Tai-Yu MA, Luxembourg Institute of Socio-Economic Research, Luxembourg
Sébastien FAYE, Luxembourg Institute of Science and Technology , Luxembourg
Public charging station occupancy prediction plays key importance in developing a smart charging strategy to reduce electric vehicle (EV) operator and user inconvenience. However, existing studies are mainly based on conventional econometric or time series methodologies with limited accuracy. In this study, we propose a new methodology based on hybrid long short-term memory neural network to predict public EV charging station occupancy. Different factors influencing charging station occupancy profiles are investigated. Given limited fields available in the datasets, we successfully generate and incorporate both historical charging state sequences and time-related features for much accurate multistep discrete charging occupancy state prediction. The model is compared to a number of state-of-the-art machine learning and deep learning approaches based on the EV charging data obtained from the open data portal of the city of Dundee, UK. The results show that the proposed method produces very accurate predictions (99.99% and 81.87% for 1 step (10 minutes) and 6 steps (1 hour) ahead, respectively, and outperforms the benchmark approaches significantly (+22.4% for one-step-ahead prediction and +6.2% for 6 steps ahead). The results show a strong potential for the improvement of charging station occupancy prediction methods allowing EV-based mobility service operators to develop smart-charging scheduling strategies.
Mots clés : electric vehicle|forecasting|Long short-term memory|charging occupancy |time series
A102382TM