Iterative method for detecting and localizing islands within sparse matrixes using DSSim-RT

Abstract

Nowadays, simulation for power flow analysis of distribution systems must consider their separation in several islands. This separation is the result of management activities such as reconfiguration, fault detection and isolation, among others. For power flow analysis the distribution system is represented as sparse matrix; however, if these islands are not considered the results can be inaccurate and inconsistent. There are several methods for detecting islands reported in literature, but these methods normally require that the Y matrix is organized in special form. This situation is not always possible because is the user who defines the order of the matrix when using simulation software. This paper presents an algorithm for detecting and localizing islands within sparse matrixes no matter the order of the elements of the matrix. By using graph theory, the incidence matrix of the power system can be transformed into its adjacent representation. With this matrix the interconnected elements are detected and classified. Then, with this new reference matrix the algorithm delivers number of islands, the suggested order for the columns of the Y matrix and more information. The aim is to simplify the localization of islands with a simple algorithm, which can be implemented using any programing language. This method is used in DSSim-RT, which is a simulator based in OpenDSS, for detecting islands and performing the multithread power flow analysis of distribution systems in Real-Time.