Based On Adaptive Chirp Mode Decomposition Method To Extract Low Frequency Oscillation Characteristics Research

With the continuous development of the power system field,the large area interconnection,the wide spread of new energy,the wide use of middle-distance power transmission and the wide use of power electronics,the current power system is characterized by strong complexity and nonlinearity.Therefore,the phenomenon of low frequency oscillation which damages the stability of power system occurs from time to time.The accurate extraction of the characteristic parameters of the low-frequency oscillation can help the field workers to make reasonable scheduling decisions,as well as the in-depth analysis of the occurrence mechanism of the low-frequency oscillation and the relevant design of the power system stabilizer.Firstly,the paper describes the core ideas of the occurrence mechanism of a series of low-frequency oscillations,such as negative damping,and briefly analyzes the existing methods of extracting parameters of low-frequency oscillations based on model classes and the methods of extracting parameters of low-frequency oscillations with the help of Wide Area Measurement System(WAMS).Under the background of the current power system structure,the disturbance signals in the low frequency oscillation time domain and frequency domain are analyzed.The disturbance types of low-frequency oscillation are divided,and the different modes of oscillation and the relative oscillation parameters are analyzed.Secondly,this paper proposes an Adaptive Chirp Mode Decomposition(ACMD)method that can preprocess low frequency oscillation identification using the data provided by Phasor Measurement Unit(PMU)in a wide area Measurement system.This method can realize the modal decomposition of multi-modal coupled non-stationary and nonlinear signals in noisy environment noise,and output several single modal components contained in the original signal.According to the single mode component obtained by decomposition,the parameter identification method of Hilbert Transform(HT)is used to extract the oscillation information that can describe the signal characteristics,such as the oscillation frequency and damping ratio of each single mode component obtained by decomposition.The instantaneous frequency was extracted by ACMD method.The validity of the method is tested with the aid of the self-synthesizing test signal.At last,the paper introduces the flow of low frequency oscillation from measurement to parameter extraction and control.Then,the system of IEEE4 unit 2 and IEEE 16 unit 5 are used to simulate the low-frequency oscillation environment of the power system to extract the low-frequency oscillation parameters and finally,the practicability is verified with the PMU measured data.The results of extraction and reconstruction are used to demonstrate the effectiveness and practicability of the proposed method in the extraction of low-frequency oscillation parameters.