Data Analytics for Load and Price Forecasting via Enhanced Support Vector Regression Technical Report for MSCS Course: Research Methodology in Information Technology RMIT CUI Fall 2018 23

Preprint (PDF Available) · December 2018with 99 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
Cite this publication
Preprints and early-stage research may not have been peer reviewed yet.
Figures - uploaded by Nadeem Javaid
Author content
All content in this area was uploaded by Nadeem Javaid
Content may be subject to copyright.
Data Analytics for Load and Price Forecasting via
Enhanced Support Vector Regression Technical Report for MSCS
Course: Research Methodology in Information Technology
RMIT CUI Fall 2018 23
Tanzeela Sultana1, Nadeem Javaid1
1COMSATS University, Islamabad 44000, Pakistan;
nadeemjavaidqau@gmail.com
Abstract—In this paper, month-ahead electricity load and price
forecasting is done to achieve accurate forecasting. The data
of electricity load is taken from the smart meter in London.
Electricity load data of five months is taken from one block along
with the weather data. Data analytics techniques are used in this
paper for month-ahead electric load and price prediction. In
this paper, forecasting is done in multiple stages. At first feature
extraction and selection is performed to make data suitable for
efficient forecasting and to reduce complexity of data. Further
regression techniques are used for prediction. Feature extraction
is done by using Singular Value Decomposition (SVD) and then
feature selection is done in two-stages, by using Random Forest
(RF) and Recursive Feature Elimination (RFE). For forecasting
Logistic Regression (LR), Support Vector Regression (SVR) is
used. Moreover forecasting is done by the proposed technique
Enhanced Support Vector Regression (EnSVR), which is modified
from SVR.
Index Terms—Load Forecasting, Price Forecasting, Data Ana-
lytics, Logistic Regression, Support Vector Regression, Enhanced
Support Vector Regression
I. INTRODUCTION
With the increasing demand of electricity along with the
growth of advanced technologies, the conventional electrical
energy markets are evolving consistently. By this the tradi-
tional approaches for power systems are decreasing and the
concept of smart meters and Smart Grid (SG) is getting high.
The electrical data in smart meters is generated more than it
is generated in the traditional systems. Traditional Grid (TG)
uses only one way transmission strategy, however, SG is the
advanced approach and use two way transmission strategy.
SG also creates a communication link between consumers
and utility markets. SG, can be defined as, an advanced
development that gathers the Information Communication and
Technology (ICT) and traditional energy systems to improve
productivity and ability of the grid by making a relation
between consumer and power systems. Increasing electricity
demand results in vast generation of electricity. Due to the
load over traditional grids and meters, it is now necessary to
manage electricity.[1]-[4] SG provides the distributed forms of
energy generation to minimize uncertainty, that often occurs
in traditional centralized grids. In traditional energy systems,
power management is considered to be a massive issue. several
benefits are provided by SG such as communication between
TABLE I: List of Abbreviation
ARIMA Autoregressive Integrated Moving Average
ABC Artificial Bee Colony
AEMC Australian Energy Market Commission
AEMC Australian Energy Market Commission
AGG Arian Golden Group
CMI Conditional Mutual Information
DA Data Analytics
DR Demand Response
DTW Dynamic Time Warping
EnSVR Enhanced Support Vector Regression
EMD Empirical Mode Decomposition
EMDSW Empirical Mode Decomposition named Slicing Window
ELM Extreme Learning Machine
EnSVR Enhanced Support Vector Regression
ENN Elman Neural Network
ERCOT Electric Reliability Council of Texas
FWPT Flexible Wavelet Packet Transform
GTA Grid Traverse Algorithm
GS Greater Sydney
IENN Improved Elman Neural Network
ICT Information Communication and Technology
LR Logistic Regression
MA Moving Average
MIMO Multi-Input Multi-Output
MLR Multiple Linear Regression
MHMM Multi-Hidden Markov Model
NA North American
NLSSVM Non-linear Least Square Support Vector Machine
NN Neural Networks
NSSO Novel Shark Smell Optimization
NSW New South Wales
PLD Power Load Decomposition
PCR Principle Component Regression
PJM Pennsylvania-New Jersey-Maryland
PQ Power Quality
PSO Particle Swarm Optimization
QLD QueensLand
RVFL Random Vector Functional Link Network
RF Random Forest
RFE Recursive Feature Elimination
RNN Ridgelet Neural Network
SG Smart Grid
SVD Singular Value Decomposition
SVM Support Vector Machine
SVR Support Vector Regression
SVC Support Vector classification
SMA Simple Moving Average
ToU Time of Use
TG Traditional Grid
TV-SABC Time-Varying Coefficients and Stumble Generation Operator
VSTLF Very Short Term Load Forecasting
utility store and consumer, security, generation cost, processing
and storing energy data [5]-[7].
