Research On Identification Algorithm Of Sustained Low Frequency Oscillation In Power System

With the application of new energy sources and the interconnection of large power grids,low frequency oscillations in power systems often occur.At present,a lot of research results have made a meaningful study on the mechanism and identification of low-frequency oscillation phenomena from the perspective of transients.In addition to analyzing the low-frequency oscillation phenomenon of amplitude exponential decay from the perspective of transients,the amplitude of the voltage or current signal of the modern power system under steady-state conditions always has a certain oscillation.Therefore,in order to understand the state of the power system and its changes,it is necessary to perform online identification of such continuous low frequency oscillation signals.This dissertation has carried out related research on this issue.It mainly includes the following contents:At first,this dissertation introduces the research background and significance of the issue,as well as the current research status of low-frequency oscillation analysis methods at home and abroad and points out the necessity of studying the phenomenon of low frequency oscillation in steady state angle.Then the signal model of the continuous low frequency oscillation in the power system and the identification task of the continuous low frequency oscillation signal are given.Through in-depth analysis of the continuous low-frequency oscillating signal model,it is known that commonly used frequency estimation algorithms can be used to identify continuous oscillation low-frequency signals and introduce the advantages and disadvantages of several commonly used algorithms.Then,based on the existing normalized adaptive notch algorithm,an improved low-frequency online identification method based on adaptive notch is proposed.The slow integration manifold,the averaging method,the Lyapunov stability theorem and the perturbation system theorem are used to analyze the stability and noise of the improved algorithm.The simulation of the mixed input with continuous low frequency oscillation signal is carried out and compared with the original algorithm to verify the effectiveness of the improved method.Theoretical analysis and simulation results show that under the case of low frequency oscillation frequency unknown and jumping,the proposed algorithm can accurately decompose the fundamental and low frequency oscillation components of the voltage/current signal,as well as the fundamental amplitude,oscillation amplitude and oscillation frequency.Next,the detector design based on the improved algorithm is given.The signal acquisition is realized by the AD7606 chip with six channels of synchronous sampling.The TMS320F28335 completes the calculation and processing of the signal.The DAC7725 N chip not only realizes the DA conversion of the fundamental wave and the oscillation component,but also perform the continuous waveform analog output.The relevant modules in the hardware design is given and include: DSP minimum system,power conversion circuit,AD detection acquisition circuit,DA conversion circuit,communication module circuit.The relevant software design and flow chart are also given while giving the hardware design module circuit diagram.It focuses on the improved adaptive notch discrete-time algorithm implemented by the classical fourth-order Runge-Kutta algorithm and provides a certain engineering basis for the detection and analysis of voltage/current continuous low frequency oscillation.Finally,the research results and research significance of this topic are summarized,and further research and development directions are proposed for the inadequacies.