Research On Power System Transient Stability Prediction Method Based On Wide Area Measurement Signals

With the gradually increasing of the power system, the structure and operationmode of the grid was more and more complex, the security and stability issues had beenprominent gradually. Due to fast transient stability and emergency control measurement,the quickly analysis of grid transient stability after disturbance occurring was veryimportant for preventing further expanding of the accidents. The phasor measurementunit (PMU) based on GPS technology could capture the grid status information quickly.It provided a new technology platform to solve the power system security and stabilityproblem, and provided a new research approach for the transient stability prediction.This thesis described a variety of transient stability prediction method which basedon the PMU measurements data, and analyzed the characteristics and disadvantages ofthose methods. Meanwhile, there existed the problems in system disturbed trajectoryprediction methods which introduced the virtual load, that a single mechanism modelwas difficult to describe the load characteristics and the portfolio variety of mechanismmodel was difficult to identify the parameter. As a compromise, the thesis attempted tointroduce non-load mechanism model. Combined with simulation results, the researchverified the effect of the non-load mechanism model which introduced the load, andanalyzed that transient stability prediction method existing the problem of predictingtime when in the practical application of engineering.According to the strict necessary and sufficient conditions (the complementarygroup of group interpersonal skills-base Standards, CCEBC) for the stability ofnon-autonomous non-linear multi-rigid operating system, transformed themulti-machine system to two machines system by reduced-order transformation,simplified the system model, and realized the fast predictive simulation and stabilityanalysis of the power system. At the specific process, based on the clustering model andthe relationship of the unstable modes, the critical generators identification using doublecriterion of the composite power angle and composite rotor angular velocity improvedthe capturing rate of the critical cluster identification and enhanced the reliability of thetransient stability prediction, For the two-machine system after reduced-order, the thesisanalyzed the effect of time-varying nonlinear character in power system to the disturbedtrajectory prediction, directly identified the dynamic differential equations parameterswhich describing the system transient process, to achieve the purpose of the system prediction and simulation. At the same time, the study combined the new stabilitycriterion, which was based on the disturbed trajectory bump identification, had thepredictive ability, and given a power system transient stability prediction methods. Thesimulation proved that the method could improve the accuracy and rapidity of thetransient stability prediction effectively.Finally, the thesis summarized the research work of the full text, and proposed thestudy contents further to achieve fast and stable controlling requirement in powersystem.