An Identification Scheme for Wideband Oscillation Parameters Based on AR-EWT and Interpolated DFT

Recently, wideband oscillations have frequently occurred in renewable power systems due to the extensive integration of power electronic devices, posing significant threats to system stability and equipment safety. During wideband oscillations, multiple oscillation modes often coexist with frequencies ranging from a few hertz to several kilohertz. Accurately identifying these modes within a limited observation window presents significant challenges. This article proposes an identification scheme based on autoregressive empirical wavelet transform (AR-EWT) and interpolated discrete Fourier transform (IpDFT) to enable rapid and simultaneous estimation of multiple oscillation modes. First, the AR-EWT method is proposed to process wideband oscillation signals within different observation windows, effectively decomposing the wideband frequency signal into subsynchronous oscillation (SSO) and high-frequency oscillation (HFO) components. Subsequently, the parameters of the SSO and HFO modes are estimated using IpDFT with a Hanning window, which effectively suppresses spectral leakage and thereby enhances the accuracy of the estimation results. Simulation results validate that the proposed method performs exceptionally well in terms of rapidity, accuracy, and robustness, demonstrating superior performance compared to conventional methods.