As the power system is moving towards SG, more electrical
data is generated. To make gird systems more advance and
for efficient management of energy, Data Analytics (DA) is
used. Different DA techniques are used to perform multiple
operations on data. For efficient electricity management DA
uses forecasting techniques. To take user rate of electricity
consumption into account, and generation by energy markets.
Forecasting of load and price is needed, to manage the
electricity consumption. The best forecasting algorithm gives
accurate forecasting results that can minimizes the issues
related to power [8] and it also helps to improve the electricity
management system. As electric data generated from the smart
meters is time series data, there are time series forecasting
paradigms for different intervals: short term, medium term
and long term [9]. Many techniques are used in literature for
forecasting both load and price in different horizon. However,
in this paper, electricity load and price forecasting is done
for month-ahead in three stages by regression techniques LR,
SVR and EnSVR by taking the dataset of one block of smart
meters in London. Load and weather data of that block is also
used for month-ahead load and price forecasting.
A. Motivation
A lot of work in literature is done for efficient management
of electricity. Many techniques are proposed to forecast elec-
tricity load and price in smart meters and smart grids for short-
term, medium-term and long-term forecasting. Many papers
proposed the hybrid model for prediction such as [2] proposes
a hybrid of Grid Traverse Algorithm (GTA) and Particle
Swarm Optimization (PSO) for short-term load forecasting,
[4] does a three-stage prediction for both load and price
through different methods at each stage, [6] mixes Non-Linear
Regression with SVM for month-ahead price forecast. [2], [5],
[6] and [7] uses SVR for forecasting. However, by taking
all the studies into consideration, the proposed paper aims
to forecast both load and price for month-ahead to achieve
accuracy. The main objective of papers is to achieve more
accuracy and performance in forecasting and to resolve peak
load and price variation in smart grids.
B. Problem Statement
[7] predicts very short-term half-hour ahead load forecast-
ing, using On-line SVR where prediction is done to overcome
the issue of memory and storage using Advanced On-line
methods. The paper used the dataset of Irish Commission
for Energy Regulation (CER) utility company and achieves
accuracy and minimizes the storage issues by using ensemble
Tree-based ensemble methods with Advanced On-line meth-
ods. However, model outperforms for storage and memory
issues, besides this model gives single output with only one-
hour ahead prediction, a further methodology is required for
the multiple outputs. Paper [7] also lacks in performance to
predict separate day. In paper [5] short-term load forecast is
predicted to calculate Demand Response (DR) for an office
building in eastern China. The model’s objective is to achieve
maximum forecasting accuracy under DR baseline using his-
torical electric load and weather data. The paper uses SVR for
accurate predictions. The proposed model forecast load and
price by two existing schemes LR, SVR and with proposed
scheme EnSVR.
C. Contribution
In this paper load and price forecast is performed in a new
way, by making a combination of schemes that are never
used before for forecasting. The major purpose is to achieve
accuracy in terms of electricity load and price prediction.
The main contributions of this paper are:
Only load data of one block smart meter, along with
weather data is given in the dataset, the price is calculated
from the given load data to predict price along with load,
Data preprocessing is done by both feature extraction and
feature selection. For extraction, SVD is used and for feature
selection, first RF and after that RFE is used.
Month-ahead forecasting for load and price is accomplished
by executing multiple regressors such as LR, SVR. The
EnSVR method is proposed to achieve more accuracy and
for better electricity forecasting.
II. RE LATE D WORK
With the generation of electricity in smart grids, the problem
of electrical load and price is increasing. To tackle the electric
load and price forecast issues, different schemes are proposed
in the literature. Paper [1] uses a 2-stage forecast model for the
short-term electric load prediction in the university campus.
