Citywide Electric Vehicle Charging Demand Prediction Approach Considering Urban Region and Dynamic Influences

TL;DR

A learning approach for citywide electric vehicle charging demand prediction, named CityEVCP, which cluster service areas by the types and numbers of points of interest in the areas and develop attentive hypergraph networks accordingly to learn non-pairwise relationships in urban areas.

Abstract

Electric vehicle charging demand prediction is important for vacant charging pile recommendation and charging infrastructure planning, thus facilitating vehicle electrification and green energy development. The performance of previous spatio-temporal studies is still far from satisfactory nowadays because urban region attributes and multivariate temporal influences are not adequately taken into account. To tackle these issues, we propose a learning approach for citywide electric vehicle charging demand prediction, named CityEVCP. To learn non-pairwise relationships in urban areas, we cluster service areas by the types and numbers of points of interest in the areas and develop attentive hypergraph networks accordingly. Graph attention mechanisms are employed for information propagation between neighboring areas. Additionally, we propose a variable selection network to adaptively learn dynamic auxiliary information and improve the Transformer encoder utilizing gated mechanisms for fluctuating charging time-series data. Experiments on a citywide electric vehicle charging dataset demonstrate the performances of our proposed approach compared with a broad range of competing baselines. Furthermore, we demonstrate the impact of dynamic influences on prediction results in different areas of the city and the effectiveness of our area clustering method.

Figures (9)

• Figure 1: The impact of different factors on charging demand.
• Figure 2: The model structure of the proposed approach, which consists of (a). Area attributes fusion module, (b). Adjacency graph fusion module, and (c). Temporal feature fusion module.
• Figure 3: Schematic of some module structures. (a). Gated residual network, (b). Variable selection network and (c). Gated temporal attention.
• Figure 4: Detailed operation of the proposed variable selection network.
• Figure 5: Lineplots of 60-min prediction results and errors for two example areas.