Study On Distribution Network Reconfiguration Based On Improved Genetic Algorithm

Distribution network is a link from generators to users. It lies in the central areas among the loads, supplying energy to users directly. When power is flowing through the net work, every component in distribution network is sure to consume some energy, especially in the distribution system leveled 10Kv, occupying a great proportion of the whole energy loss. Then great attention should be paid to the optimal operation of the distribution network.Distribution network reconfiguration is a useful way to the loss reduction and a important means to distribution system control. This paper aims at 10 kV distribution network, which objective function is the least line loss over a period of time. This paper researches distribution network reconfiguration based on improved genetic algorithm (IGA).After multifactorial analysis of many distribution network power flow algorithms, computing power loss using hybrid power flow calculation method is put forward. On the processing of load forecasting data, the paper uses the geometric means which is come from cubic spline interpolation method to represent a particular load between the two times. The method is more accurate than previously proposed algorithms.Based on improved genetic algorithm, bad body as colony strategy is proposed to avoid the best individual control of the selection process too strong, abandon the optimal mode and impact the global optimization algorithm performance at the beginning of the alogrithm. In addition, the paper puts forward topology method to judge the unfeasible solution and offer the mathematics model for the forward and backward substitution method.Finally, VB 6.0 and Microsoft SQL 2000 are used as platform in the paper. Through the simulation computation of IEEE 33 nodes system,PG&E 69 nodes system and part of 10kV distribution network in FengMan section of JiLin city, the results show that the proposed approach is rational and effictive.