Energy Storage Capacity Optimization for Improving the Autonomy of Grid-Connected Microgrid

Abstract

To support the autonomy and economy of grid-connected microgrid (MG), we propose an energy storage system (ESS) capacity optimization model considering the internal energy autonomy indicator and grid supply point (GSP) resilience management method to quantitatively characterize the energy balance and power stability characteristics. Based on these, we establish a three-stage coupled model including investment decision, day-ahead operation strategy, and real-time power fluctuation smoothing control. Investment processes solve the balance of ESS investment and internal energy autonomy. Day-ahead operational scheduling mainly solves unit commitment problems in MG, which are constrained by the ESS capacities. For real-time power fluctuation smoothing, we propose a power fluctuation smoothing control strategy, coordinated by ESS and direct load control (DLC), to achieve GSP resilience management. Furthermore, we use Monte Carlo Simulations, stochastic scenario combinations, and uncertainty set to characterize the fluctuations of both supply and demand at the above different stages, respectively. To solve the established model with multi-objective, multi-uncertainties, and multi-stage coupling, a robust counterpart method has been used to convert the uncertainty problem to the deterministic one, and then decomposition-based multi-objective evolutionary algorithm (MOEA/D) to efficiently used to solve the problem. Finally, simulations are conducted to verify the effectiveness of the proposed model.