Optimal Planning of EV Charging Stations Using Bilevel Optimization Model

Abstract

The Electric vehicles (EVs) exponential rise in urban markets poses a number of challenges, such as the need for suitable infrastructure for charging, the best locations for charging stations, and effective scheduling of charging activities. Increasing EV load substantially affects distribution networks, environmental sustainability, and economic factors. The charging and discharging patterns of EVs are compatible with demand response (DR) programs as EV loads are flexible, controllable, and can also act as distributed energy storage. An optimization-based control strategy enables dynamic adjustment and management of EV charging and discharging profiles during DR events. This paper introduces a novel bilevel optimization framework designed to effectively balance infrastructure planning, grid reliability, DR participation, economic benefits, and user experience. The Elephant Herd Optimization (EHO) algorithm is proposed to determine the optimal size, and charging profile during DR events. The numerical results obtained from the simulation show that the optimal location and sizing, along with the application of DR programs, result in a reduction of losses and total costs. The proposed EHO algorithm achieves 30.5% peak demand reduction, $149 daily DR revenue, and 54% reduction in EV user wait times, demonstrating significant improvements over uncontrolled charging scenarios. This work promotes environmental sustainability by optimizing EV charging infrastructure and DR through advanced planning and coordination, thereby reducing grid stress, enhancing energy efficiency, and supporting the large-scale integration of low-carbon electric mobility.