The Approach Of Identification And Detection Of Power System Parameters Base On SSI

On the basis of a novel theory Stochastic Subspace Identification, this paper studied the power system low frequency oscillations, harmonic and inter-harmonic parameter examination, and the dielectric loss Angle detection. The main contents are as follows:A new method was proposed based on EMD and SSI towards mechanisms which cased the low frequency oscillation in many reasons. This method can identify the modal parameters of low-frequency oscillation according to measured data. When the system is running in normal operation or has small disturbance, SSI can get the weakly damped patterns, frequency, and damping and vibration mode directly; when the system is running in fault state or abnormal operation, the measured data were filtered and smoothed by using EMD, the corresponding state matrix was identified by SSI, parameters of low frequency oscillation were identified by modal analysis. Through analyzing numerical simulation signal, four machine two area white noise interference simulation signal and Northeast Grid big disturbance signal, we can get the oscillation frequency oscillation frequency, damping ratio and vibration model, identification error in 1%, satisfy the requirement of accuracy.A method of Inter-harmonic detection based on SSI and stabilization diagram was proposed in order to accurately detect the inter-harmonic parameters. The SSI could detect system parameters directly according to the measured data which contains system information.Firstly, the number of Inter-harmonic components and the order of the model were tested by the use of the stabilization diagram, and then the actual frequency of each Inter-harmonic in the original noisy signal was accurately detected by SSI algorithm. If the signal-to-noise ratio of signal is more than 10dB, the method of least squares was applied to estimate the amplitude of the various Inter-harmonics without de-noising. The harmonic analysis of numerical simulation signal, amplitude modulation signal and electric arc furnace current show the feasibility and effectiveness of the methodThis paper introduces a novel method based on Stochastic Subspace Identification and Least Square. Dielectric loss angle is one of the effective means to determine the status of the high-tension apparatus insulation. The key to detect dielectric loss angle is correct extraction of the fundamental wave of voltage and current signal. This method can be accurately extracted fundamental wave of voltage and current signal, and get the phase of fundamental wave, then obtain the accurately dielectric loss Angle. Making simulation validation in the variety of real value of dielectric loss angle, frequency fluxion, third-harmonics, direct bias, sampling frequency, sampling dot number, have no influence to SSI-LS.When white noise is higher than 25dB, its calculation accuracy requirements. And compare with harmonic analysis method, SSI-LS is better.Through the above three topics in the numerical simulation signal and example analysis, This method has stronger anti-noise ability, the faster calculation speed, the higher precision and less influence examination by external factors, It provides a new idea and method to low frequency oscillation, harmonic parameter examination and dielectric loss factor testing and other issues of research.