Research On Optimal PMU Placement Method Based On Improving Observability Of Distribution Network

The trend that large-scale installation of distributed generations and electric vehicles changes the planning and operation mode of distribution network,and promotes the development of distribution network in a flexible,reliable,efficient and intelligent direction.The distribution network has limited measurements,and the accuracy of the data is not high.Phase measurement unit(PMU)can acquire voltage and current phasor measurements at high frequency,which can provide a secure,reliable and accurate online monitoring tool for the distribution network.This paper studies methods of optimal PMU placement in the distribution network.The main contributions are as follows:(1)Based on the type and characteristics of PMU measurements,the measurement system of distribution network is analyzed,and the methods of analyzing the observability and state estimation of distribution network are studied.On this basis,the characteristics and requirements of optimal PMU placement in the distribution network are given.(2)Aiming at improving the observability of distribution network,two methods of optimal PMU placement are proposed,which are applied to different scenarios.Considering the influence of multiple topologies and existing measurements,a new optimal PMU placement method which ensures the full observability is proposed.A network compression method based on pre-configuration is proposed to reduce the network scale and PMU candidate locations.When the number of PMUs provided can’t ensure the full observability,a method of optimal placement based on the limited number of PMUs is elaborated.A customized genetic algorithm is proposed to solve the model.All individuals become feasible solutions by performing customized crossover and mutation operations.(3)Aiming at improving the accuracy of state estimation of distribution network,a multi-objective method of optimal PMU placement is proposed,which considers supervisory control and data acquisition system and PMU measurements.The model takes the minimum number of PMUs and the lowest state estimation error as the objective functions.The non-dominated sorting genetic algorithm is applied to get the Pareto set and the optimal solution is selected by the technique for order preference by similarity to ideal solution.The methods are tested on IEEE standard bus systems and Southern Power Grid actual distribution system.The results indicate the feasibility and effectiveness of the proposed methods.