Detection of Oscillations in PMU Data Based on Multisynchrosqueezing Transform for Sub/Supersynchronous Oscillations

The current power system often uses phasor measurement units (PMUs) for fault data extraction, which is of great value for the analysis of sub synchronous and super synchronous oscillations in the power grid. In order to monitor sub/super synchronous oscillations in the power system online, facilitating fault diagnosis and power grid recovery, this paper introduces an automatic detection method for sub/super synchronous oscillations based on Multi Synchronous Compression Transform (MSST). This method mainly utilizes data obtained from PMU. Through PMU data, time and spectral information can be extracted, enabling the MSST algorithm to effectively process non-stationary signals while reducing sensitivity to noise. The application of the MSST algorithm includes using ridge tracking method for signal modal extraction and reconstruction. Finally, use the Total Least Squares ESPRIT (TLS-ESPRIT) to calculate the parameters of these patterns. Through the detection and identification of sub/ultra synchronous oscillations in doubly fed wind turbines, it has been shown that this method has better detection accuracy and anti-interference ability compared to traditional methods in identifying and analyzing sub/ultra synchronous oscillations in power systems.