Citation: Cen, S.; Yoo, J.H.; Lim, C.G.
Electricity Pattern Analysis by
Clustering Domestic Load Profiles
Using Discrete Wavelet Transform.
Energies 2022, 15, 1350. https://
doi.org/10.3390/en15041350
Academic Editors: Sergio
Nesmachnow and Islam
Safak Bayram
Received: 10 December 2021
Accepted: 11 February 2022
Published: 13 February 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Electricity Pattern Analysis by Clustering Domestic Load
Profiles Using Discrete Wavelet Transform
Senfeng Cen , Jae Hung Yoo and Chang Gyoon Lim *
Department of Computer Engineering, Chonnam National University, Yeosu 59626, Korea;
jasoncsf.7@gmail.com (S.C.); jhy@jnu.ac.kr (J.H.Y.)
* Correspondence: cglim@jnu.ac.kr; Tel.: +82-61-659-7254
Abstract:
Energy demand has grown explosively in recent years, leading to increased attention of
energy efficiency (EE) research. Demand response (DR) programs were designed to help power
management entities meet energy balance and change end-user electricity usage. Advanced real-time
meters (RTM) collect a large amount of fine-granular electric consumption data, which contain
valuable information. Understanding the energy consumption patterns for different end users can
support demand side management (DSM). This study proposed clustering algorithms to segment
consumers and obtain the representative load patterns based on diurnal load profiles. First, the
proposed method uses discrete wavelet transform (DWT) to extract features from daily electricity con-
sumption data. Second, the extracted features are reconstructed using a statistical method, combined
with Pearson’s correlation coefficient and principal component analysis (PCA) for dimensionality
reduction. Lastly, three clustering algorithms are employed to segment daily load curves and select
the most appropriate algorithm. We experimented our method on the Manhattan dataset and the
results indicated that clustering algorithms, combined with discrete wavelet transform, improve the
clustering performance. Additionally, we discussed the clustering result and load pattern analysis of
the dataset with respect to the electricity pattern.
Keywords:
demand response; discrete wavelet transform; Pearson’s correlation coefficient; principal
component analysis; clustering
1. Introduction
Smart grid technologies and applications capable of adaptive, resilient, and sustain-
able self-healing, with foresight for prediction under different uncertainties, improve the
reliability of the power system [
1
]. Furthermore, the smart grid allows bidirectional com-
munication that supports the demand response (DR) programs [
2
]. Demand response
technologies are widely applied and are constantly improving. The most common DR
programs can be categorized into the following two classes: price-based programs and
incentive-based programs. Price-based programs contain time of use (ToU), real time
pricing (RTP) and critical peak pricing (CPP), which aim to motivate the end-user to change
their consumption behavior [
3
]. On the other hand, incentive-based programs reach a
consensus with consumers to reduce electricity consumption. Examples of these schemes
are direct-load control (DLC), interruptible/curtailable service (I/C), demand bidding/buy
(DB), etc. [
4
]. Considering various end-user consumption behaviors, it required the utility
companies to design reasonable strategies. Therefore, it is necessary to analyze end-users’
consumption data to acquire the load patterns.
Advanced metering infrastructure (AMI) and smart meters have been adopted to
automatically collect energy consumption data at a fine granular interval, which is usually
in intervals of 1 h, 30 min, or even 30 s [
5
]. Most countries have vigorously deployed smart
meters because of the potential value of consumption data [
6
]. The massive amount of data
sampled by smart meters could be used for research, typically load forecasting, customer
segmentation, pricing/incentive mechanism, scheduling and control [7].
Energies 2022, 15, 1350. https://doi.org/10.3390/en15041350 https://www.mdpi.com/journal/energies