Five years data is taken from a university campus in Korea
and 2-stage prediction is done. At the first stage, by using
Moving Average (MA) method and secondly by using RF
technique. Furthermore, the performance is evaluated by the
time-series cross-validation process. Another short-term load
forecasting is done by [2] over distributed systems using SVR
and two-step hybrid optimization algorithm based on GTA
and PSO. The data is calculated from the Partner utility’s
distribution feeder and the prediction is done on seasoned
base. Load prediction is also done by taking external factors
like temperature and weather data into consideration. paper
[3] proposed a forecasting model for scheduling and optimum
planning by using a historical load with weather data and
calendar effects of Australian grids of residential areas of
New South Wales (NSW) and Greater Sydney (GS). Calendar
effects are used for prediction in several periods of the
calendar: days, months and years. Many forecasting techniques
such as Multiple Linear Regression (MLR), Neural Networks
(NN), regression trees and Support Vector Machine (SVM)
are used. [4] proposed hybrid algorithms based on demand-
side management models to predict both price and load con-
currently using data of New York Independent System Oper-
ator (NYISO), NSW and Pennsylvania New-Jersey Maryland
(PJM). Forecasting in this paper is performed in three stages;
at first stage, Flexible Wavelet Packet Transform (FWPT) and
Conditional Mutual Information (CMI) algorithms are used for
selection, then Multiple-Input Multiple-Output (MIMO) based
Non-linear Least Square Support Vector Machine (NLSSVM)
and Autoregressive Integrated Moving Average (ARIMA) is
applied to define a relationship between load and price, at third
stage enhances ABC on the basis of Time-Varying Coefficients
and Stumble Artificial Bee Colony (TV-SABC) for parameter
optimization. In [5] SVR model is employed to minimize the
risk of a load during the peak hours. Data is taken from four
office buildings in China to examine DR. Electricity price is
forecast in [6] to examine the abnormalities in the electric
power industry. Month-ahead average price forecast is done by
hybrid Non-Linear Regression and SVM on data of Electric
Reliability Council of Texas (ERCOT) electricity market in
Texas, United States of America (USA). The limitation of this
system is that it gives accurate forecasting for month-ahead
only. With the enhancement of electricity grid market, most
of the data is processed online. [7] proposed a short-term
load forecasting with online SVR to overcome the biasness
of the Irish CER grid market. The paper has compared ten
models for this purpose, however, online SVR achieved more
accuracy. For short-term load forecasting, [8] employed a
novel method Empirical Mode Decomposition (EMD)-mixed-
Extreme Learning Machine (ELM), where EMD is used for
the decomposition of complex load features to smoothen the
data. RBF and UKF kernels are used in EML method. Load
data of NSW, Victoria and Queensland (QLD) is used for
forecasting. Another hybrid approach is used for short-term
load prediction. which is comprised of EMD and Random
Vector Functional Link Network (RVFL) in [9]. For better
forecasting, ensemble method RVFL based on Dynamic Time
Warping (DTW) and EMD is presented for accuracy and
performance. Load data is collected from AEMO along with
NSW, Tasmania (TAS), QLD, South Australia (SA) and Victo-
ria. Electricity grids are often affected by weather conditions.
The PCR based method is used for prediction of temperature
on power grids in [10], the conductor temperature is collected
by the installation of a sensor over transmission lines of
TaiPower grid in Taiwan. The main objective of the paper is
to make improvement in transmission lines that are influenced
by weather conditions. PQ disturbance is one the major issues
in power systems. To tackle the PQ problem and reduce the
negative effect [11] proposed a Multi-Hidden Markov Model
(MHMM) method. MHMM based tools are used to predict
Power Quality (PQ) disturbance for future using weather data.
Also Hadoop-based PQ analysis method is proposed in [11]
minimize the computational time taking into account the PQ
data of smart grids in Chinese cities. One-step-ahead and
multi-step-ahead forecasting is done in the paper. Two-stage
forecast model based on Ridgelet Neural Network (RNN)
and Elman Neural Network (ENN) is executed by [13] and
the method is compared with different algorithms for better
accuracy measurement. However, the proposed model gives
best results and minimizes the prediction errors such as per-
centage error, semantic error, measure error and many others.
Forecasting is done on AEMC, PJM and NA electric utilities.
