Projeto de ilhamento controlado em redes de energia elétrica

Abstract

Controlled islanding is implemented in electrical power systems (EPS) to achieve the physical separation of the network into independent subsystems (sets of coherent generators) to promote the continuity of electrical energy supply in the presence of a critical fault, during which usual protection systems are not able to operate effectively and maintain the network operation in a fully interconnected manner. The islanding project, in which the elements to be shut down are selected in order to separate of the network into subsystems, can be modeled as an optimization problem in which the algebraic graph theory can be applied to associate the EPS with a branch-node incidence matrix, and use the inherent characteristics of this matrix to carry out research in the graph (associated to EPS) more efficiently and to find the cut-set that optimizes the cost (objective) function of the problem. In this work, we present an algebraic graph algorithm and a genetic algorithm, both of which are based on the lower triangular factorization of the node-branch matrix, to design controlled islanding in power networks, in order to obtain the smallest active power flow between the subsystems to be separated.