Optimal Siting and Sizing of EV Charging Stations in Distribution Networks: A Comparative Study of PSO, GWO, and HPSOGWO

Abstract

The rapid proliferation of Electric Vehicles (EVs) presents a critical challenge for existing power infrastructure, specifically concerning the integration of EV Charging Stations (EVCSs) into radial distribution networks. Uncoordinated EVCS placement significantly degrades network stability, exacerbating active power losses and inducing severe voltage deviations. This study addresses the optimal siting and sizing of EVCSs in a 30-bus radial distribution network by proposing a hybrid metaheuristic approach: the Hybrid Particle Swarm Optimization-Grey Wolf Optimizer (HPSOGWO). The optimization problem is formulated as a multi-objective task that simultaneously minimizes the total infrastructure investment cost, reduces active and reactive power losses, and ensures strict adherence to voltage profile limits. Unlike previous static models, this research introduces a mandatory 100-charger target demand constraint, forcing the algorithm to distribute load dynamically across candidate buses. The inclusion of quadratic penalty multipliers ( ) ensures that all configurations strictly adhere to the  p.u. nodal voltage operational limits. Simulation results across nine distinct weighting configurations (w₁, w₂, w₃) demonstrate that HPSOGWO effectively overcomes the premature convergence limitations of standard PSO and the late-stage stagnation observed in GWO. The hybrid approach achieves superior convergence performance, maintaining stable active power losses (18.075–18.079 MW) across all test cases. These findings establish HPSOGWO as a highly reliable and computationally efficient planning tool for modern distribution system operators, successfully balancing economic investment with the imperative requirements of grid technical stability and power system reliability in increasingly complex charging environments.