A feasibility study of excitation system parameters identification based on measured perturbation

In consideration of safety, any methods using input signals are limited in on-line electric power system identification, thus the study of excitation system parameters identification based on measured system perturbation signals on field is provided with high practicality. In this paper, the excitation system and single unite simulation system are built under Maflab/Simulink, then genetic algorithm (GA) is used to identify the excitation system parameters with M serial pseudo random binary signal (PRBS) and measured perturbation signals respectively. The feasibility of excitation system parameters identification based on measured perturbation signals is proved through the spectrum analysis of different input signals based on multi-taper method (MTM) for the first time. The identification results from same perturbation signals with various sampling frequency and recording durations are also analyzed. The influence of sampling frequency and sampling period of time on parameters identification results is studied. Some of instructive advices on excitation system parameters identification based on measured data are provided.