Multi agent system based on law of gravity and fuzzy logic for coalition formation in multi micro-grids environment

This paper presents an intelligent multi-agent system based on the law of gravity and fuzzy logic for the optimal management of the intra-MicroGrids energy. The aim is to overcome the weakness of the MicroGrid (MG) regarding the intermittent generation of its renewable source and its dependency to the main grid as the only reliable part in the power exchange. Taking into account that the power loss between each MG and the main grid is larger than that among the MGs, the aforementioned dependency causes high transfer loss. On the other hand, the failure of the single point of common coupling with the main grid makes the MG in a critical situation, especially when there is an internal lack in production. To deal with these problems, the proposed system allows the MGs which may be either generator or load to collaborate together in coalition form. So, the optimal energy distribution and power loss reduction are targeted by dynamically changing the size and the structure of coalitions of all participating MGs in the power exchange. Based on the law of gravity, the system allows the MGs that require energy to possibly specify the resources with the best energy offer and the lowest transfer loss. Besides and in order to deal with the multiple accesses to the same resources, a process based on fuzzy logic is made. Finally the JADE (JAVA Agent DEvelopment Framework) has been used and a comparison is made in order to show the impact of using the proposed system.