Bi-level programming model of location and charging electric vehicle stations in distribution networks

• Proposing a bi-level model for EV station planning • Considering the ACOPF constraints to prevent any unreal power transaction. • Dealing with a low computational burden approach for operation and planning. • Analyzing the impact of EV stations on loss minimization in distribution networks. Electric vehicles (EVs) are currently a focal point in distribution network research due to their potential benefits. Integrating EVs into distribution networks can enhance voltage profile stability. However, unplanned EV connection and charging can lead to increased peak load power consumption, voltage profile irregularities, and heightened network losses. Addressing the placement of charging stations in distribution networks is crucial to prevent voltage drop crises and mitigate other issues. This study proposes a Bi-level particle programming optimization programming approach, utilizing the Particle Swarm Optimization (PSO) algorithm to minimize losses and determine optimal station locations. The algorithm iteratively minimizes losses while considering location and operating costs until a stopping condition is met. Additionally, an EV charging planning algorithm is introduced to manage EV connections without exacerbating peak load or causing severe voltage drops. Simulations conducted on an IEEE 33-bus system using MATLAB analyze the proposed design's impact on power consumption, system losses, and network voltage. Simulations in an IEEE 33-bus system show that the proposed model shifts peak power consumption by about 4.6% to about 11.6% under different EV penetration scenarios. It achieves a reduction in the network losses by up to 0.39%, while the voltage deviations in critical buses are minimized to enhance stability during periods of high demand for EVs. Results demonstrate that the programmed charging algorithm effectively reduces power consumption at peak load, minimizes network losses, and smooths the daily voltage profile.