Online Energy Management for Highway Charging Station via Distributed Stochastic Optimization

The rapid expansion of electric vehicle (EV) deployment presents significant challenges to the operational management of highway charging stations. To address the energy management issue of these charging stations equipped with photovoltaic (PV) panels, an online energy management strategy based on distributed stochastic optimization is proposed. This strategy aims to minimize the long-term cost of purchasing electricity from the grid while ensuring the stability of the EV charging queue. Initially, a stochastic optimization model for the PV charging station is formulated, accounting for uncertain factors such as electricity prices, PV generation, and EV charging demand. Subsequently, leveraging Lyapunov optimization theory, the long-term stochastic optimization problem is transformed into a real-time decision-making problem, with a drift-plus-penalty optimization framework designed for solution. Simulation results show that the proposed strategy can dynamically adjust charging power in response to varying electricity price scenarios, achieving an optimal trade-off between cost reduction and queue length. Compared to the greedy strategy, the proposed approach reduces the total average cost by 13.6%.