Multi-Scenario Optimal Scheduling of a PV-Storage-Charging Microgrid Considering Flexible Charging and Discharging of Electric Vehicles

To address the increased load volatility and insufficient interaction stability with the main grid caused by large-scale integration of electric vehicles (EVs) into microgrids, a two-stage optimal scheduling strategy for a PV-storage-EV charging microgrid is proposed, incorporating flexible EV charging and discharging. First, in Stage 1, a piecewise logistic regression model is employed to accurately quantify users’ willingness to participate in vehicle-to-grid (V2G) services. A bi-objective optimization model is formulated to minimize both load fluctuations and user charging costs. The zero-sum game strategy is adopted to determine the weighting coefficients of the multiple objectives, thereby fully exploiting the flexible regulation potential of EVs to reduce user costs while smoothing the load profile. Subsequently, based on the results from Stage 1, Stage 2 constructs a model that minimizes both microgrid operating cost and tie-line power standard deviation, optimizing the power dispatch of internal generation units and power exchange with the upstream grid. This stage also investigates microgrid scheduling responses under low EV penetration scenarios. Finally, the mixed-integer programming problem in Stage 1 is solved using Cplex, while the multi-objectivegrey wolf optimizer—enhanced with an improved Tent chaotic map and a state-driven adaptive iterative strategy—is applied to solve the models in both stages. Simulation results demonstrate that, under various EV participation scenarios, the proposed approach enables the microgrid to simultaneously achieve economic benefits for both end-users and the microgrid operator, as well as enhanced grid stability.