Power system reactive power optimization based on adaptive group PSO

Reactive power optimization problem is typical non-linear problem with characteristics of multi-objective, uncertainty and multi-restriction. Classical mathematical techniques are inadequate and insufficient to the optimal operation of power systems due to the inherent complexity. This paper studied on PSO algorithm and reactive power optimization, and made some new ideas into PSO algorithm to improve its optimization ability. Adaptive particle swarm particle swarm algorithm was proposed for solving the problem of large-scale search failed, and presented a method of adjustment parameters for adaptive PSO algorithm according to the optimization speed. In meeting the conditions for convergence, the algorithm enabled the speed of particles in accordance with the ideal adjustment of the parameters of adaptive search. On this basis, to overcome the disadvantage that PSO maybe get in local optimization solution, an improved PSO was developed. The algorithm used statistical laws of particle fitness to classify particles, and took different evolutionary models for different kinds of particles. Finally, IEEE-6 node standard test system was computed for reactive power optimization, compared with standard PSO, the proposed PSO algorithm had more higher search efficiency and better ability of global optimization.