Parameter estimation of excitation systems from sampled data

The parameter estimation of excitation systems using a proposed indirect stochastic method is presented. In this method, the discrete-time ARMA model corresponding to each block of the excitation system is first identified from sampled input/output data. Then, the proper reduced order of the continuous-time model is determined using the concept of dominant energy modes. Finally, the parameters of the continuous-time model are found by matching its frequency response to that of the identified ARMA model. The proposed method is used to estimate the parameters of an excitation system in a pumped storage power plant. The results show that the estimated continuous-time models are rather close to those supplied by the vender.>