A Trajectory Based Waveforn-dividing Method for Recognizing Dynamic Performance of Power System

The analysis of the dynamic process is with vital importance to the security of power system. The model and parameter recognizing is prerequisite for the exact simulation of the process. As the models and parameters used in the simulation are not accurate, the simulation result will not cope with the real system trajectory. With the rapid development of wide area measurement system (WAMS), some real time dynamic trajectories of power system can be measured and recorded, Based on which a new approach of modeling and parameter recognizing can be proposed. In this paper, a waveform-dividing method, based on disturbed trajectory of power system has been proposed. The oscillatory mode in every period of the dynamic process is obtained by the analysis of measured trajectory. By comparing the difference of the simulated trajectory and the measured trajectory, the recognized models and parameters can be validated. The two case studies show the validity and simplicity of the waveform-diving method and its superiority to the Prony algorithm.