Reconfiguração de redes de distribuição de energia elétrica usando otimização por enxame de partículas aprimorado

Abstract

The present work presents a method to solve the reconfiguration problem of primary distribution networks, making use of the features of the Open Distribution System Simulator software and a new binary particle swarm (BPSO) optimization technique, named for improved selective BPSO. In addition to using the search concept for mesh or selective space,whichconsiderablyreducesthesolutionssearchspace,theproposedmethodpresents a new sigmoid function capable of control in the rate of change of the particles and of promoting a better exploration of the space of result in an improvement in the convergence of results. The solution algorithm was developed to minimize system power losses, taking into account its operational limits. For this, a program was written in the programming language C# allowing the integration between OpenDSS and the proposed reconfiguration method. Its operation has been demonstrated in a simple 5-bar system. Finally, the proposed algorithm was applied in the distribution systems of five (33-Bus and 69-Bus) and thirteen dimensions (94 nodes) available technical literature, performing validation tests and effectiveness of the developed routines. The results of the tests were compared with results of other modern techniques reported in the literature.