Socio-demographic Information Extraction from Load Profile Using Convolutional Neural Network

DOI

10.2991/978-94-6463-010-7_72How to use a DOI?

Keywords

Convolutional Neural Network (CNN); Deep Learning; One-Dimensional Convolution; Two-Dimensional Convolution; Socio-Demographic Information; Smart Meter

Abstract

Reasonable estimation of socio-demographic information by using smart meter data is the application direction of future load profile user behavior analysis. The full mining of socio-demographic information has attracted more and more attention because the socio-demographic characteristics of consumers can help energy suppliers provide consumers with personalized services, thereby gaining an advantage in business competition. Nowadays, the simplicity of the current feature extraction methods has the information content of smart meter data not fully excavated, which leads to the low accuracy of the training model. This paper uses deep learning methods to infer the possibility of household socio-demographic characteristics from consumers’ electricity smart meter data. A deep convolutional neural network (CNN) uses different feature extraction methods of one-dimensional convolution and two-dimensional convolution respectively, and some measures are used to prevent the model from overfitting. After a lot of repeated experiments, our model has a stronger identification ability than other previous models. That’s because different feature extraction methods can better decompose consumers’ heterogeneous electricity consumption behavior. Finally, we compare and discuss the results, thus supporting the modeling of users’ electricity consumption behavior and the design of a customized demand management strategy.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yibo Wang
AU  - Qian Wang
AU  - Zhengrun Wu
AU  - Bing Zhu
PY  - 2022
DA  - 2022/12/02
TI  - Socio-demographic Information Extraction from Load Profile Using Convolutional Neural Network
BT  - Proceedings of the 2022 International Conference on Artificial Intelligence, Internet and Digital Economy (ICAID 2022)
PB  - Atlantis Press
SP  - 703
EP  - 715
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-010-7_72
DO  - 10.2991/978-94-6463-010-7_72
ID  - Wang2022
ER  -