[4G2-GS-7] Agents: Agents and machine learning

[4G2-GS-7-02] Analysis of Formation Process of Charging Reservation Behavior Type Composition of Electric Vehicle Users by Machine Learning

〇Mahiro Minowa¹, Hideaki Uchida¹, Hideki Fujii¹, Shinobu Yoshimura¹ (1. The University of Tokyo)

Keywords:Electric Vehicle, Charging Station, Reservation, Behavior Model, Machine Learning

With the spread of electric vehicles, congestion at the charging station is concerned. In order to improve the

congestion, the charging station reservation system has been proposed. However, there is little information on

its effectiveness because the system has been rarely introduced yet. In this paper, the authors aim at obtaining

knowledge regarding it. Three types of charging reservation behavior were dened based on a demonstration

experiment and the impact of the composition of these types was analyzed. The simulation results showed that

there is an optimum composition of charging reservation behavior types in a specic environment. In addition, we

proposed a learning model that adaptively changes the conguration of the charging reservation behavior type of

the EV, and analyzed the formation process of the charging reservation behavior type composition in a specic

environment.