Multi-stage, Degradation-aware Day-ahead Energy Management for Prosumer-based Microgrids using Python-Gurobi

Abstract

Renewable energy sources are becoming more common in microgrids, which creates new possibilities and new challenges for providing reliable and affordable energy. The proposal of a multi-stage, day-ahead energy management model for prosumer-based microgrids is the objective of this research. The model clearly considers battery degradation and the variety of load and generation profiles. The framework is implemented with a commercial Mixed-Integer Linear Programming (MILP) solver. It examines several operational horizons, ranging from one to three days, to evaluate their economic, technical, and environmental impacts. Key performance indicators include the following: electricity cost per unit, self-consumption rate, renewable contribution, curtailed energy, carbon emissions, and energy storage aging. The findings show that using a three-day plan is the best balance between reducing costs and using renewable energy, while also keeping batteries long-life. Shorter horizons tend to improve on-site PV utilization. In contrast, extended horizons yield moderate yet consistent results. This highlights the trade-offs between conflicting objectives. The framework is a useful guide for people who plan and manage microgrids. It helps them design microgrids that are both efficient and well enough for the environment. The framework may be expanded in the future to include stochastic forecasting, demand flexibility, or hybrid AC/DC configurations, with the goal of improving resilience and operational performance.