Efficient energy management is critical for minimizing operational costs in grid-connected microgrids (MGs), particularly as renewable energy sources, such as solar photovoltaics and wind turbines, become more integrated into modern power networks. This study proposes a two-stage energy management approach to maximize cost efficiency in a grid-connected MG. The first stage, day-ahead scheduling, utilizes stochastic optimization techniques to optimize energy dispatch while accounting for uncertainties in renewable energy generation and load demand. A Monte Carlo simulation generates multiple scenarios to predict future system states, facilitating accurate decision-making for energy dispatch and grid interaction. The proposed strategy results in a reduction of operational costs from Indian rupee (INR) 12,521 to INR 12,390, and the total cost decreases from INR 158,090 to INR 14,998. The second stage, real-time scheduling, dynamically adjusts the day-ahead plan to accommodate real-time variations in demand and generation, ensuring system stability and reliability. By integrating genetic algorithms and particle swarm optimization with real-time control, the methodology effectively minimizes energy exchange costs with the grid, reduces the operational expenses of conventional generators, and enhances the utilization of renewable energy sources. Case studies validate the proposed framework’s effectiveness in lowering overall costs while maintaining grid stability and increasing renewable energy penetration. The presented strategy is adaptable to various MG configurations, offering a reliable and cost-effective solution for energy management in grid-connected systems.