Location-routing optimization problem for electric vehicle charging stations in an uncertain transportation network: An adaptive co-evolutionary clustering algorithm

Abstract

The rapid growth of Electric Vehicles (EVs) has led to issues such as insufficient and unevenly distributed charging stations, posing an increasingly severe challenge. This not only means higher economic and time costs for EV users traveling to charging stations, but also indicates that EVs need to incur additional energy consumption. To address these challenges, this paper first establishes an extended space-time-state network to describe the temporal variability and uncertainty of the traffic environment. Secondly, considering the impact of the energy consumption incurred by EVs traveling to charging stations (CSs) on the layout of CSs, a co-evolutionary optimization model of the location-routing problem is proposed. To solve the model, a two-stage Adaptive Co-evolutionary Clustering Algorithm (ACECA) integrating an adaptive clustering framework and a co-evolutionary mechanism is designed. Finally, the experiment results indicate that there exists a game between a short-term economic investment and long-term benefits of energy conservation and emission reduction. Moreover, ACECA demonstrates stronger solution performance and robustness compared to other algorithms, with the resulting CS location schemes showing superior advantages in energy conservation and in cost saving for users. The research findings can provide theoretical support for the planning of charging station locations for electric vehicles.