The Study On The Identification Of Critical Lines/Buses And Controlling Of Power System Low-Frequency Oscillations

Power system dynamics is highly complex, especially the low-frequency oscillation is becoming a serious bottleneck for increasing power transfer.With the wide applications of WAMS, synchrophasors can be accurately measured and quickly transferred to the control center which makes it possible to identify and control oscillation online. Advanced mathematical tools with the potential to identify and characterize these dynamics in near real time have been applied very successfully to power system. But the higher requirements were put forward to the power system stbability analysis methods. And how to identify the Critical Link of oscillations becomes the important issue which is to be solved urgently.Based on the structure-preserving power system model, taking excitation, dynamic load model and power equation into account, the state space equation and state space matrix preserving network information was founded. The d-q transformer was avoided in the process of derivation. The novel method for calculating low frequency oscillatory active power increment distribution by employing the eigenvalues and eigenvectors of the linearized state matrix is presented. The method not only can be applied to calculate oscillatory active power increment distribution over generators, transmission lines and loads, but also can be used to analyze the effects of load and network structure on power system low frequency oscillatory. The concept of branch brittleness index and oscillations cutsets is given. The low frequency oscillatory phenomenon is analyzed by employing network information.The identification results of Prony and ARMA are influenced strongly by signal noise and system order, and the HHT lacks basically mathematical theory. Priory, the process of HHT is very complex and consumes a long time. Based on analyzing the localization of the traditional small signal stability analysis method and identify method, this paper introduces the stochastic subspace identification (SSI) into the overcomes the influence of signal noise and order. And the time consuming in calculating is shorter than the other methods. The parameters of PSS were designed using the results of stochastic subspace identification and residue method. Comparing with other identify method, the parameters of PSS can be designed only using the observational signal. So, this method can be used to analysis and control the low frequency oscillations on-line real-time.Simulations carried on four-machines example system testify the validation of the proposed method. Test signal and physical simulation data show the strongpoint of stochastic subspace identification used in power system low frequency oscillations analysis.