Robust mean-variance optimization model for grid-connected microgrids

This paper proposes a mean-variance optimization model for the grid-connected microgrid energy management system (MG-EMS). In the proposed method, both the expected system operating cost and the tie-line power fluctuation variance are taken as the objective functions to provide a trade-off between operating benefit and risk assessment for decision makers. Further, robust optimization (RO) has been applied to eliminate the potential reliability issue due to wind power uncertainty. Therefore, the proposed method can effectively generate a robust optimal day-ahead operating schedule for wind power, energy storage (ES), and load management under worst-case scenarios in the microgrid. Finally, the case study on a revised IEEE 14-bus system has been implemented to demonstrate the validity and effectiveness of the proposed method.