Reactive Power Optimization Based On Particle Swarm Optimization Algorithm With Metropolis Rules

The reactive power optimization is mainly aimed to find a solution to optimal dispatching the available voltage of generators, transformer tap-settings and shunt reactive sources to get minimal power loss while subjecting to some operational constraints, and it is an important method for improving electricity quality and reducing the power loss in power system.This paper presents a model for the optimization first. And then, in order to solve the problems, including high possibility to converge at local optimum and difficulty to choose parameters such as penalty factors in optimizing searching and study factor. This paper presents some strategies. The real number coding and integer coding are adopted in this paper. In the optimizing model of this paper, penalty factors are adjusted adaptively. The values of penalty factors increase with the iteration. It enhances the stability of the PSO converges in the feasible area. And then, the proposed method which combined particle swarm optimization algorithm (PSO) and Metropolis Rule (M-PSO) can avoid efficiently falling into the local optimization.According to this thesis, a program based on MATLAB is designed for power optimization in power system. The simulation results of IEEE-30 and IEEE-118 power system verify the effectiveness of the proposed model and algorithm.