcult
[4,5].
The existing models for load forecasting can be broadly
classied into conventional (statistical) approaches, articial intelligence based models and hybrid/combination models. The survey
papers [4,69] give an account of different techniques devoted to
electrical load analysis and forecasting. The conventional models
Corresponding author. Tel.: +91 0891 2739144, +91 08885043344.
E-mail addresses: g.sudheer@gvpcew.ac.in (G. Sudheer), suseelatha.a@gmail.
com (A. Suseelatha).
1
Tel.: +91 09866000118.
http://dx.doi.org/10.1016/j.ijepes.2014.07.043
0142-0615/ 2014 Elsevier Ltd. All rights reserved.

are linear and make certain assumptions regarding the characteristics of the load series. Among these models, the autoregressive
integrated moving average (ARIMA) model has excellent natural
statistical characteristics and is the most popular [10]. However,
these models are known to show some weakness in the presence
of special events and nonlinearities [11,12]. The low adaptability
of the statistical techniques to deal with nonlinear series has generated an increasing interest in using Articial intelligence based
techniques [13,14]. Among the AI based techniques, articial neural networks (ANN) have been successful applied to STLF [8]. However, a certain regularity of the data is an important precondition
for the successful application of Neural Networks (NN) [15] and
in addition they suffer from a number of weaknesses such as network construction problem, over tting issue, connection weight
estimation and the need for a large number of data for training
[16]. Moreover, these methods cannot capture well the rapid
changes in load [17]. Currently the emphasis is on hybrid models
that blend different technique to enhance the performance and
eliminate the limitations of existing individual models [18,19].
Both theoretical and empirical ndings in the literature show that
combining different methods can be an effective and efcient way
to capture the different patterns in the data and improve the forecasting performance [3,20].
Recently, hybrid models that integrate wavelet transforms with
other techniques [2,15,16,2125] have been extensively used for
short term load forecasting. In these studies, wavelet transforms
have been effectively employed to extract/analyze the characteristic

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

features embedded in the load time series at different frequencies.

In [15], two strategies for embedding the DWT into NN based STLF
model is proposed. The rst strategy consists of creating a model
using information from the original series and the wavelet domain
subseries. The second strategy involves predicting the loads future
behavior by independently forecasting each subseries in the wavelet
domain. In [22], a hybrid forecasting method composed of wavelet
transform, a two step correlation analysis, NN, evolutionary algorithm and cross validation is proposed for hourly load forecasting.
In [2], the hourly load was predicted via an adaptive NN structure
with the NN weights adjusted by the particle swarm optimization
algorithm. The wavelet processed data was fed as input for the adaptive NN structure. Some intelligent hybrid models utilizing only low
frequency components of load data is proposed in [23]. The proposed models include wavelet decomposition based radial basis
function model, wavelet decomposition based time series model
and wavelet decomposition based fuzzy inference NN model. These
studies bring out the importance of wavelet transform as a multiscale decomposition tool that can be used to capture/differentiate
the different characteristics of load data for easy integration with
other forecasting techniques.
A novel technique proposed recently [16] utilizes the wavelet
transform, ANN and ARIMA model to propose a hybrid method
for STLF. The technique is based on the idea that a time series
can be considered to be made up of two parts: a linear component
and a nonlinear component. The linear component is modeled
using the ARIMA model and the residuals of ARIMA model consisting of nonlinear components is decomposed via the wavelet transform into their details and approximation parts so that each of
them could be modeled by the approximate ANN. The forecast values of the linear and nonlinear components were composed to
obtain the forecast value of the load. A slightly similar approach
is presented in [3] where in the seasonal ARIMA (SARIMA) is used
to address the linear relationship while the back propagation neural network (BPNN) model is used to handle the nonlinear patterns
in the wavelet denoised load series. A variancecovariance
approach is then used in the model to combine the predicted values of SARIMA and BPNN.
In this work, a hybrid method is proposed for STLF by integrating Haar wavelet transform with Triple Exponential Smoothing
(TES) and weighted nearest neighbor (WNN) techniques. The basic
idea of the proposed hybrid method consists in assuming that the
load series can be considered to be made up of two parts: a deterministic component and a stochastic component. The deterministic
component that represents the general pattern of the load signal is
modeled using TES model while the WNN technique is used to nd
and weight similar data in the stochastic/uctuation component.
The TES model accounts for the trend and the seasonality of the
data and is optimal for a seasonal ARIMA model. The WNN model
is capable of accounting for both the nonlinearity and nonstationarity in the given data [26]. The Haar wavelet transform is used to
separate the faster dynamics (uctuation component) of load data
from the slowly varying characteristics (deterministic components) of data using suitable wavelet coefcient thresholds. During
thresholding, the wavelet transform will separate the faster changing data in the wavelet domain and inverse wavelet transform will
then retrieve the slowly varying characteristics of data with little
loss of detail. The maximum time scale level and threshold is
obtained by testing on the load series data. The feasibility of the
proposed hybrid method is evaluated using historical load data
from the California and Spain energy markets. The results evaluate
day ahead prediction.
The paper is organized as follows: section Theory deals
with some basic theoretical aspects of the wavelet transform,
triple exponential smoothing method and weighted nearest neighbor approach. In section Proposed methodology, the proposed

