Study On Optimal Allocation Of Distributed Generation And Reconfiguration Of Distribution Network

Distributed generation, as an aid for traditional power supply, which makes full use of new and clean energy and is very efficient and flexible according to the local situation, is widely utilized and playing a more and more important role in power system.The introduction of distributed generation, by which the distribution system is changed into a source owned system, may cause series of influences on distribution system operation and security. Loss problem of distribution system is researched in this paper.First, the background, the characters, the actuality and the future of distributed generation are introduced in this paper. Then, the methods of classification of distributed generation and some common distributed generation technologies are introduced, such as wind-turbine system and photovoltaic cell system, fuel cell system and micro-turbine system.Based on simple model of distribution system, the influence of distributed generation on power losses is discussed and concluded in this paper, considering three related factors: the capability of distributed generation, the location of distributed generation and power factor. Models of several common distributed generations in power flow calculation are analyzed. The back/forward sweep algorithm is improved using injecting reactive power, which can dispose PV node effectively and is available to calculate distribution power flow with distributed generations. The improved back/forward sweep algorithm is proved effectively by simulating on a 33-node distribution network.Three optimal methods of genetic algorithms, tabu search and improved tabu search are compared and researched on optimal allocation of distributed generations in distribution system in the paper. Tabu search is improved with mutation of genetic algorithms. Objective function of loss minimization and code method are advanced. By simulating on a 33-node distribution network, the capability and the location of distributed generations are calculated optimally. By comparing the optimal results of the two methods, improved tabu search application for finding the optimal allocation of DGs from a viewpoint of loss minimization is better than genetic algorithms.Considering the power of wind-turbine system depending on the randomness of wind speed where the wind turbine is located, an index based on scenario analysis is presented. Power loss is taken into consideration in this index. A new model including the optimal index for reconfiguration is proposed. And a new node numbering method which can be applied to reconfiguration and programming is advanced. The method can judge whether the solution is feasible. Finally, by simulating on a 33-node distribution network, the model of reconfiguration, new node numbering method and improved Tabu search are feasible.