Research On Online Analysis Method For Dynamic Stability Of Islanded Distribution System Based On ?PMU

As the large-scale distributed generations(DG)integrate into distribution systems,random fluctuations and intermittencies could be found more frequently.Moreover,DGs are connected with power electronic interfaces,which significantly weaken the inertia of the system,making the dynamic stability problem more severe.When an emergency such as a fault occurs,the distribution system will be disconnected from the transmission system and switched to islanded mode,continuing to supply power to the important loads.However,the dynamic stability problem will be highlighted due to the lack of support from large power system.Therefore,it is critical to detect and analyze the dynamic stability of islanded distribution network.The traditional power system dynamic stability analysis mainly relies on the eigenvalue analysis method based on the small signal model,but this method requires a large number of accurate system parameters and can only be analyzed via offline.With the application of the ?PMU,massive accurate synchronized real-time data of the system can be obtained online,which provides effective support to the online analysis of the dynamic stability of islanded distribution system.In this paper,two kinds of signal analysis methods are performed to analyze two typical types of signals that could be collected by ?PMU in islanded distribution systems:the ringdown signals and the ambient signals.After obtaining the system oscillation mode information including the frequency and damping ratio,the dynamic stability level of the islanded distribution system is able to evaluated.Firstly,the causes of dynamic stability problems in the islands of distribution network are described.A simulation model of islanded distribution network based on PSCAD/EMTDC platform is built and the corresponding small signal model is constructed.The existence of dynamic stability problems of islanded distribution network is verified.Then,two commonly used signal analysis methods,Prony method and ARMA method,are introduced in detail.Both methods are used to analyze the ringdown signal and ambient signal obtained by the simulation model,and the analysis results are compared with the eigenvalue obtained by the small signal model.The applicability of the two methods in the dynamic stability analysis of islanded distribution system is compared.Finally,an idea of combining Prony method and ARMA model method to analyze the dynamic stability of islanded distribution system is proposed.According to the required features and the multi-thread mechanism,an online analysis module is designed and developed,thus realizing the real-time evaluation of the dynamic stability level of the system by using data collected by ?PMU.