Research On Power System Reactive Power Optimization Based On Improved Particle Swarm Optimization

Reactive power optimization not only can reduce the active power loss, but also keep the grid under security, reliability and economy. Reactive power optimization is a typical non-linear programming problem with multi-control variables, multiple constraints condition and so on. Conventional optimization algorithm for solving such problems has greater limitations, compared to that, intelligent optimization algorithm show a clear advantage, this thesis conducts in depth analysis and research for PSO, starting from different point of view, several methods is presented.Firstly, from economic consideration, this thesis takes the minimization of active power loss as an objective function, establish reactive power optimization model and propose an improved adaptive chaotic particle swarm optimization algorithm (AC-PSO).In the initialization of the particle swarm, the algorithm introduces chaotic idea and increase the diversity of the population of particles, in the process of the particle swarm optimization, the algorithm use adaptive dynamic inertia weight and learning factors to improve the local and global search ability and get better solution quality. Secondly, to further enhance the exploration ability of particle swarm optimization algorithm in the solution space, this thesis propose two improved algorithms start from the perspective of variation:he accelerating factor of the particle swarm optimization (AF-PSO) and the random variability of the particle swarm optimization (RM-PSO).By simulation, the performance of two algorithms have been analyzed and compared, at last, using reactive power optimization calculation verify the validity of the algorithm.Finally, from comprehensive consideration of the economy and security, this thesis take active power loss minimum, the minimum voltage deviation and static voltage stability margin as the objective function and establish the fuzzy multi-objective reactive power optimization model.Hybrid optimization algorithm-adaptive chaotic random variation of particle swarm optimization algorithm (AC-RM-PSO) optimization is proposed to solve the optimization problem, the fuzzy solution of the multi-objective optimization solve the problems between the different dimensions of the multi-objective; AC-RM-PSO optimization algorithm avoid falling into local optimum and global convergence and achieve good results.