Analysis And Suppression Of Low Frequency Oscillation Of Prony Power System Based On Improved EMD De-noising

Research on the online oscillation feature identification algorithm is an important theoretical basis for realizing on-line monitoring of low frequency oscillation of power system and wide area damping control.The traditional empirical mode decomposition method has a large error in denoising the low frequency signal of low signal to noise ratio,which will seriously interfere with the accuracy of Prony algorithm.On the basis of wide area measurement system,a new method of improved EMD denoising Prony algorithm for parameter identification of low frequency oscillation of power system.The normalized autocorrelation function is obtained by using the intrinsic mode function(IMF)of the EMD decomposition of the low frequency oscillation signal of the power system.By using the characteristics of the normalized autocorrelation function,the noise dominant mode And the demarcation signal of the dominant mode of the signal,the noise-dominated mode is denoised and the signal-dominated mode is reconstructed to obtain the low frequency oscillation signal of the power system after denoising.The reconstructed signal is subjected to Prony analysis to extract the low frequency oscillation characteristic The The simulation results show that the improved EMD method has the advantages of low SNR low frequency oscillation signal denoising ability,and it is proved that the improved EMD method combined with Prony algorithm is proved by the simulation experiment of ideal signal,stand-alone infinite system and3-machine 9-node system The low frequency oscillation signal of the power system is extracted,which has the advantages of fast speed,high resolution and good fitting effect.The extraction of the low frequency oscillation of the power system is accurate.Finally,the power system stabilizer(PSS)is used to quickly suppress the low frequency oscillation of the power system.The simulation results show that the method is effective and effective in identifying and suppressing the low frequency oscillation.This paper comes from the Liaoning Provincial Key Laboratory Project: Power System Fault Identification and On-line Diagnosis(LJZS003).