Distribution Network Reconfiguration Using Genetic Algorithms With Sequential Encoding: Subtractive and Additive Approaches

This paper deals with radial distribution network reconfiguration for loss and switching mitigation. Its main contribution is the presentation of a genetic algorithm (GA) with two network encodings, capable of representing only radial connected solutions without demanding a planar topology or any specific genetic operator. The code was named sequential because the evaluation of the i th gene depends on the information forthcoming from all the previous genes. In addition to a full description of proposed techniques with examples, the paper presents a survey of existing network codifications including their search space and chromosomes for a 33-bus example system. In order to validate the proposed algorithms, these and other two techniques in the state of the art are applied to reconfigure three sample systems which have been broadly published in the technical literature. Performance is assessed by using the results of implemented representations and the reconfiguration literature.