A Proxy Electricity Purchase Prediction Method Based on User Profile Analysis and Data Augmentation

View at publisher Summarise with VIRAT
Authors: Dan Wu, Wanru Zhao, Ting Lei, Fangzheng Sheng, Sizhuang Chen
Cited by: 0
References: 9

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

With the reform of the electricity market, the proxy electricity purchase mechanism has become an important way to optimize power resource allocation. This paper proposes a novel prediction method based on user profile analysis and data augmentation to address the electricity consumption forecasting problem in proxy electricity purchase scenarios. First, user profiles are constructed by analyzing proxy users' basic attributes, electricity consumption behavior characteristics, and sensitivity to external factors, enabling multi-dimensional user classification. Subsequently, to mitigate the common challenges of limited and low-quality historical electricity consumption data among proxy users, the TimeGAN model is employed to perform data augmentation sequentially for each user group. Finally, dedicated electricity consumption prediction models are developed for each user category to achieve accurate proxy electricity purchase forecasting. Experimental results demonstrate that the proposed method significantly improves both the accuracy and stability of proxy electricity purchase predictions after user profile analysis and data augmentation, providing reliable technical support for the stable and economic operation of power systems.

Fields of Study & Subjects

Energy Load and Power Forecasting Smart Grid Energy Management Electric Vehicles and Infrastructure

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
1748–1752
Publication date
28 January 2026
Language
EN