341

forecasting model is described. Numerical results concerning the

application of the proposed approach to the two electricity
markets are shown in section Results and discussion. Finally,
section conclusion presents the main conclusions of this work.
Theory
Discrete wavelet transform
The Discrete Wavelet Transform (DWT) whose main idea is the
process of multi-resolution analysis is one of the most appropriate
techniques to make a joint timefrequency analysis of discrete signals. The DWT [27] utilizes two set of functions /(t) and w(t), each
associated with the low pass and high pass lters respectively to
decompose the signal in terms of approximations and details.
The scaling function /(t) is associated with the low pass lters with
lter coefcients fhn; n 2 Zg and the wavelet function w(t) is
associated with the high-pass lters with lter coefcients
fgn; n 2 Zg. The DWT is usually computed using the pyramid
algorithm [28] and is realized by means of the lters h[k], g[k] that
are related to each other through

gk 1k1 hN  k  1 for k 0; 1; . . . ; N  1

where N is the length of the lter. These lters are constructed from
the wavelet kernel w(t) and the companion scaling function /(t)
through the relations

/t

pX
2 hk/2t  k

wt

pX
2 gk/2t  k

Using the wavelet lters, the data can be decomposed into a set
of low and high frequency components named approximations
aj[k] and details dj[k] respectively. They are given by

dj1 k

X
aj ng2k  n

aj1 k

X
aj nh2k  n

If xk a0;k k 1; 2; . . . ; l is the original series and if

aJ;k ; dj;k ; j 1; 2; . . . ; J; k 1; 2; . . . ; l are respectively the approximations and detail series obtained after J levels of decomposition

then xk aJ;k

J
X
dj;k k 1; 2; . . . ; l

j1

Here aJ;k k 1; 2; . . . ; l presents the tendency of the series and is

characterized by slow dynamics, while the details dj,k presents the
local details of time series and has fast dynamics [29,30]. The Haar
wavelet lter, which is a lter of length N = 2, g 0 g 1 p12 is used in
the present work to decompose the load signal into its deterministic
and uctuation components. The Haar wavelet coefcients within
each time scale are statistically independent and they can be
directly understood and described in simple terms of up and down
variations from one region to the next [31]. Moreover, Haar wavelet
is the only symmetric compactly supported orthonormal wavelet
[32].
Triple exponential smoothing
There are many forms of exponential smoothing methods of
which the HoltWinters family of exponential smoothing methods
is the one most commonly used [33]. Exponential smoothing is a
pragmatic approach to forecasting whereby the prediction is

342

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

constructed from an exponentially weighted average of past observation [34]. The method gains its popularity due to its simple
model formulation and good forecasting results [35]. An excellent
survey of the state of art in exponential smoothing models is
brought out in [36]. The application of this method to load forecasting problem can be found in [1,3739]. In particular, the triple
exponential smoothing (Holt-Winters) method, allows us to deal
with univariate time series that contain both trend and seasonal
factors. The HoltWinters exponential smoothing approach [40]
includes methods for both additive and multiplicative seasonal
patterns. The additive form of the TES method works with the following recursive updating equations:

