.png){kind=icon}
返回
Day-ahead bus load forecasting method based on fully connected spatial-temporal graph attention network
.png){kind=icon}
PRE
AI
DOI：10.1016/j.epsr.2024.111294.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Bus load forecasting is significant to operating electric power systems and renewable energy consumption. Aiming at the problem of day-ahead bus load forecasting, this paper proposes a day-ahead bus load forecasting method based on the local and global graph attention modules. The local and global graph attention modules are composed of the dynamic graph attention networks and ProbSparse self-attention layers, respectively. Firstly, we use the Pearson correlation coefficient to analyze the correlation between the bus load and meteorological features and the correlation among loads on buses within the same supply area. Then, the fully connected spatialtemporal graph, according to the supply area's single line diagram, is constructed, including the spatial correlations between buses within the same supply area and the temporal correlations within the bus load series. The local and global spatial-temporal correlations in the fully connected spatial-temporal graph are captured by the local and global graph attention modules, respectively. In this way, our model captures the information in the mutually coupled spatial-temporal correlations and models the genuine spatial and temporal changes in reality. The experiment results show that the proposed method has higher forecasting accuracy than the comparative baseline models.
Keyword:
Dynamic graph attention network
Probsparse self-attention
Day-ahead bus load forecasting
Fully connected spatial-temporal graph
Graph neural network
期刊
IF:
4.2
论文数:
1.2W
被引数:
2.2W
机构
引用论文
A Multi-Scale Spatial-Temporal Graph Neural Network-Based Method of Multienergy Load Forecasting in Integrated Energy System基于多尺度时空图神经网络的综合能源系统多能负荷预测方法
PREAI
分享收藏Spatial-Temporal Residential Short-Term Load Forecasting via Graph Neural Networks基于图神经网络的时空住宅短期负荷预测
PREAI
分享收藏TCAMS-Trans: Efficient temporal-channel attention multi-scale transformer for net load forecasting
PREAI
分享收藏分享收藏The impact of heat pumps on day-ahead energy community load forecasting热泵对日前能源社区负荷预测的影响.png)
APPLIED ENERGY
OAAI
APPLIED ENERGYIF11
OAAI分享收藏
查看更多内容