Identification of excitation systems with the generator online

A methodology to identify excitation systems where the experiments to collect the data are carried out with the generator online and the excitation control system in closed loop is reported in this paper. In this methodology the IEEE standard model structures are considered for identification since they are widely adopted, and a pseudo random binary sequence signal is applied as a perturbation signal. The effect of online operation of the generator on the identification results of excitation systems is analyzed. Particularly, the effect of the closed loop configuration on the excitation system input signal in an identification experiment is discussed. The proposed estimation methodology is validated experimentally. An index to quantify the quality of the system input signal in contrast with the designed perturbation signal is proposed. In addition, an operation region in the generator capability curve, high active power and absorbing reactive power, where the identification experiment produces better identification results is located.

► A methodology to estimate excitation system parameters is proposed. ► The identification experiments are carried out when the generator is online. ► A region in the generator capability curve is suggested to carry out experiments. ► Generating high active power and absorbing reactive power is the suggested area. ► Experimental results validate the methodology and the suggested area.