arrow
Return

A New Binary Encoding Method for Energy Consumption Patterns Quantification

delete2024-01-01
delete0
PRE
AI
H
Hongliang Fang
J
Jiang‐Wen Xiao *
王燕舞 cover
王燕舞 (Yan‐Wu Wang)
C
C. Y. Chung
DOI:10.1109/TIM.2024.3370803delete
deleteOriginal
deleteShare
deleteSave
Abstract

Abstract

En 中文
Extracting users' energy consumption patterns (ECPs) from smart meter data is an important work for retailers. The existing literature usually describe these patterns by clustering the daily load curves (DLCs), but lack a clear and quantified representation to explain what the exact schema of a user is. Therefore, this article proposes a new binary encoding method for ECPs quantification. Specifically, first, both time and value intervals are divided for dimensionality reduction based on the similarity of adjacent timestamp loads. Then, a binary aggregate approximation (BAX) method is proposed to encode each DLC into a BAX word, and the BAX words of users are merged to obtain the schemas with a three-element alphabet. Finally, based on the schemas, the stability scores of users' patterns are quantified and are used to select target users for demand response (DR) measures. Case studies on a real dataset with 5566 users show that each target user averagely contributes to 0.172% of peak reduction, while each unselected user only contributes to 0.026%. Furthermore, to obtain target schemas and to find new users in future DR measures, a $K$ -means symbolic algorithm is designed to cluster BAX words of target users. The proposed encoding method and the findings can provide guidance of finding typical target users for DR measures.
Keywords:
Binary aggregate approximation (BAX)
dimension reduction
energy consumption pattern (ECP)
K-means symbolic algorithm
stability quantification

Journal

IEEE Transactions on Instrumentation and Measurement cover
IEEE Transactions on Instrumentation and Measurement
IF:
5.9
Papers:
2.0W
Citations:
7.2W

Organization

H
Hong Kong Polytechnic University
Scholars:
2.0K
Papers: 999
Citations: 4.1W
Cited Papers

Cited Papers

errShare
errSave
An Online Learning Framework for Targeting Demand Response Customers
err2022-01-01
err17
PREAI
errSchneider, Ian; Roozbehani, Mardavij; Dahleh, Munther
errShare
errSave
Detecting Energy Theft in Different Regions Based on Convolutional and Joint Distribution Adaptation
err2023-01-01
err4
PREAI
errWang, Jiangzhao; Zhu, Yanqing; Gao, Yunpeng; Cai, Ziwen; Sun, Yichuang; Peng, Fenghua
errShare
errSave
Matching consumer segments to innovative utility business models
err2021-03-01
err28
PREAI
errHall, Stephen; Anable, Jillian; Hardy, Jeffrey; Workman, Mark; Mazur, Christoph; Matthews, Yvonne
errShare
errSave
A review on the deployment of demand response programs with multiple aspects coexistence over smart grid platform
err2022-07-01
err30
PREAI
errIbrahim, Charles; Mougharbel, Imad; Kanaan, Hadi Y.; Abou Daher, Nivine; Georges, Semaan; Saad, Maarouf
errShare
errSave
Decarbonization efforts hindered by China's slow progress on electricity market reforms
err2023-04-27
err17
PREAI
errYu, Yang; Wang, Jianxiao; Chen, Qixin; Urpelainen, Johannes; Ding, Qingguo; Liu, Shuo; Zhang, Bing
errShare
errSave
A systematic review of building electricity use profile models
err2023-02-01
err26
PREAI
errKang, Xuyuan; An, Jingjing; Yan, Da
errShare
errSave
Electric Load Clustering in Smart Grid: Methodologies, Applications, and Future Trends
err2021-01-01
err64
errOAAI
errSi, Caomingzhe; Xu, Shenglan; Wan, Can; Chen, Dawei; Cui, Wenkang; Zhao, Junhua
errShare
errSave
A Scalable Ensemble Approach to Forecast the Electricity Consumption of Households
err2023-01-01
err8
errOAAI
errBotman, Lola; Soenen, Jonas; Theodorakos, Konstantinos; Yurtman, Aras; Bekker, Jessa; Vanthournout, Koen; Blockeel, Hendrik; De Moor, Bart; Lago, Jesus
errShare
errSave
researcher View more