Data-Driven Mobility on Demand Fleets Charging Demand Modeling

With the increasing concern about the energy security and environment pollution, the development of electric vehicles (EVs) has been paid more attention. At present, the large-scale access of EVs will aggravate the load peak-valley difference and make the load power fluctuate greatly, which brings challenges to the stability of the power grid. Therefore, it is necessary to effectively predict the charging demand distribution. In this paper, we propose data-driven mobility on charging demand modeling. Considering the travel time and travel location, we first establish the historical travel rule utilizing the data mining and clustering method. After that, the travel route model is designed to determine the shortest the travel distance. Besides, the charging demand model with different driving characteristics of vehicles is provided to simulate the charging behavior. Finally, a case study with multiple date type scenarios is provided to verify the validity of the proposed model based on the actual travel data from Didi data center.