Optimal Reconstruction of Photovoltaic Distribution Networks Based on an Enhanced Random Weight Particle Swarm Algorithm

Abstract

This study proposes an innovative and improved random weighted particle swarm optimization algorithm with the goal of effectively resolving the static reconfiguration issue of PV distribution networks. The algorithm, tailored to suit the unique characteristics of distributed PV distribution grids, not only enhances the search's efficiency and precision, but also successfully avoids the algorithm's premature convergence by incorporating the Tabu search approach. This significantly augments the comprehensive capability for global search and local optimization. Simulation is utilized to authenticate the algorithm's efficacy, and a comprehensive comparison, such as search speed and accuracy, is conducted with the particle swarm algorithm, the improved particle swarm algorithm, and the stochastic weighted particle swarm algorithm, all of which are based on the IEEE 33-node distribution network model with distributed photovoltaic. The comparison results demonstrate that the upgraded algorithm exhibits remarkable performance in terms of search efficiency, global search aptitude, and accuracy, far surpassing the other comparison algorithms.