A Method Of Fault Location Algorithm For Distribution Networks With DG Based On Harmony Algorithm

With the change of energy structure and the user’s growing demand for electricity, the technology of distributed generation with clean environmental protection, low energy consumption, flexible power advantage obtained the fast development. But with the access of distributed power supply, power distribution network gradually developed from the traditional single power supply radial network to the power of complicated power network, random variation trend direction, quickly and accurately fault location has brought great influence, need to study new algorithms to realize with distributed power distribution network fault location for improving power supply reliability, reduce the scope of power has important practical significance.This paper researched the fault location algorithm for DG distribution power network through the theoretical research and simulation model of Matlab/Simulink.The distributed power access distribution network for the impact on short circuit current has made the detailed analysis and verification respectively, the concluded that the access of distributed power supply might be on the short circuit current shunt, help increased and reverse flow, etc. Secondly based on the structural feature of complex containing DG distribution network, established a strong adaptability and single, multiple power supply general switching function model. In order to improve the search efficiency, introduction about the "tree" zoning method, eliminate the trouble-free current passive and reduce the dimension of branches; Then use the characteristics of rapid and accurate algorithm for global optimization of harmony; Finally through the concrete case analysis to verify the accuracy and rapidity of the method.In this paper, the harmonic convergence speed and fault tolerance of the algorithm is studied. Harmony algorithm is developed in recent years a new intelligent optimization algorithm, compared with the traditional optimization algorithm, the algorithm is simple in concept, the advantages of easy to implement, but the algorithm of harmonic memory is random, uncertainty, and the memory initialization has a strong dependence. In order to improve the positioning speed and fault tolerance of the algorithm, using the highly parallel, the advantages of robust genetic algorithm (GA) produce harmony library initial solution; At the same time to improve the way of harmony of library data processing, overcomes the initial harmony memory strong dependence of random selection and parameter uncertainty; The last program using Matlab to compile corresponding algorithm, respectively for the same optimization computation and analysis of the distribution network model, the results show that the optimized harmony search algorithm can effectively improve the convergence speed and precision of fault location algorithm, under the condition of lack of information, distortion and so on, it has a good fault tolerance.