Dynamic model validation via error indexes

An error index to support the validation stage of dynamic models is proposed in this paper. The result of the index quantifies the percentage of the relative difference between an experimental signal and an estimated signal. The index is calculated for several test signals, which are selected and designed to exhibit controlled differences in contrast with a reference signal. Also, other seven indexes are calculated and compared with the proposed one. The results of the proposed index are the same controlled differences in the test signals, allowing an accurate interpretation of the results in a validation stage of a dynamic model. Additionally, experimental data taken from an excitation system for a synchronous generator are used to evaluate and validate the proposed index.