A Graph-Based Approach for Optimal Sizing and Siting of Electric Vehicle Charging Stations in Distribution Networks

Abstract

The rapid global adoption of electric vehicles (EVs) is introducing substantial new challenges for distribution system operators (DSOs), particularly in accommodating additional charging demand within existing distribution networks. This paper proposes a two-stage methodological framework for the optimal siting and sizing of electric vehicle charging stations (EVCSs), where siting is expressed in terms of bus identification and sizing in terms of feasible power-demand increments. The proposed method is grounded in a graphbased representation of the distribution system, in which buses and branches are modeled as nodes and weighted edges, respectively, thereby enabling the explicit incorporation of network topology, electrical distance, and spatial separation into the planning process. From a theoretical perspective, the framework combines graph-theoretic network modeling with mixed-integer linear programming to identify candidate buses, followed by a binary search procedure to determine the maximum feasible load increment at each selected bus subject to operational constraints. The proposed approach is applied to the IEEE 123-bus system and the 533-bus system using MATLAB. Results show the effectiveness of the proposed approach. The results show total feasible EVCS capacities of 277 kW and 5465 kW, respectively. Compared with the baseline cases, the method accommodates these additional EVCS loads while keeping the expanded systems feasible; total real-power losses increase from 186.4 kW to 224.1 kW in the IEEE 123-bus system and from 175.1 kW to 555.2 kW in the 533-bus system, illustrating the additional loading impact captured by the proposed planning framework.