Lt axt  Stc 1  aLt1 bt1

bt bLt  Lt1 1  bbt1

and St cxt  Lt 1  cStc

The h step ahead forecast Fn+h is

F nh Ln hbn Snhc

10

Here a, b and c are the smoothing parameters associated with level,

trend and seasonality respectively, while c is the length of the seasonal cycle. The parameters a, b and c are estimated by minimizing
the sum of squares of errors or any suitable error function. The
TES method is used here to model the deterministic component of
the demand series. The estimates of different seed values, one for
the level, one for the trend and seasonal cycle must be made. The
seed values adopted in the present paper is outlined in the
methodology.
Weighted nearest neighbors technique
Among the simplest and most intuitively appealing classes of
non-probabilistic classication procedures are those that weight
the evidence of nearby sample observations most heavily. The rst
formulation of a rule of the nearest neighbor type proposed in [41]
considers the k-nearest neighbor rule which assigns to an unclassied sample that class most heavily represented among its k nearest neighbors. A weighting function which varies with the distance
between the sample and the considered neighbor, in such, a manner that the value decreases with increasing sample to neighbor
distance is proposed in [42]. The weighted nearest neighbor
(WNN) technique [43] elaborates on these ideas and is outlined
in this section.
The observed hourly data is considered up to day d and the
prediction of 24 hourly load data corresponding to day d + 1 by
WNN technique [43] is elucidated in this section. Let Li e R24 be a
vector composed of the 24 hourly data corresponding to an
arbitrary day i i.e. Li = [l1, l2, . . ., l24]. The associated vector
LLi e R24m is the hourly data contained in a window of m consecutive days preceding day i LLi Lim1 ; Lim2 ; . . . ; Li1 ; Li where m is
a parameter to be determined. Using the Euclidean norm, the
distance for any couple of days i and j can be dened as

disti; j kLLi  LLj k

11

The k nearest neighbors of day d is rst determined using the

metric dist and based on closeness of the distance values obtained,
the neighbor set is formed as N = {q1, q2,. . ., qk}, q1 and qk are the
rst and kth neighbor in order of distance.
The prediction is given by

Ld1 P

1
i2N ai

X
i2N

ai Li1 where ai

distqk ; d  disti; d
distqk ; d  distq1 ; d

12

Proposed methodology
The discrete Haar wavelet transform is applied to the historical load data to decompose it into several other scales with different levels of resolution. Applying the wavelet lters on the
data at each scale transfers the noisy characteristics of the data
into a set of coefcients whose absolute values are smaller than
the rest of the coefcients [44]. When the Haar wavelet lter is
applied to the load signal, the components under different time
scale levels show different characteristics. Both the values and
variation characteristics of wavelet coefcients in the deterministic series and uctuation series are obviously different and an
accurate wavelet coefcient threshold will separate the two series [45]. Using wavelet thresholding, the uctuation coefcients
can be separated to recover the deterministic component of
the data. Thus utilizing the Haar DWT together with wavelet
de-noising technique the demand series is decomposed into
deterministic and uctuation components. The deterministic
component contains the trend and seasonality factors and is
therefore modeled using TES method. The uctuation component
that contains the rapid changing dynamics of data is modeled
using WNN method. The proposed forecast strategy is summarized in the following steps:
1. Decompose the historical load data (X) via the DWT into two,
three, four levels of decomposition. In the rst level decomposition, the Haar lters splits up the load series into a low
frequency component (approximation) and high frequency
component (details). The approximation represents a general pattern of the load signal where as the details can be
viewed as a difference between two successive approximations. In the second level of decomposition, these approximation coefcients are further decomposed into
approximation and detail coefcients. In this manner, the
Haar wavelet splits up the load signal into one approximation series and some detail series in the wavelet domain.
The best resolution level is chosen based on the approximation signals ability to describe the general pattern of the
load series.
2. The detail/wavelet coefcients are subjected to thresholding
which separates the variation below a certain value. Here
the wavelet coefcients are thresholded adaptively at each
level using Steins Unbiased Risk Estimation (SURE) method
[46,47].
3. After thresholding the coefcients, wavelet reconstruction of
the decomposed levels is carried out using the inverse wavelet transform to obtain the deterministic component (Xd) of
the data. The uctuation series (Xf) is then obtained using
the relation Xf = X  Xd.
4. The deterministic series Xd is now modeled using the TES
method through Eqs. (7)(10). The strategy for obtaining
the seed values to be used is the following:
i. Compute L24

