Effective Customization of Evolutionary Algorithm-Based Energy Management System Optimization for Improved Battery Management in Microgrids

The growing penetration of renewable energy sources into electricity grids, along with the problems linked to the electrification of rural areas, has drawn more attention to the development of microgrids. Their Energy Management Systems (EMSs) can be based on evolutionary optimization algorithms to identify efficient scheduling plans and improve performance. In this paper, a new approach based on evolutionary algorithms (EAs) is designed, implemented, and tested on a real microgrid architecture to evaluate its effectiveness. The proposed approach effectively combines heuristic information with the optimization capabilities of EAs, achieving excellent results with reasonable computational effort. The proposed system is highly flexible, making it applicable to different network architectures and various objective functions. In this work, the optimization algorithm directly manages the microgrid Energy Management System, allowing for a large number of degrees of freedom that can be exploited to achieve highly competitive solutions. This method was compared with a standard scheduling approach, and an average improvement of 11.87% in fuel consumption was achieved. After analyzing the differences between the solutions obtained, the importance of the features introduced with this new approach was demonstrated.