Research On Parameter Identification Of Synchronous Generator Under Different Short-circuit Fault

With the interconnection of the national grid, the state monitoring of grids between different regions will become very important. As the generator is an important component in the grid, whose operation state will produce a great influence on the whole electrical power system, the identification of parameters of the generator by application of the data measured by PMU devices will provide data basis for the stability and security dispatch of the electrical power system. In order to obtain a more accurate and reasonable identification parameter, and make the structure of motor model closer to the actual motor, the identification objects applies the highest sixth order model in the practical application to carry on parameter identification under different short circuit fault disturbances.Firstly, this paper studies the identifiability of the synchronous generator, deduces the identifiability of sixth order model electrical parameters in the d and q-axis by use of analytical method, and conclude that all the electrical parameters of the sixth order model are identifiable. Secondly, the method of trajectory sensitivity is applied to verify the analysis, which provides a basis for parameter identification. Then, the paper presents the time domain analysis method based on nonlinear least squares method for parameter identification of the synchronous generator, studies the selection of identification initial values, discusses the processing method of steady parameters, and states practical process and method of specific identification. In order to compare the effect of fault types on the parameter identification results, a Simulink single-machine system is built to simulate the short circuit fault of synchronous machine, by means of which the actual identification data before and after the disturbance could be acquired. Through writing the corresponding identification program of the synchronous generator, the identification results under three kinds of fault disturbance can be verified, which proves the correctness and rationality of the identification method.