[14] proposed a new methodology for forecasting short-term
electrical load. For feature selection enhanced version of EMD
named slicing window Empirical Mode Decomposition Slicing
Window (EMDSW) is computed, forecast engine algorithm is
used with more bearing and least duplication measures through
Pearson’s Correlation Coefficient. Forecasting is executed by
Improved Elman Neural Network (IENN) and Novel Shark
Smell Optimization (NSSO) algorithms. Paper [15] proposed a
hybrid model for both linear and non-linear predictions. Linear
prediction is done by Auto-Regressive Modeling using tem-
perature data and non-linear by Gaussian Process Regression
for weather data. Dataset is taken from GIST campus electric
power consumption for day-ahead load prediction.
III. SYS TE M MOD EL
In this section, the proposed system model of the paper is
presented in detail. The flow of the system model is shown in
figure 3.
A. Dataset Description
The dataset of smart meters in London is used. In the whole
dataset, data of household electric load of multiple blocks
is given. Also, load data of half-hourly and hourly basis is
presented along with the daily temperature data and hourly
weather data of three years, from 2013 to 2015.
In this paper, for accurate forecasting of load and price,
electrical load data of only one block smart meter is considered
along with weather data of that block is used in this paper.
Load and weather data of five months is used to forecast one-
month ahead load and price of that block. The main focus of
the paper is to forecast price and load. As the dataset does
not have any price data of smart meter, the paper calculated
price from the given load data to predict the future price for
the smart meter of that block.
B. Feature Importance of Data
Feature importance of data is calculated for of every feature
in the dataset. It is examined by observing the effect of every
independent feature in the data for the target feature. Figure 1,
shows the feature importance for a load. For load forecasting,
price feature shows more influence on the target than other
features. However, the feature that does not affect the target
is humidity.
Like load, feature importance for Price is also calculated. The
figure 2, shows the feature importance of every feature to the
target. For price, pressure shows more influence than other and
humidity does not shows any effect also for price forecasting.
C. Proposed System Model
Figure 3 demonstrates the flow of presented system model.
In the proposed model, first, the load data of one block smart
meter is taken along with the weather data of the block.
Then the price is calculated for price forecasting and data is
preprocessed. Feature extraction is done by SVD. After that
feature selection is executed by two techniques, RF and RFE
for getting the best features for accurate forecasting. In the
TABLE II: Summary of Related Work
Technique used Features Dataset Region Limitations
RF, ANN [1] Academic year, load University campus Seoul, Korea -
SVR, PSO [2] historical load Partner’s Utility US -
MLR [3] Calender effect, load,
weather data
Ausgrid NSW and GS Minimize accuracy due to
large dataset
3-Stage methodology [4] Price, demand NYISO and PJM US -
SVR [5] Temperature, electricity
load, weather
Office building Eastern China Predict few hours of one
building only
Hybrid Non-Linear Re-
gression, SVR [6]
Price, reading rate, aver-
age wages
ERCOT Texas, USA Model works well only for
month-ahead forecast
On-line SVR [7] half-hour load, time,
weekdays, weather data
Irish CER Ireland Model does not perform
well with separate day
values
EMD-ELM [8] Half-hour electric load NSW, Victoria, QLD Australia
DWT, EMD, RVFL [9] year, max, median, mean,
standard deviation
AEMO Australia
PCR [10] Weather data, grid load,
transmission line load,
Tai power grid Taiwan Model works only for sin-
gle tranission line
MHMM, Hadoop-based
PQ [11]
PQ disturbance, weather
condition data
PQ data Chinese Cities Only one-step ahead pre-
diction
Shape-based DTW, PLD
[12]
Household load curves,
device ToU
Opower corporation US Minimum convex
optmizaiton
RNN, ENN [13] Historical load, tempera-
ture and humidity data
AEMC, PJM and NA util-
ity markets
Australia, America
EMDSW, IENN, NSSO
[14]
Load demand AGG Ardabil-Iran Model performs best only
for small-scale load pre-
diction
Auto-Regressive,
Gaussian Process
Regression [15]
Power consumption,
weather, temperature
GIST power consumption South Korea -
Fig. 1: Feature Importance for Load
proposed model, forecasting is done by LR, SVR and the SVR
is modified to EnSVR.
IV. SIMULATION RESULTS AN D REASONING
In this section, results of electricity load and price forecast
are presented with reasoning.
A. Load and Price Forecasting
In this paper, month-ahead load and price is predicted by
taking five month historical data of smart meter in one block.
Fig. 2: Feature Importance for Price
Fig. 3: Flow of System Model
Load and price forecast are shown in figure 4 and figure 5.
