Investment in renewable energy is expanding globally due to global warming and rising energy costs. Because renewable energy depends largely on weather conditions, load balancing and managing power surpluses are important. Coalitions for researching electric power management are increasing. A microgrid coalition is comprised of demanders and suppliers, so it is necessary to properly match supply and demand. Appropriate matching minimizes power loss and satisfies demand. Power loss is determined by the amount of power and the distance of the power transmission. In order to calculate the minimum power loss, it is necessary to calculate all combinations, including the power transmission destination. This is computationally difficult, so it requires a heuristic algorithm that operates at a high speed. It is important to use the features of a power grid to provide appropriate power supply and demand; however, most studies do not consider the power grid's features. In this paper, we propose a distributed power managing method based on the power grid's centrality. We compare our proposed method with an existing method by power simulation and confirm that the proposed method is better at matching demand with supply because transmission loss in the power grid network is reduced.