基于改进同步重分配变换的风电场次/超同步振荡参数辨识OA

With the large-scale integration of wind farms into power grids,the issues of sub-synchronous and super-synchronous oscillations have become increasingly prominent.Accurate identification of oscillation parameters is of great significance for ensuring the safe and stable operation of power system equipment.However,existing identification methods generally suffer from poor noise robustness and modal aliasing problems.To address these issues,this paper proposes an improved synchro-reassigning transform(ISRT)-based decomposition method,combined with the Hilbert transform(HT),for the identification of sub-synchronous and super-synchronous oscillation modal parameters. Firstly,the wide-area measurement data are processed by short-time Fourier transform(STFT)to obtain the time-frequency coefficient matrix of the signal.Subsequently,noise-induced spurious modes are removed by applying mode energy weight.Then,a three-step selection rule is applied to extract the instantaneous frequency trajectories of the oscillation modes from the time-frequency matrix,enabling accurate separation and time-domain reconstruction of each mode.Subsequently,the Hilbert transform is employed to accurately extract the characteristic parameters of each oscillation mode,including oscillation frequency,damping factor,and amplitude.Finally,the effectiveness of the proposed method is validated through tests on synthetic signals,electromagnetic transient simulation signals,and field-measured power grid data.The simulation results demonstrate that the proposed method achieves higher identification accuracy and better noise robustness compared with STFT,SET,FSST and MSST algorithms. The main conclusion of this paper can be summarized as follows: (1)The proposed method extends one-dimensional time-domain signals to the two-dimensional time-frequency domain for analysis.By introducing a mode energy weight threshold,it effectively eliminates noisy pseudo-modes.Furthermore,a three-step selection criterion is applied to retain time-frequency coefficients corresponding to true oscillation modes.This approach significantly mitigates the scale ambiguity issue inherent in traditional time-frequency analysis methods,thereby improving the accuracy of time-frequency decomposition for measured signals and enhancing the precision of sub/super-synchronous oscillation parameter identification. (2)Compared with existing methods,such as STFT,FSST,SET,MSST,the proposed approach achieves higher concentration in the time-frequency representation,offers superior time-frequency resolution,effectively alleviates mode mixing and redundant information interference,and exhibits strong robustness against noise.These advantages contribute to improved accuracy in identifying sub/super-synchronous oscillation parameters under high-noise conditions. (3)The proposed method was validated using self-synthesized signals,electromagnetic transient simulation signals,and real-world power grid sub-synchronous oscillation data.The results demonstrate that the proposed method can accurately identify sub/super-synchronous oscillation parameters.