Noise-like data low-frequency oscillation identification method based on double-covariance random subspace

Abstract

The invention discloses a noise-like data low-frequency oscillation identification method based on a double-covariance random subspace. The noise-like data low-frequency oscillation identification method comprises the steps: 1), processing a noise-like signal of a power system through the double-covariance random subspace, obtaining poles of two groups of features, and defining the poles as a verification group H1 and a reference group H2; 2) screening same-order poles of the two groups of poles to obtain physical poles, and forming a stable graph; and 3) performing system clustering on physical poles obtained through screening so as to obtain the final real modal parameters. According to the noise-like data low-frequency oscillation identification method, the defects that an existing recognition method is insufficient in data size and cannot automatically determine orders are overcome, and efficient and accurate beforehand low-frequency oscillation parameter recognition is achieved through the double-covariance random subspace and the system clustering method.

Description

technical field [0001] The invention relates to the technical fields of low-frequency oscillation parameter identification, signal processing and mode identification, in particular to a noise-like data low-frequency oscillation identification method based on double-covariance random subspace. Background technique [0002] At present, the scale of the power grid continues to expand, the degree of interconnection between systems continues to increase, and more and more high-magnification fast excitation devices are put into use, resulting in a higher probability of weakly damped low-frequency oscillation (LFO) in the system [1]-[2]. The appearance of weak damping or negative damping LFO greatly endangers the stability of the power grid, and also limits the maximum transmission capacity of the interconnection system. Therefore, monitoring and analyzing low LFO modal parameters is of great significance to ensure the safety and stability of the power system. [0003] At present,...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and propose a low-frequency oscillation identification method for noise-like data based on double covariance random subspaces, which breaks through the shortcomings of the existing identification methods based on post-identification under obvious disturbances, and utilizes dual The stochastic subspace of covariance and system clustering method realize efficient and accurate identification of low-frequency oscillation parameters beforehand, and improve the effective information for the control strategy of suppressing low-frequency oscillation

Images