Research Of Distribution Network Planning Based On Modified Particle Swarm Algorithm

The growing urban agglomerations in our country,and the distribution network is an important urban infrastructure, scientific distribution network planning must keep economic development era. Distribution network construction and development in addition to many restriction of technical factors, also need to consider a lot of external uncertainties. So the development of power distribution network should not only adapt to the development of the load, more requirements for such as urban environment, industrial structure, space resources,energy resources and other external conditions, can be regarded as a solving combinatorial optimization problems. Due to the development of the national urbanization progress, the distribution network scale is bound to increase, programming problem is particularly complicated. Better economic benefits, reducing the power grid invest funds, to allow more rational allocation of resources, is the contribution of each grid workers, and therefore a good distribution network planning is very important.This paper first introduces the present commonly used method of distribution network power flow calculation, selection of backward / forward iterative algorithm for network programming power flow calculation at this paper. Then studies the relevant parameters selection of particle swarm algorithm. Finally in the paper, the improved particle swarm optimization algorithm was applied to the actual distribution network programming examples,and verifies the feasibility of the algorithm. In the paper, the main research results and the concrete content is as follows:(A) Briefly introduces several kinds of commonly used power system power flow calculation mothod, comparing various power flow calculation method, then elaborated the present stage using common backward / forward iterative algorithm. In distribution network planning, must analysis the constantly changing network operation mode, this paper analyzes the improvement of backward / forward iterative algorithm, through node systems and compares various algorithms, proved that the method applicable widely scopes and need less number of iterations.(B) In view of the standard particle swarm algorithm easy to fall into local optimum,this paper analysis the research of previous scholars,improved the acceleration factor,making it the asymmetric arc cosine form.Application of standard test functions to verify the improvement scheme, which indicates that the search performance is good. Improved method in particle search has good diversity at first, make the particles have better global search ability, and in the late calculation, also has a better performance, convergence precision is greatly improved.(C)Improved particle swarm optimization algorithm was applied to optimization of distribution network programming, select minimum planning in annual comprehensive costs as the objective function. Choose a 10 KV distribution network programming examples to verify the above woek.Application Matlab software for example, the simulation experiment results show that the improved algorithm effectively overcomes the optimization iteration process is easy to fall into local optimal problem, and get a better convergence accuracy,faster search rate.