Energy demand pattern analysis in South Korea using hidden Markov model‐based classification

• Published date: 01 September 2024
• Author: Jaeyong Lee,Beom Seuk Hwang
• Date: 01 September 2024
• DOI: http://doi.org/10.1111/asej.12338

ORIGINAL ARTICLE

Energy demand pattern analysis in South Korea using

hidden Markov model-based classification

Jaeyong Lee | Beom Seuk Hwang

Department of Applied Statistics, Chung-Ang University, Seoul, Korea

Correspondence

Beom Seuk Hwang, Department of Applied Statistics, Chung-Ang University, 84 Heukseok-ro, Dongjak-gu,

Seoul, 06974, Korea.

Email: bshwang@cau.ac.kr

Funding information

National Research Foundation of Korea, Grant/Award Number: NRF-2019R1C1C1011710; Korea Institute of

Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry \& Energy (MOTIE)

of the Republic of Korea, Grant/Award Number: 20199710100060

Abstract

Understanding energy demand patterns in the residential sector is crucial for improving

energy efficiency through demand-side management. Load curve classification is a useful

method for analyzing energy demand patterns. In this paper, we employ a hidden

Markov model (HMM)-based classification to residential load curves in South Korea.

We also investigate how the number of hidden states affects classification performance

by allowing HMM to train with a different number of hidden states for each class. We

compare our HMM-based method with several state-of-the-art models and find that it

outperforms other competing models in multiple datasets. Additionally, we use the fitted

HMM model to make inferences about the load curves, gaining deeper insights into

energy demand patterns.

KEYWORDS

energy demand pattern, hidden Markov model, residential load curve, time series classification

JEL CLASSIFICATION

C15, C22, Q41

[Correction added on 22 September 2025, after first online publication: The copyright line was changed.]

DOI: 10.1111/asej.12338

© 2024 East Asian Economic Association and John Wiley & Sons Australia, Ltd

Received: 13 May 2023; Accepted: 29 June 2024

404 Asian Econ. J. 2024;38:404–428.

wileyonlinelibrary.com/journal/asej

1|INTRODUCTION

Understanding demand patterns is fundamental in numerous economic

contexts, from resource allocation to policy formulation. Energy demand, par-

ticularly in the residential sector, is a significant component of this framework,

accounting for up to half of the total energy demand in many countries. Analyz-

ing these patterns yields crucial insights not only for energy producers and

consumers but also for economic planners and policymakers aiming to enhance

efficiency and sustainability. Given the recent advancements in demand-side

management (DSM) strategies, the characterization of demand patterns has

gained heightened importance. The primary objective of DSM is to provide

cost-effective and reliable energy to customers by optimizing electricity usage at

the user end, yielding benefits for both consumers and producers (Premkumar

et al., 2022). Moreover, energy demand pattern analysis influences housing and

urban development by informing infrastructure planning and supporting smart

city initiatives, promoting energy-efficient housing to reduce overall consump-

tion and costs. In energy pricing and subsidies, it aids in designing fair tariff

structures and tailoring subsidies for energy-efficient appliances. Environmental

policies benefit from emission reduction efforts and energy conservation

programs, promoting renewable energy and efficient technologies in homes.

Since understanding demand patterns is essential in implementing effective

DSM, there have been various attempts at characterizing load curves.

Balasubramanian and Balachandra (2021) used a simulation-based clustering of

load curves and demonstrated its effectiveness. Within the residential sector,

home appliances contribute significantly to energy consumption. According to

Firth et al. (2008), the quantity and types of these appliances within a household

exert a notable influence on electricity demand. Several studies have focused on

investigating the effect of seasonal appliances, such as air conditioners,

on household electricity consumption. Hu et al. (2019) suggested that the use of

air conditioners or other weather-related devices is on an increasing trend. As

such, seasonal home appliances can be useful in understanding energy demand

patterns of the residential sector.

In this study, we classify the electricity load curves of South Korean house-

holds based on the types and quantities of home appliances. Four types of home

appliances are considered: seasonal, video/audio, kitchen, and general devices.

A survey was conducted to gather information on the specific devices owned for

each of the four types. We aggregated the number of devices for each type and

used them to analyze the electricity load curves of each household. Given that

energy demand is likely to be lower for households with fewer electronic devices

and higher for those with more, we divided the distribution of the number of

possessions into two, creating two distinct classes for our analysis. This classifi-

cation provides insights into household energy behavior, which is vital for devel-

oping targeted DSM strategies and economic policies.

ASIAN ECONOMIC JOURNAL 405