Detection of Step Changes in Real-World Synchrophasor Measurements

This paper presents a practical method for detecting step changes in real-world synchrophasor measurements based on three fundamental theories. Step changes can be caused by shunt switching, generator set point changes, and other changes in grid apparatus, and detecting them is important for understanding the grid’s response to inverter-based resources. To precisely localize step changes, a nonorthogonal discrete wavelet transformation (DWT) based on smoothed gradient estimation is used. A detector based on a multiscale point-wise product of wavelet coefficients is proposed, which takes advantage of the broadband characteristic of step changes in the wavelet coefficient space. This product also suppresses undesirable signal components, thereby reducing false positives. Finally, Rosin’s unimodal thresholding is used to provide an adaptive threshold for the step detector. The effectiveness of the proposed approach is demonstrated on synthetic signals and real-world synchrophasor data obtained from Dominion Energy’s power grid.