Algorithms for estimation of model parameters of excitation system of an electrical machine

Estimation of parameters of mathematical model of an electromachine excitation system is presented, carried out by selected optimisation algorithms. The estimation is based on real and simulated measurement of waveforms obtained during test disturbances of the steady-state operation of the generating unit. Measurements of dynamic waveforms during step changes in the reference voltage of the voltage regulator of the generator operating in no-load state at the voltage controller operation (automatic control) were taken and recorded in Power Plant Rybnik. The problem of determining the considered model parameters is formulated in the form of the objective function, which was minimised using genetic, gradient and hybrid algorithms. MATLAB-Simulink package was used for computations.