Two-Stage Stochastic Optimization for Cost-Effective Energy Management in Grid-tied Microgrids

Abstract

Effective energy management is essential for minimizing operational costs in grid-connected microgrids (MGs), particularly as renewable energy sources such as solar photovoltaics and wind turbines are increasingly integrated into modern power systems. This paper presents a two-stage energy management strategy aimed at minimizing the total cost of a grid-connected MG. In the first stage, day-ahead scheduling, energy dispatch is optimized using stochastic optimization techniques while accounting for uncertainties in renewable generation and load demand. A Monte Carlo simulation generates multiple scenarios to assess future states, facilitating precise decision-making for grid interaction and local generation. As a result, the total operational cost is reduced from Rs. 12,521 to Rs. 12,390, and the total cost is reduced from Rs. 158,090 to Rs. 14,998. The second stage, real-time scheduling, refines the day-ahead plan by adjusting for real-time fluctuations in demand and generation, ensuring system balance and reliability. By integrating metaheuristic algorithms with real-time control, the proposed strategy minimizes energy exchange costs with the grid, reduces operational expenses of conventional generators, and maximizes the utilization of renewable energy. Case studies validate the effectiveness of the proposed methodology in reducing overall costs, maintaining grid stability, and enhancing renewable energy penetration. The method is adaptable to various MG configurations, offering a robust and cost-efficient solution for energy management in grid-connected systems.