24
1 X
xi
24 i1

13

where (x1, x2,. . ., x24) is the deterministic part in the hourly load of
the rst day of the test week. This value is taken to be the initial/
seed value L0 of the level term (Lt) in Eq. (7).
ii. The deterministic part (x1, x2,. . ., x48) in the rst two days of
the hourly load is used to initialize the trend term b0 in Eq.
(8). It is computed as

b0

1 hx25  x1 x26  x2
x48  x24 i

...
24
24
24
24

14

343

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

iii. To nd the hourly seasonal indices, one year of hourly load

data is considered and the initial seasonal indices S1,
S2,. . .S24 of the seasonal component in Eq. (9) are calculated
as follows:
Let lij denote the load at hour i on day j i = 1, 2, . . ., 24 and
P
j = 1, 2, . . ., 365. Calculate dav j 24
i1 lij the daily average load.
The hourly seasonal index Si is obtained as
365
1 X
lij
Si
365 j1 dav j

MAPE
i 1; 2; . . . ; 24

15

5. With the initial values L0, b0, S1, S2, . . .S24 determined, each of
the parameters a, b, c are varied in steps of 0.01 from 0 to 1
to determine the forecast values of the deterministic
component.
6. The uctuation series (Xf) is modeled using WNN. The
parameters window length (m) and number of neighbors
(k) are determined by sliding a window of length (m) along
the training data and minimizing the forecast error. For a
training data of two weeks, m is varied from 1 to 12 and k
from 2 to 14 m. For each pair of values m, k the forecast
values of the uctuation component is obtained.
7. For every possible combination of the parameters
a, b, c, m, k, the 24 h ahead load is forecast by composing
the forecast values of the deterministic and uctuation components. The best parameters are those resulting in the
smallest 24 h ahead forecasting errors. The minimum value
of the mean squared forecast error determines the optimum
values of the parameters.
8. With these optimum values and Eqs. (10) and (12), the 24 h
^ d and uctuation compoahead deterministic component X
^ f are determined. The next day load is obtained using
nent X
the relation

^X
^d X
^f :
X

The deterministic series and uctuation series obtained in the

wavelet preprocessing stage are forecast using TES and WNN technique respectively. The individual forecasts obtained are summed
up to obtain the 24 h ahead load. The performance of the model
is evaluated using the Mean Absolute Percentage Error (MAPE)
metric. The metric is computed as follows

16

Results and discussion

The hourly load data of California [48] and Spain [49] energy
markets are considered in this real world case study. The electrical
load for every hour of the year 2000 is considered for California
market and of the year 2002 for the Spain market. The zonal variations in the climatic conditions are wider in Spain than in California. Therefore, to illustrate the behavior of the proposed model,
results comprising four weeks corresponding to the four seasons
of the year 2002 are considered for Spain market while results
comprising three weeks corresponding to three seasons of the year
2000 are presented for the California market. In this manner, representative results for the whole year are proposed. The considered
weeks in the California market are January 17th to January 23rd in
winter, April 17th to April 23rd in spring and July 17th to July 23rd
in summer. In case of Spain market, the considered weeks are January 21st to January 27th in winter, April 15th to April 21st in
Spring, July 15th to July 21st in Summer and October 21st to October 27th in fall. The proposed forecast model for each of the considered weeks is built by considering the hourly load data of 14 days
(2 weeks) previous to the day of the week whose load is to be forecast. The proposed methodology is applied to the training data of
2 weeks and 24 h ahead load is forecast. The two week training
window is shifted one day ahead and the forecasts for the next
24 h are obtained. In this way, the forecast for the considered
weeks in both the markets are obtained.
The DWT is applied to the training data to decompose, threshold and reconstruct the data at different levels of resolution
using the Haar wavelet. In this paper, three levels of resolution
have been considered for both decomposition and reconstruction.

