Modal Analysis And Source Locating Of Low Frequency Oscillation Based On Wide Area Measurment System

With the development of smart grid and the increasing scale of power systems in China,Estimating the characters of low frequency oscillations and locating the oscillation source accurately and fastly can ensure the security of large-scale interconnected power systems.The traditional methods on low-frequency oscillation analysis are based on the detailed model of the system.However,for the actual power systems,the operating state of the system is variable,and the model-based analysis method may not accurately reflect the true operating state of the system.Therefore,this paper focuses on modal analysis and source locating of low frequency based on wide area measurment data,which can correctly reflect the current operation condition of the power systems.Firstly,this paper introduces the Poly-Reference Complex Exponential(PRCE)method to analyze the low-frequency oscillations based on the ring-down signals after disturbance.The analysis result of the algorithm complexity shows that the PRCE method is faster than the SSI method in calculating speed.At the same time,the input signal of this method can be multiple groups,so the oscillation mode can be identified,which provides more reference parameters for low-frequency oscillation analysis.The simulation results of the 16-machine system show that the PRCE method can effectively identify the low-frequency oscillation mode.The comparison with the TLS-ESPRIT method shows that the PRCE method performs better in conditions with noise.At the same time,the PRCE method can be based on multichannel signals and can identify oscillation modes.The comparison with the SSI method shows that the estimation accuracy of the PRCE method and the SSI method are similiarly,but the identification efficiency of the PRCE method is higher.The results of identification based on PMU measured data show that the PRCE method can be used to identify the lowfrequency oscillation modes of the actual large power grid.Secondle,the input signal of the PRCE method must be a ring-down signal,and this signal can only be obtained after a disturbance occurs in the system.In order to identify the system's low-frequency oscillation mode in a wider time range,this paper combines random decrement technique and natural excitation technique with PRCE method to realize low-frequency oscillation mode identification based on random response signals.The analysis of the 16-machine system shows that the proposed method performs better than NExT-TLSESPRIT under conditions with noise.Estimating results using measured data shows that both RDT-PRCE and NEx T-PRCE methods can identify long-term oscillation modes of about 0.35 Hz in the power grid.Finally,a method for locating the oscillation source of multi-machine systems is proposed.The electromagnetic torque expressions of various generators in a multi-machine system are deduced,and it is found that in each oscillation mode,the electromagnetic torque can be decomposed into a damping torque and a synchronous torque.Based on this,an oscillation source positioning scheme based on decoupling mode is proposed.The 4-machine and 2-zone examples show that the proposed scheme can effectively locate the oscillation source in the multi-oscillation mode.Compared with the existing methods,the positioning method of the proposed method is more accurate.