A New Spectral Estimator for Identifying Dominant Modes and Detecting Events in Power Systems

In power systems, electromechanical modes are of interest due to their relationship to system stability. Under ambient conditions, spectral analysis methods can be applied to synchrophasor measurements to examine the modes. However, the presence of forced or ringdown oscillations in the data can make it difficult to appropriately interpret results from conventional methods. The new spectral estimator proposed in this paper addresses this challenge by automatically detecting and removing ringdowns from consideration, while also suppressing the spectral content of forced oscillations (FOs). This new estimator is termed the Robust Modified Daniell-Welch (RMDW) periodogram. Applications of the RMDW estimator include monitoring changes in the system's dominant modes and detecting large system events. The methods are validated with simulated and measured power system data.