Particle Swarm Optimization Algorithm And Its Application In The Optimal Operation Of Power System

At present, the total installed capacity of the power system in China has already ranked the second place in the world. Therefore, it has been one of the hot focuses concerned by the scholars to make short-time generation plans and cost optimization of power system under the condition of current technique equipments and normal industry level, which is to properly distribute the assignment among the generating units so as to avoid mass waste of the energy caused by the unnecessary on-off and the idling of the machines. On the other hand, with the rapid development of national economy and the increasing power demand, the economical operation of the power networks gets more and more attentions. While, the reactive power optimization is considered effective to guarantee the safety and the economical operation of power system. Making rational optimization to the reactive component not only can improve the voltage level of the systems operation, but also can decrease the active and the reactive network loss of power system and make its operation more effective.In this paper, several emerging intelligent optimization methods are further analyzed, based on the current research status of the unit commitment of power system and the reactive power optimization of distribution system. Then the Particle Swarm Optimization (PSO) which has better search efficiency and convergence property is selected to be modified. Moreover, new solutions are suggested, which are based on the Modified Particle Swarm Optimization (MPSO) and aiming at the unit commitment of power system and the reactive power optimization of distribution system. In sum, the work in this paper comprises the two parts as follows:1. The algorithm of Modify Particle Swarm Optimization (MPSO) which bases on the method of inertia weight is attempted, and a lot of discussion is made on the selecting principle of correlated parameters to effectively overcome the prematurity problem of Particle Swarm Optimization and to improve the optimization property of it. These are tested by the optimizing situations of some typical complicated functions, and further discussion is made on how the sampling of every parameter of the algorithm affects the optimizing process.