Spatio-Temporal Graph Convolutional Networks for EV Charging Demand Forecasting Using Real-World Multi-Modal Data Integration

2025-10-10

AI Summary

To address grid stress and planning challenges arising from spatiotemporally imbalanced electric vehicle (EV) charging demand, this paper proposes TW-GCN—a hybrid spatiotemporal forecasting framework integrating Graph Convolutional Networks (GCNs) and 1D Convolutional Neural Networks (1D-CNNs). It jointly models traffic flow, meteorological conditions, and charging behavior as multi-source drivers. Innovatively, regional topology is encoded as a spatial graph structure, while temporal dynamics are explicitly coupled to achieve balanced stability and responsiveness in medium- to long-term predictions. Experiments on real-world data from Tennessee demonstrate that TW-GCN achieves state-of-the-art 3-hour-ahead prediction accuracy; the 1D-CNN component consistently outperforms alternatives across diverse configurations; and region-specific analyses align closely with empirical charging demand patterns. The framework thus enables precise EV charging infrastructure deployment and enhances grid resilience planning.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Transportation remains a major contributor to greenhouse gas emissions, highlighting the urgency of transitioning toward sustainable alternatives such as electric vehicles (EVs). Yet, uneven spatial distribution and irregular utilization of charging infrastructure create challenges for both power grid stability and investment planning. This study introduces TW-GCN, a spatio-temporal forecasting framework that combines Graph Convolutional Networks with temporal architectures to predict EV charging demand in Tennessee, United States (U.S.). We utilize real-world traffic flows, weather conditions, and proprietary data provided by one of the largest EV infrastructure company in the U.S. to capture both spatial dependencies and temporal dynamics. Extensive experiments across varying lag horizons, clustering strategies, and sequence lengths reveal that mid-horizon (3-hour) forecasts achieve the best balance between responsiveness and stability, with 1DCNN consistently outperforming other temporal models. Regional analysis shows disparities in predictive accuracy across East, Middle, and West Tennessee, reflecting how station density, population, and local demand variability shape model performance. The proposed TW-GCN framework advances the integration of data-driven intelligence into EV infrastructure planning, supporting both sustainable mobility transitions and resilient grid management.

Problem

• Forecasting EV charging demand using spatio-temporal graph networks
• Addressing uneven charging infrastructure distribution and utilization
• Integrating multi-modal real-world data for grid stability planning

Innovation

• Spatio-temporal graph convolutional networks for EV demand
• Integrates real-world traffic, weather, and proprietary data
• Mid-horizon forecasts balance responsiveness and stability