Skip to Main Content
This paper proposes a method for online identification of modes corresponding to low-frequency oscillations in a power system. The proposed method has considered the effect of colored Gaussian noise produced due to filters used for signal preprocessing. In order to mitigate the effect of colored Gaussian noise, the paper first proposes a modified total least square estimation of signal parameters via rotational invariance techniques (TLS-ESPRIT) that utilizes first and second rotational shift invariance properties of the signal. In the next step, the modified TLS-ESPRIT utilizes the signal transformed in an orthogonal basis. The proposed method has been compared with the improved Prony, the TLS-ESPRIT and the fourth-order cumulant-based TLS-ESPRIT (4CB-TLS-ESPRIT) using a test signal for identification of the modes at different noise levels. Robustness of the proposed method is established in the presence of colored Gaussian noise through Monte Carlo simulations. Estimation of modes for a two-area power system, using the proposed method, is carried out in the present work. Comparison of the proposed method with other methods is also performed on real-time probing test data obtained from the Western Electricity Coordinating Council (WECC) network.