A New Two-step Method to Estimate the Sequence Parameter of Transformer Based on Median Criterion

This paper proposes a new two-step robust identification method for the transformer sequence parameter based on median estimation, the method combines multiple sets of data of different time to identify the transformer's ratio and positive sequence parameters step by step and uses the median method to realize identification. First, this paper establishes equivalent circuit model of the transformer which depends only on the PMU (Phasor Measurement Units) measurements and the identified parameters on each side. Second, the actual ratio can be estimated by the median method, then this paper combines the estimated ratio, least square method and median criterion to identify the parameters. This method can effectively reduce the impact of noise and biased data. It is worth noticing that this paper eliminates the neutral point voltage in the model so as to solve the problem that the neutral point and the excitation branch of the transformer are not measurable or the low voltage side of the three-winding transformer is not measured. Finally, the simulation results show the effectiveness of the proposed method and the anti-noise and robustness of the algorithm, and the practicality in engineering is proved by the field PMU data.