Based On Second-order Synchroextraction Wavelet Transform And Empirical Envelope Demodulation Modal Analysis Of Mixed Oscillation

Hydropower generation is taking the lead in the development of renewable energy because of its fast response to load transformation.Large-scale hydropower stations and pumped storage power stations have been gradually applied in long-distance,inter-regional transmission and asynchronous interconnection systems,such as Southwest Power Grid and Yunnan Power Grid.This kind of delivery system can not only improve the power grid hydropower channel delivery capacity,but also reduce the structural security risks of long-chain power grid.After asynchronous networking,mixed low frequency and ultra-low frequency oscillations will appear successively in the system with high permeability of hydropower units.Therefore,how to extract the characteristic parameters of multiple types of oscillations signals accurately is of great significance for the safe and stable operation of power system.Aiming at the inaccurate identification of characteristic parameters of mixed oscillating signals,the existing Synchroextracting Transform(SET)is extended to wavelet transform and a second-order instantaneous frequency estimate is introduced to decompose the signals in this paper.Then the parameters of each group of oscillatory components are identified by using the empirical envelope demodulation technique.Considering that modal aliasing and information loss are common in the process of extracting signals with traditional analysis methods,a Second-order Sychroextracting Wavelet Transformer(SSEWT)algorithm is introduced in this paper.The wavelet transform can convert the signal waveform into the time-frequency domain,and the time-frequency coefficients form the timefrequency graph and form the ridge with the signal energy.The second-order synchronous extraction operator can process the fuzzy curve of the original time-frequency graph to select the time-frequency coefficients at the ridge and construct a clearer time-frequency spectrum,which has a certain anti-noise property while accurately obtaining the frequency of the signal to be measured.The mode of mixed oscillating signal can be accurately extracted.Then,the single component of SSEWT decomposition was extracted with the empirical envelope demodulation technique.The empirical envelope demodulation technique is proposed based on the empirical amplitude modulation and frequency modulation decomposition.The method is simple in calculation,does not require complicated procedures in the application process,and does not require special treatment to the extreme point,so it has good applicability.Finally,the proposed algorithm is applied to three sets of simulation including self-synthesizing signal simulation,10-machine 39-node system simulation and measured signal simulation,which proves the feasibility and accuracy of the proposed method.The simulation and comparison with Variational Mode Decomposition(VMD)and Short-Time Fourier Transform(STFT)algorithms show that the proposed method can get better time-frequency aggregation.Accurately decomposed signal components;Compared with Prony algorithm and Hilbert-Huang Transform(HHT)algorithm,it is proved that the proposed method can effectively suppress noise and end effects,accurately identify the characteristic parameters of mixed oscillating signals,and has certain advantages over traditional methods in processing accuracy.