Analysis of low frequency oscillations in power system using EMO ESPRIT

Identification of poorly damped low frequency oscillations present in the densely interconnected power system is of paramount importance to maintain its stable operation. Estimation of signal parameters via rotational invariance technique (ESPRIT) is a parametric method used for analysing such signals even under noisy conditions. However, this method requires precise information about the number of modes present in the signal. Hence, this work uses a combination of Exact Model Order (EMO) algorithm and ESPRIT for analysing these low frequency oscillations. The performance of the proposed method is tested using various synthetic signals with different levels of noise and PMU reporting rates. Further, the robustness of the proposed method towards noise resistance is compared with modified Prony, TLS-ESPRIT and ARMA methods. Finally, the proposed method is tested using real time probing test data obtained from Western Electricity Coordinating Council (WECC) network. Results reveal that the proposed method is accurate, precise and outperforms the other methods.