N
1X
jli  ^li j
 100
N i1 li

17

where li and ^li are the actual load and the forecast load at hour i
respectively and N is the number of predictions. The forecasts for
the test weeks are further compared on the two metrics MAPEWeek
and weekly error variance EVWeek [50,51]. The weekly MAPE is calculated as

MAPEWeek

168
1 X
jli  ^li j
168 i1 lWeek

18

P168
1
where lWeek 168
i1 li . The weekly error variance which is an index
of the uncertainty of a model is given by

EVWeek

168
1 X
jli  ^li j

 MAPEWeek
168 i1 lWeek

!2
19

The forecast errors (MAPE) for different seasons in both the

markets for various models are given in Tables 1 and 2. The proposed model results are compared with the results of the following
three models:
(i) Direct application of TES on training data denoted HW
method.
(ii) Direct application of WNN method on the training data
denoted, the WNN model.
(iii) Decompose, threshold and reconstruct the training data into
deterministic and uctuation series using Haar DWT and
then forecast each series separately using WNN. The resulting forecasts are summed up to obtain 24 h ahead load forecast. This is denoted HWNN model.
From the results presented in Tables 1 and 2, it is observed that
the proposed model performs fairly well on weekdays. Towards
the weekend i.e. on Sunday, the forecast accuracy given by HWNN
model is slightly better in the spring season of California market
and in the summer and fall seasons of Spain market. However,
from the weekly MAPE and error variances presented in Tables 3
and 4, it is clear that the proposed model performs well across
all seasons in both the markets. The per unit weekly error
(MAPEWeek) presented in Table 3 brings out the superiority of the
proposed model over other considered models irrespective of
seasons and markets. The weekly error variances presented in
Table 4 brings out the variability of what is still unexplained after
tting the model [50]. These results also show the superiority of
the proposed technique over the other considered models. The
forecasting behavior of the proposed technique is illustrated in
Figs. 1 and 2 for the California and Spain markets respectively.
Fig. 1(a) provides actual load and forecast load using proposed
technique for the winter week in California market. The corresponding plots for the seasons of spring and summer are provided
in Fig. 1(b) and (c). In the same manner, the actual load and proposed model load estimates for the test weeks in the seasons of
Winter, Spring, Summer and Fall of Spain market are given in
Fig. 2(a)(d) respectively.
In summary, it can be stated that the proposed technique shows
the usefulness of the wavelet preprocessor considered in this
paper. In addition it brings out the capability of the triple
exponential smoothing model in capturing the trend and

344

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

Table 1
The daily forecast errors (MAPE) for the analyzed weeks in California energy market.
Winter (January 17th23rd)

Spring (April 17th23rd)

Summer (July 17th23rd)

TES

WNN

HWNN

Proposed model

TES

WNN

HWNN

Proposed model

TES

WNN

HWNN

Proposed model

1.7188
1.4312
0.6060
0.6510
1.0042
2.5742
2.5820

2.7163
1.3947
0.8391
0.7904
1.3361
2.0349
1.0618

2.601
1.4332
0.8179
0.5964
1.0731
2.0349
1.0607

1.2093
1.1833
0.6040
0.4594
0.6793
1.8400
0.9827

1.2485
1.2225
0.6704
0.7271
1.3304
2.2315
2.4814

2.0983
2.0669
1.4034
1.7177
1.617
2.2248
1.786

1.8632
1.5373
0.9655
1.3536
1.6183
1.7234
1.5315

1.0672
0.9696
0.5146
0.5154
1.0819
1.6620
1.6416

2.0506
1.2747
0.7819
0.5084
0.8176
1.7130
2.3591

1.3655
2.1205
3.8194
1.2548
1.5807
1.5671
3.5665

1.4303
2.2422
3.1618
1.2346
1.5309
1.4204
2.866

1.2755
0.8408
0.7226
0.4915
0.6879
1.5682
1.4231

Table 2
The daily forecast errors (MAPE) for the analyzed weeks in Spain energy market.
Winter (January 21st27th)

Spring (April 15th21st)

