AST-GIN: Attribute-Augmented Spatiotemporal Graph Informer Network for Electric Vehicle Charging Station Availability Forecasting

Abstract

Electric Vehicle (EV) charging demand and charging station availability forecasting is one of the challenges in the intelligent transportation system. With accurate EV station availability prediction, suitable charging behaviors can be scheduled in advance to relieve range anxiety. Many existing deep learning methods have been proposed to address this issue; however, due to the complex road network structure and complex external factors, such as points of interest (POIs) and weather effects, many commonly used algorithms can only extract the historical usage information and do not consider the comprehensive influence of external factors. To enhance the prediction accuracy and interpretability, the Attribute-Augmented Spatiotemporal Graph Informer (AST-GIN) structure is proposed in this study by combining the Graph Convolutional Network (GCN) layer and the Informer layer to extract both the external and internal spatiotemporal dependence of relevant transportation data. The external factors are modeled as dynamic attributes by the attributeaugmented encoder for training. The AST-GIN model was tested on the data collected in Dundee City, and the experimental results showed the effectiveness of our model considering external factors’ influence on various horizon settings compared with other baselines.

Comparison of Small EV Charging Station's Load Forecasts and it's Impact on the Operational Costs

“…Existing research on charging station (CS) forecasting can be categorized based on the algorithms used and in terms of their focus. Predicted parameters include occupancy [2]- [5], load demand and parking duration of an individual charging session [6]- [8]. The following focuses on work that develops forecasts of the load of CSs.…”

“…The following focuses on work that develops forecasts of the load of CSs. The range of studies predicting charging power varies from entire regions [4], [5], [9] or entire cities [5], [10] to individual charging stations. Kim et al [11] compare predictions for these three observation levels and conclude that predictions at the city and regional levels provide good results, while predictions at the charging station level have room for improvement.…”

Deep Learning Based Optimization Model for Energy Consumption of New Electric Vehicles

“…The (AST-GIN) structure developed by Luo et al [21] enhances prediction accuracy and interpretability by transport data. During training, attribute augmented encoders model the environmental effects as a set of varying attributes.…”

Feature Papers in Vehicular Sensing

“…The development and deployment of electric vehicles involve several considerations relating to conventional vehicles and traffic, and some uncertainties are proposed in [5,6]. For example, electromagnetic fields present problems that are specific to electric vehicles.…”

“…To enhance prediction accuracy and interpretability, Ref. [6] proposes an Attribute-Augmented Spatiotemporal Graph Informer structure by combining the Graph Convolutional Network layer and the Informer layer to extract both the external and internal spatiotemporal dependence of relevant transportation data.…”