Multi-period planning of locations and capacities of public charging stations

Appropriate charging resources allocation is critical to ensure charging convenience and charging station operation efficiency. However, the temporality of electric vehicle penetration, the development of chargingrelated technologies, and the randomness of charging behaviors bring highly spatiotemporal dynamics to the charging demands distribution in cities. In this paper, the multi-period and multi-scenario spatiotemporal distribution of charging demands is evaluated based on real-world operation data of electric vehicles in Beijing. A three-period charging stations locations and capacities planning model is proposed to deploy charging stations reasonably based on high-resolution spatiotemporal charging demands distribution at a spatial resolution of 0.46 km side length hexagon units and time resolution of 15 min to satisfy dynamic multi-period charging demands. The model takes minimizing the total costs of charging stations and electric vehicles during all the planning periods as the optimization objective. The capacity-constrained M/M/c/N charging queuing theory combined with the sensitivity analysis and optimization of the charging arrival rate is introduced into the capacity designing process to determine the corresponding charging pile quantity reasonably. Suggestions are given on the charging stations construction locations and the corresponding configurations of types and quantities of charging piles during different planning periods, an actual case study in the Haidian district, Beijing is conducted to validate the proposed planning models.