TES

WNN

HWNN

Proposed model

TES

WNN

HWNN

Proposed model

1.4830
0.6998
0.4017
0.3483
0.5443
3.1206
2.6959

1.5742
1.5738
0.5027
0.6051
1.2139
1.9353
3.465

1.2489
1.0239
0.3716
0.5801
1.0659
1.5939
4.1628

1.3374
0.6452
0.3040
0.4111
0.5485
1.1651
2.3995

1.6730
1.2779
0.3520
0.4050
0.7121
4.0101
3.1665

2.4037
1.4595
0.6188
1.0022
0.8962
2.583
3.0863

2.254
1.3018
0.5341
0.9311
0.7708
2.583
3.012

1.6107
0.7196
0.3040
0.3565
0.5883
2.4004
2.7681

Summer (July 15th21st)

Fall (October 21st27th)

TES

WNN

HWNN

Proposed model

TES

WNN

HWNN

Proposed model

2.2078
0.6878
0.6282
0.2806
0.3588
2.3799
2.2180

2.6488
0.7636
1.0742
0.3758
0.9567
1.8789
2.087

2.4542
0.5656
0.9139
0.4559
0.9152
1.8789
1.7056

1.8085
0.5504
0.6084
0.2723
0.3049
1.3946
1.8313

2.6176
1.0408
0.6183
0.4264
0.5404
3.5552
3.9553

2.1434
0.9244
0.9499
1.092
0.6428
1.4396
2.0888

2.1434
0.9123
0.8437
1.0095
0.6528
1.4396
2.0396

2.2177
0.7314
0.5656
0.4066
0.5225
1.4267
2.4188

Table 3
Weekly forecast errors for the analyzed weeks of different energy markets.
Energy market

Typical week in

MAPEWeek
TES

WNN

HWNN

Proposed model

Winter
Spring
Summer

1.5097
1.4160
1.3579

1.4533
1.8449
2.1821

1.3739
1.5132
1.9837

0.9940
1.0646
1.0967

Winter
Spring
Summer
Fall

1.3276
1.6567
1.2516
1.8220

1.5529
1.7214
1.3978
1.3273

1.4353
1.6267
1.2699
1.2915

0.9730
1.2497
0.9672
1.1842

California

Spain

Table 4
Weekly error variances for the analyzed weeks of different energy markets.
Energy market

Typical week in

EVWeek
TES

WNN

HWNN

Proposed model

Winter
Spring
Summer

0.000157
0.000123
0.000130

0.000160
0.000165
0.000328

0.000155
0.000161
0.000246

0.000085
0.000075
0.000086

Winter
Spring
Summer
Fall

0.000234
0.000316
0.000816
0.000330

0.000182
0.000211
0.000136
0.000120

0.000181
0.000199
0.000120
0.000122

0.000102
0.000155
0.000134
0.000156

California

Spain

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

345

Conclusion
A new approach for forecasting the 24 h ahead electricity load is
presented. The Haar wavelet transform is used as a preprocessor in
the approach to decompose the original load series into deterministic and uctuation series. The seasonal oscillations and trend factors contained in the deterministic component is modeled using
TES method. The faster dynamics of data contained in the uctuation component is modeled using the WNN technique. The proposed approach is illustrated using the load data of the energy
markets of California and Spain for the years 2000 and 2002
respectively. The accuracy of the technique which is based on using
the Haar wavelet combined with HoltWinters additive seasonal
model and WNN model outperforms the direct use of HW model,
WNN model and Haar wavelet combined with WNN model. The
performance of the other wavelet transforms such as the Daubechies wavelets in improving the forecast accuracy will be investigated in the future.

References

Fig. 1. Actual load and forecast load using proposed model for typical weeks in (a)
winter, (b) spring and (c) summer seasons of California energy market.

Fig. 2. Actual load and forecast load using proposed model for typical weeks in (a)
winter, (b) spring, (c) summer and (d) fall seasons of Spain energy market.

seasonality factors of the deterministic component of load data.

Further, the usefulness of the WNN technique in modeling the uctuation component of data is reected in the results of this work.

