Power System Reactive Power Optimization Based on MIPSO

DingPing Li,GuoLiang Shen,WenDong Guo,Zhi Zhang,BaoNing Hu,Wei Gao

DOI: https://doi.org/10.1016/j.egypro.2011.12.1012

2012-01-01

Energy Procedia

Abstract:In order to improve power quality and reduce network losses, this paper proposes a modified immune particle swarm algorithm for power system reactive power optimization. To overcome the disadvantages in the traditional PSO algorithm about low accuracy and easy to fall into local optimal, MIPSO adopts sinusoidal changing inertia weight strategy to make particles explore the local and global optimization more efficiently. Introduce convergence acceleration factor to improve the convergence rate. Modified immune principle enhances the search capability to avoid premature convergence. Finally, compare the reactive power optimization results of Wuzhong of Ningxia system by MIPSO with other classical PSO algorithms. The experimental results show that the MIPSO algorithm is an efficient and feasible approach for reactive power optimization.