Research On Dynamic Phasor Measurement Algorithm For Power System

Synchronized Phasor Measurement Unit(PMU)provides accurate phasor estimation data for wide area measurement systems,monitors the dynamic process of the power system effectively,and improves the safe and stable operation of the power grid.Phasor estimation algorithm is the basis and core of PMU,the existing algorithm has good measurement accuracy under static conditions,but its measurement performance needs to be improved under dynamic conditions.In this paper,in order to improve the accuracy and real-time performance of phasor measurements,phasor measurement algorithm under different dynamic conditions is studied.For the dynamic signal with frequency deviation,an improved DFT algorithm suitable for frequency offset is proposed.Firstly,the error expression of phasor calculation in traditional DFT under frequency offset is deduced,and the error is divided into the sinusoidal phasor error and the fixed phase error.Then,the vir tual phasor of B phase and C phase is constructed by translating the current sampling window forward and backward to 1/3 fundamental period,and then the virtual positive sequence phasor is calculated to reduce the sinusoidal phasor error.Finally,considering the phase angle error of sinusoidal variation,the signal frequency is estimated by using the phase angle difference of the virtual phasor of phase B and phase C,so as to correct the fixed phase error.Through simulation verification of MATLAB numerical signal,compared with the virtual positive sequence DFT algorithm,this paper proposed method does not increase the sampling time delay,increase only very limited computation load,and it can not only accurately estimate the signal frequency,but also greatly reduce the fixed phasor error.For power oscillating signals and signals with sudden changes in amplitude or phase,a dynamic phasor measurement algorithm based on Strong Tracking Taylor-Kalman filter(STKF)is proposed in this paper.First,in or der to fully characterize the time-varying characteristics of the signal,a linear state space model for dynamic signals is established using a second-order Taylor expansion of the phasor.Then,in order to improve the ability of Taylor-Kalman filter(TKF)to track mutation signals and robustness to model uncertainty,the idea of strong tracking filter was introduced in which the estimation covariance matrix is adjusted adaptively according to the suboptimal fading factor based on the theoretical residual a nd the actual residual mismatch degree.Through simulation verification of MATLAB numerical signal,compared with the TKF algorithm,the STKF algorithm proposed in this paper has higher measurement accuracy and better dynamic response performance.In view of the existing phasor measurement algorithm that filters out decaying DC component has a defect which is only suitable for the case without frequency offset,time-varying characteristics of signal and decaying DC component are taken into account in this paper,so a strong tracking filter algorithm for dynamic phasor measurement considering decreasing DC component is proposed.Firstly,the decaying DC component is expressed by its second-order Taylor polynomial,and the first and second derivative of decaying DC component,as well as the angular frequency and amplitude of the fundamental component,are added to state variable matrix to establish nonlinear state space model of fault current,which can reduce estimation model error.Then,in order to ensure the tracking ability of the algorithm,using strong tracking filter to estimate each state variable recursively.Through simulation verification of MATLAB numerical signals,ATP-EMTP fault simulation signals,and actual fault recording signals,the proposed method can accurately estimate the attenuation of DC and improve the accuracy of phasor measurement.The algorithm is not only affected by the frequency offset,but also has a fast dynamic response.