[1] Taylor JW. Triple seasonal methods for short-term electricity demand
forecasting. Eur J Oper Res 2010;204:13952.
[2] Bashir ZA, El-Hawary ME. Applying wavelets to short-term load forecasting
using PSO-based neural networks. IEEE Trans Power Syst 2009;24(1):207.
[3] Wang J, Wang J, Li Y, Zhu S, Zhao J. Techniques of applying wavelet denoising
into a combined model for short-term load forecasting. Int J Elect Power
Energy syst 2014;62:81624.
[4] Hahn H, Nieberh MS, Stefan Pickl. Electrical load forecasting methods: tools for
decision making. Eur J Oper Res 2009;199:9027.
[5] Ceperic E, Ceperic V, Baric A. A strategy for short term load forecasting by
support
vector
regression
machines.
IEEE
Trans
Power
Syst
2013;28(4):435664.
[6] Gross G, Galiana FD. Short term load forecasting. Proc IEEE 1987:155873.
[7] Tzafestas S, Tzafestas E. Computational intelligence techniques for short-term
electric load forecasting. J Intell Robot Syst 2001;31:768.
[8] Hippert HS, Pedreira CE, Souza RC. Neural networks for short term load
forecasting: a review and evaluation. IEEE Trans Power Syst 2001;16(1):4455.
[9] Alfares HK, Nazeeruddin M. Electric load forecasting: literature survey and
classication of methods. Int J Syst Sci 2002;33:2334.
[10] Fard AK, Fard FK. A new hybrid correction method for short-term load
forecasting based on ARIMA, SVR and CSA. J Exp Theo Artif Intell
2013;25(4):55974.
[11] Moghram I, Rehman S. Analysis and evaluation of ve short term load
forecasting techniques. IEEE Trans Power Syst 1989;4(4):148491.
[12] Amina M, Kodogiannis VS, Petrounias I, Tomtsis D. A hybrid intelligent
approach for the prediction of electricity consumption. Int J Elect Power
Energy Syst 2012;43:99108.
[13] Romere EG, Jaramillo-Moran MA, Diego CF. Monthly electric energy demand
forecasting based on trend extraction. IEEE Trans Power Syst
2006;21(4):194553.
[14] Jaramillo-Maran MA, Romere EG, Diego CF. Monthly electric demand
forecasting with neural lters. Int J Elect Power Energy Syst 2013;49:25363.
[15] Reis AJR, da Silva APA. Feature extraction via multiresolution analysis for
short-term load forecasting. IEEE Trans Power Syst 2005;20(1):18998.
[16] Fard AK, Akbari-Zadeh MR. A hybrid method based on wavelet, ANN and
ARIMA model for short-term load forecasting. J Exp Theo Artif Intell
2014;26(2):16782.
[17] Chen Y, Luh PB, Guan C, Zhao Y, Michel LD, Coolbeth MA, et al. Short-term load
forecasting: similar day-based wavelet neural networks. IEEE Trans Power Syst
2010;25(1):32230.
[18] Zhu Suling, Wang Jianzhou, Zhao Weigang, Wang Juie. A seasonal hybrid
procedure for electricity demand forecasting in China. Appl Energy
2011;88:380715.
[19] Hanmandlu M, Chauhan BK. Load forecasting using hybrid models. IEEE Trans
Power Syst 2011;26(1):209.
[20] Cagdas Hakan Aladag, Egrioglu Erol, Kadilar Cem. Forecasting nonlinear time
series with a hybrid methodology. Appl Math Lett 2009;22:146770.
[21] Lee Cheng-Ming, Ko Chia-Nan. Short term load forecasting using lifting
scheme and ARIMA models. Expert Syst Appl 2011;38(5):590211.
[22] Amjady N, Keynia F. Short term load forecasting of power systems by
combination of wavelet transform and Neuro-evolutionary algorithm. Energy
2009;34(1):4657.
[23] Pandey AK, Singh Devender, Sinha SK. Intelligent hybrid wavelet models for
short-term load forecasting. IEEE Trans Power Syst 2010;25(3):126673.
[24] Suseelatha A, Sudheer G, Bharathi D. A Simple Hybrid Model for Short-Term
Load Forecasting, J Engg, Hindawi 2013; Article ID 760860. doi:10.1155/2013/
760860.

