Genetic Algorithm Approach based on Graph Theory for Location Optimization of Electric Vehicle Charging Stations

Abstract

Nowadays, with the increase in the economic and ecological disadvantages of internal combustion engine vehicles, the use of electric vehicles is becoming widespread. In addition, the widespread use of electric vehicles creates the need for optimal locating electric vehicle charging stations in urban areas. In this study, a Genetic Algorithm approach based on graph theory is developed to locate a certain number of charging stations in urban areas. In the proposed method, some points in the urban area are accepted as reference nodes. By considering the path relation and distance information between these reference nodes, the urban area is expressed with a weighted graph. The minimum cost paths between the reference points in the urban area expressed by graph were determined by the Dijkstra method. The obtained minimum cost paths are used in the fitness function of the GA. Then, a GA-based method is developed to distribute the charging stations homogeneously at the urban area. The proposed method is tested on a virtual urban area with 20 reference nodes. The obtained results showed that the proposed method is a suitable method for distance-based charging station location determination. © 2021 IEEE.