Electricity load and price forecasting is done by using mul-
tiple techniques. Feature extraction is done by SVD. Feature
selection is performed by using two techniques, RF and RFE.
Regression techniques are used for load and price forecasting.
Fig. 4: Load Forecast
Fig. 5: Price Forecast
Existing techniques LR, SVR and proposed scheme EnSVR
is used for prediction. At first step, LR is used and at second
stage, SVR is used. At the final stage, EnSVR is used to
achieve prediction accuracy.
1) Logistic Regression (LR): Logistic Regression (LR) is
a statistical model. LR shows relationship of one dependent
variable with other independent variables by using logistic
function.
2) Support Vector Regression (SVR): Support Vector Re-
gression (SVR) is the regression technique, based on Support
Vector Machine (SVM) model. SVR model is produced by
Support Vector Classification (SVC) that depends on training
data. SVR works based on the functions of SVM model.
3) Enhanced Support Vector Regression (EnSVR): To
achieve accurate forecasting results, this paper proposed
EnSVR technique. EnSVR is modified from original SVR
technique. EnSVR is retrieved by making changes in SVR
parameters such as cost, epsilon, gamma and tolerance. To
achieve accuracy, parameter values are changed by cost=1.0,
epsilon=0.001, gamma=10, tolerance=0.1. Forecasting results
shows that EnSVR shows more accuracy than SVR for load
and price as illustrated in accuracy tables. and it also shows
best forecasting as shown in fig 4 and fig 5.
B. Performance Evaluation
The error value of forecasting models is calculated by
performance evaluators such as MAPE , MAE, RMSE, and
MAE. TABLE III shows the error percentage for LR, SVR
and EnSVR for load. MAPE percentage is more than other
measurements. TABLE IV shows the error measurement for
price. Also for price, MAPE error is more that MAE, MSE
and RMSE.
TABLE III: Performance Evaluation for Load
Method RMSE% MSE% MAE% MAPE%
LR 0.0024 0.0006 0.0021 inf
SVR 0.0005 2.704 0.0004 22.142
EnSVR 0.0001 1.065 8.3451 5.706
TABLE IV: Performance Evaluation for Price
Method RMSE% MSE% MAE% MAPE%
LR 0.0304 0.0927 0.0168 114.12
SVR 0.0071 0.0051 0.006 1893.3
EnSVR 0.002 0.0004 0.001 7.164
C. Accuracy Measurement
The goal of this paper is to achieve more accuracy for load
and price forecasting on the data of smart meters of London.
The accuracy is calculated for both load and price by using
LR, SVR and EnSVR. Table shows the accuracy rate for load
where LR achieves 89% SVR has 60% accuracy. However our
proposed technique EnSVR achieves more accuracy for load
which is 94% more than other techniques used.
TABLE V: Accuracy for Load
Method LR SVR EnSVR
Accuracy 89% 60% 94%
Accuracy is also examined for price on the base of every
technique used. Table shows that LR has 74% and SVR
gets 60% accuracy for price. Whereas EnSVR achieves 91%
accuracy that is more than other techniques.
TABLE VI: Accuracy for Price
Method LR SVR EnSVR
Accuracy 74% 60% 91%
V. CONCLUSION
With the rapid growth of electricity it is becoming very
difficult for utility market to handle the issues relating to
forecast for future load and price. In this proposed paper, short
term load and price forecasting is done to achieve accuracy
by taking load data of smart meters in London with weather
data of about five months. There are many strategies proposed
for forecasting. However in this paper, short-term load and
price forecasting is done by using multiple techniques and
with the proposed technique. For forecasting first LR and SVR
is used. After that the forecasting is done by the proposed
scheme EnSVR. By comparing all the techniques it clearly
shows that the proposed technique outperforms the existing
techniques. EnSVR is the modified form of SVR and it gives
more accurate forecasting for load and price.
REFERENCES
[1] Moon, Jihoon, Kyu-Hyung Kim, Yongsung Kim, and Eenjun Hwang.“A
Short-Term Electric Load Forecasting Scheme Using 2-Stage Predictive
Analytics.”In Big Data and Smart Computing (BigComp), 2018 IEEE
International Conference on, pp. 219-226. IEEE, 2018.
