Low frequency oscillation identification method for power system based on O3KID algorithm

Abstract

The invention relates to a low frequency oscillation identification method for a power system based on an O3KID algorithm. The method embeds an observer in the stochastic model of the power system, and uses the basic equation of the O3KID algorithm and the least squares method to estimate the Markov parameters and residuals of the observer, and transforms the stochastic system identification of the power system into an identification problem of the deterministic system. The introduced device is equivalent to the Kalman filter. The Hankel matrixes are respectively constructed by using the output of the observer and the residual time series. The orthogonal projection and singular value decomposition methods of the deterministic system are used to effectively identify the reduced-order modelof the power system, and accurately extract the frequency, damping ratio and vibration mode parameter information of the dominant mode of the low-frequency oscillation. The method provided by the invention is suitable for the low-frequency oscillation mode analysis of the power system for the WAMS synchronous measurement environment excitation signal and the transient ring-down signal, and the IEEE-39 node system simulation and the US Eastern Power Grid WAMS measured data analysis verify the effectiveness of the method.

Description

technical field [0001] The invention relates to the technical field of power system low-frequency oscillation analysis, in particular to a method based on O 3 KID algorithm based power system low frequency oscillation identification method. Background technique [0002] With the rapid development of modern power systems, the rapid excitation system is widely used in synchronous generators, the interconnection of large-scale power systems, and the proportion of installed capacity of new energy power generation systems continue to increase. It restricts the power transmission of the transmission line and affects the stable operation of the power system and the control of power quality. At present, the wide application of wide area monitoring system (WAMS), especially the continuous improvement of the sampling frequency and accuracy of phasor measurement units (PMU), provides a basis for near real-time identification of low-frequency oscillations based on data measurement. in...

AI Technical Summary

Problems solved by technology

The normalized innovation vector output by the Kalman filter has the characteristics of zero-mean white noise. Through the statistical analysis of the innovation power spectral density, the rapid detection of the change of the modal damping parameter of the low-frequency oscillation of the power system is realized, but the essence is still a kind of nonparametric identification method

Kalman filtering and its extension method analyze the transient low-frequency oscillation response signal in a noisy environment by establishing a linear filter, and extract the dominant oscillation mode parameters. An accurate power grid model is still required, and low-frequency oscillation analysis based on data measurement is not applicable.

Images