Binary PSO-Based Reconfiguration Strategy for Loss Reduction in Three-Phase Unbalanced Distribution Networks

The presence of unbalanced loads and distributed energy resources in modern distribution systems poses significant challenges for maintaining power quality and minimizing losses. Traditional reconfiguration techniques often assume balanced loading, leading to suboptimal solutions when applied to real-world, unbalanced systems. This paper presents a novel approach to reconfiguring unbalanced radial distribution networks using Binary Particle Swarm Optimization (BPSO) with the objective of minimizing total active power losses. The network switches are encoded in a binary string representing candidate topologies, and BPSO is employed to search for the optimal configuration. The evaluation of each particle's fitness is based on a detailed three-phase unbalanced load flow analysis, which accurately captures the effects of load asymmetry. The proposed method is tested on the Gejayan Network unbalanced distribution system. Results show BPSO reduces power losses by 47 % and decreases the unbalanced current by 57 %, while satisfying all operational constraints, including radial topology preservation and voltage limits.