[2] Jiang, Huaiguang, Yingchen Zhang, Eduard Muljadi, Jun Jason Zhang,
and David Wenzhong Gao. “A short-term and high-resolution distribu-
tion system load forecasting approach using support vector regression
with hybrid parameters optimization.”IEEE Transactions on Smart Grid
9, no. 4 (2018): 3341-3350.
[3] Lusis, Peter, Kaveh Rajab Khalilpour, Lachlan Andrew, and Ariel
Liebman. “Short-term residential load forecasting: Impact of calendar
effects and forecast grSanularity.”Applied Energy 205 (2017): 654-669.
[4] Ghasemi, Ali, Hossien Shayeghi, Mohammad Moradzadeh, and Mo-
hammad Nooshyar. “A novel hybrid algorithm for electricity price and
load forecasting in smart grids with demand-side management.”Applied
energy 177 (2016): 40-59.
[5] Chen, Yongbao, Peng Xu, Yiyi Chu, Weilin Li, Yuntao Wu, Lizhou Ni,
Yi Bao, and Kun Wang. “Short-term electrical load forecasting using
the Support Vector Regression (SVR) model to calculate the demand
response baseline for office buildings.”Applied Energy 195 (2017): 659-
670.
[6] Ziming, M. A., Haiwang Zhong, X. I. E. Le, X. I. A. Qing, and K. A.
N. G. Chongqing. “Month ahead average daily electricity price profile
forecasting based on a hybrid nonlinear regression and SVM model:
an ERCOT case study.”Journal of Modern Power Systems and Clean
Energy 6, no. 2 (2018): 281-291.
[7] Vrablecov´
a, Petra, Anna Bou Ezzeddine, Viera Rozinajov´
a, Slavom´
ır
ˇ
S´
arik, and Arun Kumar Sangaiah.“Smart grid load forecasting using
online support vector regression.”Computers and Electrical Engineering
65 (2018): 102-117.
[8] Chen, Yanhua, Marius Kloft, Yi Yang, Caihong Li, and Lian Li.
“Mixed kernel based extreme learning machine for electric load fore-
casting.”Neurocomputing (2018).
[9] Qiu, Xueheng, Ponnuthurai Nagaratnam Suganthan, and Gehan AJ Ama-
ratunga. “Ensemble incremental learning Random Vector Functional
Link network for short-term electric load forecasting.”Knowledge-Based
Systems 145 (2018): 182-196.
[10] Jiang, Joe-Air, Yu-Ting Liang, Chia-Pang Chen, Xiang-Yao Zheng,
Cheng-Long Chuang, and Chien-Hao Wang. “On dispatching line am-
pacities of power grids using weather-based conductor temperature
forecasts.”IEEE Transactions on smart Grid 9, no. 1 (2018): 406-415.
[11] Xiao, Fei, and Qian Ai. “Data-Driven Multi-Hidden Markov Model-
Based Power Quality Disturbance Prediction that Incorporates Weather
Conditions.”IEEE Transactions on Power Systems (2018).
[12] Teeraratkul, Thanchanok, Daniel O’Neill, and Sanjay Lall. “Shape-
based approach to household electric load curve clustering and predic-
tion.”IEEE Transactions on Smart Grid 9, no. 5 (2018): 5196-5206.
[13] Ghadimi, Noradin, Adel Akbarimajd, Hossein Shayeghi, and Oveis
Abedinia. “Two stage forecast engine with feature selection technique
and improved meta-heuristic algorithm for electricity load forecast-
ing.”Energy 161 (2018): 130-142.
[14] Mohammadi, Mohsen, Faraz Talebpour, Esmaeil Safaee, Noradin
Ghadimi, and Oveis Abedinia. “Small-scale building load forecast based
on hybrid forecast engine.”Neural Processing Letters 48, no. 1 (2018):
329-351.
[15] Song, Junho, and Euiseok Hwang. “Hybrid Day-ahead Load Forecasting
with Atypical Residue based Gaussian Process Regression.”In Proceed-
ings of the Ninth International Conference on Future Energy Systems,
pp. 631-634. ACM, 2018.
[16] Zhang, Yang, Ce Li, and Lian Li. “Wavelet transform and Kernel-
based extreme learning machine for electricity price forecasting. ”Energy
Systems 9, no. 1 (2018): 113-134.
