As nations accelerate electrification of transportation under carbon neutrality targets, EV charging infrastructure has become a cornerstone of energy transition planning. Integrated stations combining solar generation, battery storage, and fast-charging capabilities offer a solution, yet their optimal operation remains challenging. A key problem is scheduling the flow of power between solar panels, battery systems, and vehicle chargers to minimize both electricity procurement costs and disruptive grid impacts.
Traditional optimization methods struggle with this multi-objective problem. Conventional particle swarm optimization algorithms use random initialization that generates suboptimal starting solutions, leading to slow convergence and subpar final results. Researchers have addressed these limitations by developing an algorithm that incorporates prior knowledge about electricity pricing patterns and battery state-of-charge dynamics.
The improved approach, termed PriceSOC-PSO, strategically initializes the optimization by leveraging time-of-use pricing structures and battery scheduling principles. Rather than random starting points, the algorithm generates high-quality candidate solutions informed by the economic principle of valley-period charging and peak-period discharging. This mimics the operational strategy a human dispatcher might employ: charge during low-price, low-demand hours and discharge during peak periods when electricity is expensive.
Simulation results demonstrate substantial performance gains. When weighted to emphasize cost reduction, the method cuts daily power purchase expenses by 14.8% and grid fluctuations by 62.6% relative to stations without storage. Under weight settings favoring grid stability, power purchase costs fall 11.7% while grid fluctuations decrease 74.9%.
The work also compares performance against other metaheuristic algorithms including Grey Wolf Optimizer, Harris Hawks Optimization, and Sparrow Search algorithms, establishing that the informed initialization strategy outperforms competing methods. For utilities and charging network operators, this research provides a practical framework to design station schedules that balance economic efficiency with grid reliability—a critical capability as transportation electrification accelerates worldwide.
