Development of a Modified Particle Swarm Optimization Algorithm for Voltage Profile Enhancement in Distribution Network Reconfiguration

Abstract

Distribution network reconfiguration (DNR) is the most economical method to reduce power losses in distribution network. It involves altering the topological structure of feeders by changing the switching states of normally opened switches. Previous optimization algorithms in the time past have not developed significantly enhanced the voltage profile of distribution network reconfiguration, hence the motivation for this research paper. This paper therefore develops a modified particle swarm optimization (PSO) algorithm for voltage profile enhancement in distribution network reconfiguration. In the modified PSO algorithm, the vector containing the nodal voltages is defined and the element of the vector is assigned a value of 1 p.u with zero angle. The currents of the network branding were calculated. The voltage of the node near to the reference node was calculated by subtracting the corresponding voltage drop from the voltage of the previous nodes until convergence was attained. Distribution load flow based on backward forward sweep was used to compute the losses while a number of iterations were performed until convergence was attained. The modified PSO algorithm was tested with a 28-bus radial distribution network using MATLAB programming language. The results of the work showed that the loss reduction obtained was 7.53% more than the existing approach with good convergence characteristics. The percentage power loss reduction was 47.58% which indicated a load balance in the network with a 12.32% improvement in voltage profile as a result of the efficiency and good convergence characteristic of the modified PSO algorithm, thus making it useful in large-scale network.

Keywords: Distribution Network Reconfiguration (DNR), Particle Swarm Optimization (PSO), Modified PSO Algorithm, Distribution Network, Backward- forward Sweep, Power Loss Reduction, Voltage Profile, Line Losses.