Distribution Network Dynamic Reconfiguration Based On Rare Measured Data

Distribution network reconfiguration problem is a large,mixed and nonlinear optimization problem.Under load balancing, the aim of the distribution network reconfiguration is to reduce system power loss and to satisfy capacity constraint and voltage constraint, which is realized by changing status of switches.The vast majority of distribution network reconfiguration is to reduce system power loss, which takes a lot of data of distribution transformers.In China,distribution network load(distribution transformers) are not generally measured,the voltage and power can only be got from FTU.So,measurement data of distribution network is seriously lacked.Loading balance can be conducted as the goal for distribution network reconfiguration by FTU.In this paper, distribution network reconfiguration which aims to load balance and distribution network reconfiguration which aims to reduce system power loss are correlatively analysized.The results show that they have strong correlation.So the former can instead of the latter to conduct distribution network reconfiguration.Distribution network reconfiguration can be divided into static and dynamic reconfiguration according to optimization time.On that base of the correlation analysis,a time recursive merging method is proposed.The proposed method can make the most use of the existing research results on static reconfiguration to solve dynamic reconfiguration.In this method,a time zone is divided by one hour and distribution network reconfiguration is conducted in each hour. Whether the switch operation is to meet the constraints can be determined. If not , the time zone is divided by two hours.The switch operation is determined.If the switch operation is not to meet the constraints,the time zone is divided by three hours and so on.Until the switch operation meets the constraints, the best option is got.The program was written out.Examples were given to show the feasibility of the proposed methods.