346

G. Sudheer, A. Suseelatha / Electrical Power and Energy Systems 64 (2015) 340346

[25] Rahmat-Allah H, Amooshahi Habib, Parastegari M. A hybrid intelligent

algorithm based short-term load forecasting approach. Int J Elect Power
Energy Syst 2013;45:31324.
[26] Lora AT, Santos R, Exposito G, Ramos M. Time series prediction: application to
the short-term electric energy demand. Lect Notes Aritic Intell
2004;3040:57786.
[27] Strang G, Nguyen T. Wavelets and lter bank. Wellesley-Cambridge Press;
1995.
[28] Mallat S. A theory for multiresolution signal decomposition: the wavelet
representation. IEEE Trans Pattern Anal Mach Intell 1989;11(7):67493.
[29] Soltani S. On the use of the wavelet decomposition for time series prediction.
Neurocomputing 2002;48:26777.
[30] Mabrouk AB, Abdallah NB, Dhifaoui Z. Wavelet decomposition and
autoregresssive model for time series prediction. Appl Math Comput
2008;199(1):33440.
[31] Price R, Vincent S, Lebohec S. Haar wavelet as a tool for the statistical
characterization of variability. Astro Phys 2011;34:8717.
[32] Daubechies I. Ten lectures on wavelets. CBMS-NSF series in applied
mathematics, 61. Philadelphia: SIAM; 1992.
[33] Stelwagon E. Exponential smoothing: the workhorse of business forecasting.
Foresight: Int J Appl Forecast 2012;27:238.
[34] Weron Rafal. Modeling and forecasting electricity load and prices: a statistical
approach. The Wiley Finance series; 2006.
[35] Gardner ES. Exponential smoothing: the state of the art. Int J Forecast
1985;4:128.
[36] Gardner ES. Exponential smoothing: the state of the art part II. Int J Forecast
2006;22:63766.
[37] Taylor JW, de Menezes LM, McSharry PE. A comparision of univariate methods
for forecasting electricity demand up to a day ahead. Int J Forecast
2006;22:116.
[38] Siddharth A, Taylor JW. Short-term forecasting of anomalous load using
rule-based triple seasonal methods. IEEE Trans Power Syst 2013;28(3):
323542.

[39] Gould PG, Koehler AB, Ord JK, Snyder RD, Hyndman RJ, Vahid-Araghi F.
Forecasting time-series with multiple seasonal patterns. Eur J Oper Res
2008;191:20722.
[40] Winters P. Forecasting sales by exponentially weighted moving averages.
Manag Sci 1960;6:32442.
[41] Fix E, Hodges J.L. Discriminatory analysis, nonparametric discrimination:
consistency properties. USAF School of Aviation Medicine Randolgh eld, Tx,
Project No. 21-49-004 1951;4, Contract No. AF41(128)-31.
[42] Dudani SA. The distance-weighted nearest-neighbor rule. IEEE Trans Syst Man
Cyber 1976:3257.
[43] Lora AT, Santos R, Exposito G, Ramos M. Electricity market price forecasting
based on weighted nearest neighbors techniques. IEEE Trans Power Syst
2007;22(3):1294301.
[44] Akbaryan F, Bishnoi PR. Smooth representation of trends by a wavelet-based
technique. Comput Chem Eng 2000;24:191343.
[45] Sang YF, Wang D, Wu JC, Zhu QP, Wang L. Entropy based wavelet denoising
method for time series analysis. Entropy 2009;11(4):112347.
[46] Gencay Ramzan, Seluk Faruk, Whitcher Brandon. An introduction to
wavelets and other ltering methods in nance and economics. Academic
press; 2002.
[47] Donoho DL, Johnstone IM. Ideal spatial adaptation via wavelet shrinkage.
Biometrika 1994;81:42555.
[48] http://www.ucei.berkely.edu/ucei/datamine/ucei-data.
[49] http://www.omel.es/.
[50] Conejo AJ, Plazas MA, Espnola R, Molina AB. Day-ahead electricity price
forecasting using the wavelet transform and ARIMA models. IEEE Trans Power
Syst 2005;20(2):103542.
[51] Tan Zhongfu, Zhang J, Wang J, Jun Xu. Day-ahead electricity price forecasting
using wavelet transform combined with ARIMA and GARCH models. Appl
Energy 2010;87(11):360610.