Stochastic optimization of integrated electric vehicle charging stations under photovoltaic uncertainty and battery power constraints

Optimal scheduling based on accurate power state prediction of key equipment is vital to enhance renewable energy utilization and alleviate charging electricity strain on the main grid in the integrated electric vehicle charging station (EVCS) with photovoltaic (PV) and battery energy storage system (BESS). However, the multi-source power prediction uncertainty of PV and the complex discharging/charging state of power (SOP) constraints of BESS were less explored in previous studies, which might result in suboptimal or even infeasible scheduling. This study proposes a novel stochastic scheduling optimization framework for the PV-BESS-EVCS integrated system. In this framework, a hybrid characterization method is developed to describe the multi-source uncertainty of PV power, and a piecewise linear model is developed to estimate the dynamic SOP constraints of BESS. The effectiveness of the proposed framework is demonstrated by a set of typical scenarios for commercial regions. The results indicate that compared to optimal results in the deterministic model, the proposed framework could improve operational profits by 0.41 % while reducing equivalent carbon emissions and power curtailment rates by 0.61 % and 3.05 %, respectively. It is thus illustrated that the proposed framework could provide a promising solution for the optimal scheduling of the PV-BESS-EVCS integrated system under uncertainty.