Power distribution network reconfiguration for loss minimization using a new graph theory based genetic algorithm

A planned and effective distribution system requires the best topological configuration to cope up with the increasing power demand with minimum power loss. This paper presents the solution of a power distribution network reconfiguration problem using a graph theory based genetic algorithm. The new approach is framed where improvements are suggested on chromosome coding and evaluation of fitness function. Moreover, a new load flow method is used to implement the proposed genetic algorithm which is independent of any particular node numbering scheme and input data. Also, it has an inbuilt mechanism of checking connectivity of all nodes to substation node(s) and radial topology of a distribution network. The effectiveness of the algorithm is tested on 33-bus, 69-bus, 84-bus and 136-bus distribution systems and the results are presented.