Integrated approach for optimal capacity and site of electric vehicle charging stations incorporated with photovoltaic systems and allocation of capacitor banks

Electric vehicles (EVs) offer a promising solution for reducing environmental pollution. However, the rapid expansion of electric vehicle charging stations (EVCSs) creates significant technical challenges for distribution networks, including increased power losses and voltage instability. Strategic EVCS planning is essential to mitigate these impacts. This paper presents a comprehensive 2-stage framework for optimally siting and sizing EVCSs with photovoltaic (PV) systems and capacitor banks (CBs) in distribution networks. The approach employs the Dandelion Optimizer (DO), a recent metaheuristic algorithm, to solve a multi-objective optimization problem. It is validated on the large-scale IEEE 118-bus distribution system, divided into 3 geographical zones based on land cost. In the first stage, DO determines optimal locations and capacities for EVCSs per zone (an EVCS with 50 kW fast chargers and another one with 100 kW fast chargers) and PV units by minimizing active and reactive power losses, average voltage deviation index, and EVCS land costs while maximizing the voltage stability index. The second stage optimally places and sizes CBs, either centralized per zone or distributed-based mitigation solutions, to ensure voltage regulation, minimizing total reactive compensation instead of land costs. Moreover, the framework considers time-varying loads and PV generation profiles using OpenDSS, incorporates Pareto-based multi-objective optimization to explore trade-offs among competing objectives, and accounts for zone-specific land costs and varying distribution of fast-charging connectors. Comparative analysis demonstrates DO's superior convergence and solution quality with respect to Jaya optimization, particle swarm optimization, marine predators algorithm, artificial gorilla troops optimizer, teaching-learning-based optimization, and ant lion optimizer. Sensitivity analysis further validates the framework's robustness under varying land cost scenarios, providing valuable insights for both distribution network operators and charging station operators.