Reactive Power Optimization Based On Modified PSO

Reactive power optimization of power system has a precondition that it should meet the commands of loads and various constraints formulas when the system is running. It aims to raise the voltage quality and to reduce the net loss. Reactive power optimization of power system is a nonlinear programming problem with a large number of variables and constraint conditions, and the control variables are made of continuous and discrete variables, so reactive power optimization problem is complex and hard to solve with conventional methods.According to the above-mentioned characteristics of reactive power optimization, Particle Swarm Optimization (PSO) algorithm has been applied to solve the question. PSO algorithm is a kind of heuristic algorithm based on swarm intelligence which is easy to implement and has fast convergence performance. However, the convergence precision of the basic PSO algorithm is low, and the algorithm can easily fall into its local minimum value. Aiming at these shortcomings of PSO algorithm, the population's convergence degree is used to define an inertial factor which can adapt with the swarm so that the convergence precision can be raised, and the algorithm can be protected from falling into its local minimum value. In the process of solving the maximum-exceeding problem of the population's convergence degree, this paper modified the method of regenerating exceeding-part swarm. According to the static distribution law of swarm after interation, the exceeding-part swarm is regenerated based on normal distribution function, so the population's convergence degree can meet the constraint. Thus the inertial factor can be dynamically regulated. Based on the above-mentioned method, the PSO algorithm can be modified. When applied to classic optimization function, the Modified Particle Swarm Optimization (MPSO) algorithm proves superiority in global convergence and convergence precision compared to standard PSO.This MPSO algorithm is applied to IEEE-6, IEEE-14 and IEEE-30 standard systems, considering all factors affecting reactive power flow such as generator terminal voltage, output of reactive power sources and transformer tap positions, taking the real power loss as optimization goal ,meeting voltage constraints and reactive power constraints, setting different parameters in different systems to meet the different demands. Compared with the basic PSO algorithm, the algorithm of MPSO performs better, and the convergence precision has been greatly improved.