[17] F¨
orderer, Kevin, Mischa Ahrens, Kaibin Bao, Ingo Mauser, and Hartmut
Schmeck. “Towards the Modeling of Flexibility Using Artificial Neural
Networks in Energy Management and Smart Grids: Note. ”In Proceed-
ings of the Ninth International Conference on Future Energy Systems,
pp. 85-90. ACM, 2018.
[18] Gao, Wei, Ayda Darvishan, Mohammad Toghani, Mohsen Mohammadi,
Oveis Abedinia, and Noradin Ghadimi. “Different states of multi-block
based forecast engine for price and load prediction.”International Journal
of Electrical Power and Energy Systems 104 (2019): 423-435.
[19] Nowotarski, Jakub, and Rafał Weron. “Recent advances in electricity
price forecasting: A review of probabilistic forecasting.”Renewable and
Sustainable Energy Reviews (2017).
[20] Dong, Xishuang, Lijun Qian, and Lei Huang. “Short-term load forecast-
ing in smart grid: A combined CNN and K-means clustering approach.
”In Big Data and Smart Computing (BigComp), 2017 IEEE International
Conference on, pp. 119-125. IEEE, 2017.
[21] Bramer, Lisa M., J. Rounds, C. D. Burleyson, D. Fortin, J. Hathaway, J.
Rice, and I. Kraucunas. “Evaluating penalized logistic regression models
to predict Heat-Related Electric grid stress days. ”Applied Energy 205
(2017): 1408-1418.
ResearchGate has not been able to resolve any citations for this publication.
  • Article
    Short-term load forecasting is of major interest for the restructured environment of the electricity market. Accurate load forecasting is essential for effective power system operation, but electricity load is non-linear with a high level of volatility. Predicting such complex signals requires suitable prediction tools. This paper proposes a hybrid forecast strategy including novel feature selection technique, and a complex forecast engine based on a new intelligent algorithm. The electricity load signal is first filtered by feature selection technique to select appropriate candidates as input for the forecast engine. Then, the proposed two stage forecast engine is implemented based on ridgelet and Elman neural networks. All forecast engine parameters are chosen based on a novel intelligent algorithm to improve its accuracy and capability. Different electricity markets were considered as test cases to compare the proposed method with several current algorithms. Additionally, the proposed forecasting model measures the absolute forecasting errors in this work (among seven types of measurements i.e., absolute forecasting errors, measures based on percentage errors, symmetric errors, measures based on relative errors, scaled errors, relative measures and other error measures). The results validate the effectiveness of the proposed method.
  • Article
    Power quality (PQ) disturbance in power systems has been a concern for operators and customers. The purpose is to locate and forecast the presence of PQ disturbances to suppress or avoid their negative effects on power grid and appliances. This study, using a multi-hidden Markov model (MHMM), motivates data-driven tools to achieve situation awareness of PQ disturbance. We first design a modified adaptive-sorted neighborhood method that consists of blocking and merging phases to locate PQ disturbance sources from a large volume of PQ records. We then group and discretize the data on disturbance sources and weather conditions according to regions. The capability of MHMM-based tools to predict future PQ disturbance level can be improved by clustering the training set of time series of PQ disturbance levels based on weather conditions. A Hadoop-based PQ analysis framework is proposed to reduce computational times, considering the volume of PQ data in a realistic power grid is large. We utilize numerical case studies that use real data collected from a power system of a Chinese city to investigate the correctness and feasibility of the proposed method.
  • Article
    Short term electric load forecasting, as an important tool in the electricity market, plays a critical role in the management of electric systems. Proposing an accuracy and optimization method is not only a challenging task but also an indispensable part of the energy system. More and more accurate forecasting methods are needed by different people in different areas. This paper proposes a novel short-term electric load forecasting method EMD-Mixed-ELM which based on empirical mode decomposition (EMD) and extreme learning machine (ELM). EMD-Mixed-ELM first uses the empirical mode decomposition to decompose the load series for capturing the complicated features of the electric load and de-noising the data. Considering that the performance of extreme learning machine (ELM) is greatly influenced by the choice of kernel, the mixed kernel method is proposed for ELM. The mixed kernel combines the RBF kernel and the UKF kernel. The forecasting results of the EMD-Mixed-ELM are proved to be better than all the other three methods (RBF-ELM, UKF-ELM and Mixed-ELM) and other existing methods (MFES, ESPLSSVM and Combined method). To verify the forecasting ability of the EMD-Mixed-ELM, half-hourly electric load data from the state of New South Wales, Victoria and Queensland in Australia are used in this paper as a case study. The experimental results clearly indicate that for this three datasets, the forecasting accuracy of the proposed method is superior to other methods.
