Spatial connection cost minimization of EV fast charging stations in electric distribution networks using local search and graph theory

Abstract

Fast charging stations for electric vehicles require a high-power demand, meaning that electricity distribution companies must define the connection locations within the distribution network to guarantee adequate power supply levels. Due to electric vehicle users' driving patterns and equipment's high costs, these stations must be concentrated in certain regions. This paper presents a methodology for assisting electricity distribution companies in identifying candidate connection points for fast charging stations to reduce new installations and network reinforcement investments. First, possible connection points are analyzed with graph theory to find the least costly connection; this strategy prioritizes the current network elements' unused capacity. As a second step, the electric distribution network is analyzed after fast-charging stations have been connected, evaluating the networks' operational limits. The methodology is applied in a Brazilian city combining spatial information with a realistic representation of the network and network total supply capability to connect new loads. Model outcomes are spatial maps that help identify suitable connection locations, determine new capacity values, and calculate the necessary investment. We compare the proposed methodology with other conventional approaches, demonstrating how the developed methodology can assist distribution companies in reducing overall investment and operational costs of fast charging stations for electric vehicles.