Análise de Métodos para Agrupamento de Cargas Elétricas em Redes de Distribuição de Grande Porte usando Sistema de Informação Geográfica

Abstract

The growing integration of renewable energies and the new technologies adoption make the coupling between supply and demand a challenge. Furthermore, the allocation of distributed energy resources on the electrical grid allows islands configurations, as microgrids (MG). In a MG, when the consumption exceeds the generation capacity, it leads to an inefficient operation of the system, therefore, it is necessary to know the electrical demand patterns. However, electrical load grouping of a study area into categories is a technique that allows finding consumption relationships within and between groups. This article presents and compares two methods to identify electric load clusters in distribution networks using geographic information systems (GIS), DBSCAN and K-means methods are tested. Electric load density and electrical distance prove to be effective setting factors. The performance of the methods is evaluated in the coverage area of three distribution substations of an Ecuadorian electric utility. The results show clusters, that maximize electrical and geographical proximity, which represents a support tool in the MG planning process.