  • Conference Paper
    The prediction accuracy of electric power consumption plays a crucial role for the efficiency of a smart grid. Hybrid approaches that jointly account for the linear and nonlinear portions of the electric load have shown promising performance because of the mixture of memory effects and random environmental perturbations. Especially for day-ahead short-term prediction, the potentially long time gap between the measurements and prediction point degrades the linear prediction performance, while the nonlinear prediction based on the weather forecast may supplement the degradation. This paper proposes a residue-based hybrid model that uses linear prediction by auto-regressive modeling and nonlinear prediction by Gaussian process regression with atypical residue of the weather forecast, particularly the difference of weather station forecasted and linear predicted local temperatures. Since the typical memory effect of the temperature can be double counted by both models, atypical residue without its linear prediction contribution is employed for the Gaussian process regression step. To verify the performance of the proposed scheme, a GIST campus electric power consumption dataset is evaluated. As expected, the linear prediction residue shows larger correlation to the atypical residue of temperature than the temperature itself. Consequently, hybrid model with the atypical residue temperature based Gaussian process regression shows improved performance in the day ahead load prediction.
  • Conference Paper
    This paper presents a novel approach to the representation and communication of the energy flexibility of distributed energy resources. The approach uses artificial neural networks (ANNs) to represent the devices and act as surrogate models. The main benefit of this approach is its potential to represent arbitrary energy flexibilities and the resulting universal applicability in various usage patterns, some of which are presented in detail in this paper. Furthermore, the flexibility represented by an ANN can be conditioned on the state of the corresponding devices and their environment, such that only a small state update needs to be communicated to construct feasible load profiles by a third party. Therefore, in contrast to other approaches, such as support vector data description, new ANNs only need to be constructed once the device configuration changes.
  • Article
    Full-text available
    With the deregulation of the electric power industry, electricity price forecasting plays an increasingly important role in electricity markets, especially for retailors and investment decision making. Month ahead average daily electricity price profile forecasting is proposed for the first time in this paper. A hybrid nonlinear regression and support vector machine (SVM) model is proposed. Off-peak hours, peak hours in peak months and peak hours in off-peak months are distinguished and different methods are designed to improve the forecast accuracy. A nonlinear regression model with deviation compensation is proposed to forecast the prices of off-peak hours and peak hours in off-peak months. SVM is adopted to forecast the prices of peak hours in peak months. Case studies based on data from ERCOT validate the effectiveness of the proposed hybrid method.
  • Article
    Short-term electric load forecasting plays an important role in the management of modern power systems. Improving the accuracy and efficiency of electric load forecasting can help power utilities design reasonable operational planning which will lead to the improvement of economic and social benefits of the systems. A hybrid incremental learning approach composed of Discrete Wavelet Transform (DWT), Empirical Mode Decomposition (EMD) and Random Vector Functional Link network (RVFL) is presented in this work. RVFL network is a universal approximator with good efficiency because of the randomly generated weights between input and hidden layers and the close form solution for parameter computation. By introducing incremental learning, along with ensemble approach via DWT and EMD into RVFL network, the forecasting performance can be significantly improved with respect to both efficiency and accuracy. The electric load datasets from Australian Energy Market Operator (AEMO) were used to evaluate the effectiveness of the proposed incremental DWT-EMD based RVFL network. Moreover, the attractiveness of the proposed method can be demonstrated by the comparison with eight benchmark forecasting methods.
  • Article
    Electricity load forecasting plays an important role for optimal power system operation. Accordingly, short term load forecast (STLF) is also becoming an important task by researchers to tackle the mentioned problem. As a consequence of the highly non-smooth and volatile trend of the load time series specially in building levels, its STLF is even a more complex procedure than that of a power system. For this purpose, in this paper we proposed a new prediction model based on a new feature selection algorithm and hybrid forecast engine of enhanced version of empirical mode decomposition named sliding window EMD bundled with an intelligent algorithm. The proposed forecast engine is combined with novel shark smell optimization to increase the prediction accuracy. All weights of this forecast engine have been optimized with an intelligent algorithm to find better prediction results. Effectiveness of the proposed model is carried out to real-world engineering test case in comparison with other prediction models.