| Title |
Fast Estimation of Low Frequency Parameter for Real-Time Analysis in Wide Area Systems |
| Authors |
김은주(Kim, Eun-Ju) ; 심관식(Shim, Kwan-Shik) ; 김용구(Kim, Yong-Gu) ; 김의선(Kim, Eui-Sun) ; 남해곤(Nam, Hae-Kon) ; 임영철(Lim, Young-Chul) |
| Keywords |
Time signal ; Low frequency ; Parameter estimation ; Fourier spectrum ; Discrete Fourier transform |
| Abstract |
This paper presents a Fourier based algorithm for estimating the parameters of the low frequency oscillating modes. The proposed methods estimates various parameters(frequency, damping factor, mode magnitude, phase) by fitting Fourier spectrum and phase with a damped exponential cosine function. Dominant frequency is selected by taking frequency corresponding to the peak spectrum, and damping factor is estimated using the left/right spectra of Fourier spectrum. In addition, mode magnitude is calculated by the normalized peak spectrum, and phase is estimated from spectrum phase. Also, we introduce an accuracy index in order to determine the accuracy of the estimated parameters, and the index is calculated using the deviations of the peak spectrum and the left/right spectra. The parameter estimation methods proposed in this paper include very simple arithmetical processes, so the algorithms are simple and the calculation speed is very fast. The proposed methods are applied to test functions with two dominant modes. The results show that the proposed methods are highly applicable to low frequency parameter estimation. |