Publications

Abstract : This paper presents a robust dynamic approach for the monitoring and estimation of electromechanical oscillatory modes in the power system in real-time with less computational burden. Extensive implementation of phasor measurement units (PMU) and the utilization of advanced signal processing techniques help in identifying the dynamic behaviors of oscillatory modes. Conventional nonstationary analysis techniques are computationally weak to handle a larger quantity of data. This research utilizes the time-varying filter based empirical mode decomposition method for signal decomposition, which is highly tolerant to noise and computationally more robust. Low frequency modes are estimated by analyzing the power spectral density of the most suitable decomposed mode, the selection of which is done using correlation analysis. Instantaneous mode shapes of the signals are determined using cross-power spectral density functions, which will give the operator much information about the nature of oscillations and provide proactive steps to improve the operation of the power system. The proposed approach has been tested using signals obtained from two areas Kundur system and actual PMU data recorded from Power System Operation Corporation (POSOCO) Limited of the Indian power grid. The results confirm the superior viability and adaptability of the proposed approach in estimating the electromechanical modes. The instantaneous mode shapes are analyzed accurately with less computational complexity compared to the existing nonstationary strategies which are used for power system mode estimation.