基于图注意力网络-门控循环单元的多源异构特征融合电动汽车充电负荷预测方法

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (979 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract To mitigate the decline in prediction accuracy caused by the limited diversity of input features and the insufficient extraction of spatiotemporal correlations in existing electric vehicle (EV) charging load forecasting models, a novel spatiotemporal forecasting framework is proposed, in which a graph attention network (GAT) is integrated with a gated recurrent unit (GRU) to effectively capture complex spatial and temporal dependencies. The model constructs a graph structure based on geographical proximity, incorporating diverse features such as historical load, weather conditions, calendar dates, and holidays. A multi-head GAT is employed to extract spatial dependencies, while the GRU models temporal dynamics. The final prediction is generated through a fully connected layer. Experimental results demonstrate that the proposed method significantly outperforms traditional methods and mainstream deep learning approaches in terms of forecasting accuracy, while also maintaining robust performance in cross-regional scenarios. This method offers data support for the dynamic scheduling of urban power grids and the orderly management of EV charging.

Key words： electric vehicle load forecasting graph neural network spatiotemporal correlation features attention mechanism

Received: 26 June 2025

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	WEN Changbao
	WU Benhuang
	SUN Jieru

Cite this article:

WEN Changbao,WU Benhuang,SUN Jieru. A graph attention network-gated recurrent unit based method for electric vehicle charging load prediction with multi-source heterogeneous feature integration[J]. Electrical Engineering, 2026, 27(1): 1-8.

URL:

https://dqjs.cesmedia.cn/EN/Y2026/V27/I1/1

[1] 郭剑锋, 张雪美, 曹琪, 等. 电动汽车助力我国能源安全与“碳达峰、碳中和”协同推进[J]. 中国科学院院刊, 2024, 39(2): 397-407.
[2] 黄学良, 刘永东, 沈斐, 等. 电动汽车与电网互动: 综述与展望[J]. 电力系统自动化, 2024, 48(7): 3-23.
[3] 陈中, 万玲玲, 张梓麒. 面向共享电动汽车的用户助推与充电协同调度[J]. 电工技术学报, 2025, 40(11): 3572-3590.
[4] 高云, 解超, 张高航, 等. 计及电动汽车接入的园区综合能源系统低碳协调运行调度策略[J]. 电气技术, 2025, 26(2): 14-25, 34.
[5] 王海鑫, 袁佳慧, 陈哲, 等. 智慧城市车-站-网一体化运行关键技术研究综述及展望[J]. 电工技术学报, 2022, 37(1): 112-132.
[6] 吴奇珂, 程帆, 陈昕儒. “电能替代”战略中电动汽车的推广潜力及经济性分析[J]. 电气技术, 2016, 17(9): 88-92, 97.
[7] 刘动, 孟晨旭, 潘正阳, 等. 居民小区经营性电动汽车充电站投资建设研究[J]. 电气技术, 2022, 23(6): 104-108.
[8] 张晨彧, 丁明, 张晶晶. 基于交通出行矩阵的私家车充电负荷时空分布预测[J]. 电工技术学报, 2017, 32(1): 78-87.
[9] RASHID M, ELFOULY T, CHEN Nan.A com- prehensive survey of electric vehicle charging demand forecasting techniques[J]. IEEE Open Journal of Vehicular Technology, 2024, 5: 1348-1373.
[10] QU Haohao, KUANG Haoxuan, WANG Qiuxuan, et al.A physics-informed and attention-based graph learning approach for regional electric vehicle charging demand prediction[J]. IEEE Transactions on Intelligent Trans- portation Systems, 2024, 25(10): 14284-14297.
[11] 马帅, 刘建伟, 左信. 图神经网络综述[J]. 计算机研究与发展, 2022, 59(1): 47-80.
[12] 王健宗, 孔令炜, 黄章成, 等. 图神经网络综述[J]. 计算机工程, 2021, 47(4): 1-12.
[13] 庞昊, 高金峰, 杜耀恒. 基于多神经网络融合的短期负荷预测方法[J]. 电力自动化设备, 2020, 40(6): 37-43.
[14] LUO Ruikang, ZHANG Yicheng, ZHOU Yao, et al.Deep learning approach for long-term prediction of electric vehicle (EV) charging station availability[C]//2021 IEEE International Intelligent Transportation Systems Conference (ITSC), Indianapolis, IN, USA, 2021: 3334-3339.
[15] 胡博, 张鹏飞, 黄恩泽, 等. 基于图WaveNet的电动汽车充电负荷预测[J]. 电力系统自动化, 2022, 46(16): 207-213.
[16] BHAT F A, TIWARI G Y, VERMA A.Preferences for public electric vehicle charging infrastructure locations: a discrete choice analysis[J]. Transport Policy, 2024, 149: 177-197.