Reconfiguration of Distribution Network For Power Loss Minimization & Reliability Improvement using Binary Particle Swarm Optimization

This paper describes a method to evaluate optimal configuration of distribution system with minimum active power losses and maximum reliability indices. The optimal configuration is obtained by using Binary Particle Swarm Optimization (BPSO). Simple Backward-Forward Load Flow (BFLF) method is applied to calculate active power loss and basic cut-set method is used to evaluate reliability indices. Graph theory-based incidence matrix concept is used during optimization so as to maintain the radial topology of the network and to discard any looped or unconnected configuration. Depth-first Search (DES) algorithm is utilized to check the connectivity of all nodes and to get intermediate nodes between load point and substation. Detailed study of two distribution networks is done and presented. Various possible configurations are compared based on their power loss and reliability indices. Based on results, recommendation is made for the adoption of network operational topology for desired outcomes.