A new hybrid algorithm for solving distribution network reconfiguration under different load conditions

Abstract

Distribution Network Reconfiguration (DNR) is a significant problem for keeping the network running under normal conditions. In this study, for preventing the premature convergence issue, also to improving the searching ability of the Binary Particle Swarm Optimization (BPSO) algorithm, chaotic strategy is incorporating with BPSO algorithm to create a new hybrid algorithm called Chaotic BPSO (CBPSO). Undeniably, the chaotic strategy enables the hybrid CBPSO algorithm to slip from the local optima and also to reach optimal solution in fewer number of iterations compare to BPSO due to the remarkable behavior and ergodic of the chaos strategy than random search in BPSO algorithm. The CBPSO algorithm is presented as a advantageous optimization tool for solving DNR. In this problem, decreasing of real power loss (PL) is an objective function while node voltage, system radially and line current have been utilized as a constrains of the system. The search space in this problem for the presented technique is a group of lines (switches) that are normally opened or closed. Two types of loads are presented: the constant and variable loads for testing the efficacy of the CBPSO method for tackling DNR problem when the load is changes. The proposed technique is implemented on IEEE Node−33 system by utilizing MATLAB R2013b software for verifying the efficacy of CBPSO technique. The simulation results confirm that CBPSO technique has high ability in reducing PL and raising the voltage profile of the grid compared to BPSO and other procedures in the literature.