System approximation by matching the impulse response energies (Q1093604)

A model reduction method based on matching the energies of the impulse response and of a suitable number of its derivatives is presented. In particular, the denominator parameters of the reduced-order transfer function are determined with reference to the impulse response of the system without zeros: then, the numerator parameters are computed by taking into account the contribution of the zeros. The method ensures the stability of the reduced model of a stable system and is computationally simple. The practicality of the method is illustrated by an example. The results obtained show that the time response of the reduced model satisfactorily fits the original system response.