Stochastic EV Charging Dispatch in Unbalanced Three-phase Networks Based on Interpretable Fuzzy Representation of User Preferences

Electric Vehicle (EV) charging dispatch can effectively coordinate EV charging demand to mitigate power unbalances and bus voltage violations in distribution networks. However, the existing EV charging dispatch models neglect users' fuzzy and uncertain preferences regarding the features of different charging stations and mainly operate in three-phase balanced distribution networks. Hence, a novel EV charging dispatch scheme is proposed to achieve EV charging management in an unbalanced three-phase network with the consideration of users' preferences. This scheme deploys linguistic term sets to capture imprecise preferences in an easily understandable way for ordinary users. A preference-constrained stochastic optimization model is then proposed for day-ahead EV charging dispatch, achieving an optimal three-phase EV charging schedule to enhance PV utilization while considering user preferences. Additionally, a real-time stage EV charging dispatch model is proposed to adapt to changes in charging requests and user preferences during operation. Case studies indicate that the proposed method enables EVs to support the operation of an unbalanced three-phase network and meet users' charging demands and fuzzy preferences. © 2024 IEEE.