Locality-balanced Energy Community Aggregations Considering Net Energy Predictions of Microgrids

Microgrids are small entities in the smart grid, which consisting of distributed energy resource, energy consumers, and energy storage systems, etc., and can connect to the main grid or work in off-grid mode. This paper focuses on the aggregations of microgrids to form energy communities, which is beneficial to the management and the locality-balance of energy. Considering the time-varying character of photovoltaic power, we propose a novel algorithm to aggregate microgrids using the predicted data of net energy instead of historical data. The proposed algorithm firstly clusters the microgrids with positive net energy into different communities, based on Kmeans method, and then add the remaining microgrids with negative net energy into one selected cluster one by one according to the information of both location and total net energy of the cluster. Moreover, two evaluation factors are proposed to validate the efficiency of the proposed aggregation algorithm. The experimental results demonstrate that the proposed aggregation algorithm can realize the energy balance of all communities as well as guarantee more communities with positive net energy and minimize the average distance of microgrids to the community center than SECs.