Heuristic Algorithm for Distribution Networks Reconfiguration for Loss Reduction Based on Graph Theory

One of the ways of reducing active power losses in the distribution networks is using the network reconfiguration for different load levels. Since the number of possible configuration of a distribution network is usually very much and then review of all of them (especially in the network of large-scale) to finding the optimal configuration is not possible simply, therefore in this paper a new algorithm based on the graph theory and the its developed algorithms are presented for the reconfiguration of distribution networks. The proposed algorithm start with a meshed distribution system which obtains by considering all switches closed. The distribution system models as a super graph and proportionate weights to losses are attributed to each branch of this graph. Due to the nature of reconfiguration operations, the most important factor in the distribution networks is to reach the optimal configuration in the shortest possible time. The developed algorithms of graph theory have appropriate speed and considering that the number of load flow calculation in the proposed algorithm is little as compared to other methods. Therefore the proposed algorithm of this paper, as compared to other methods has good speed in reaching the desired optimum. The proposed algorithm has been implemented on the sample distribution networks. The obtained results show superiority and efficiency of proposed algorithm as compared to with other methods.