A Combined State-space Identiication Algorithm Applied to Data from a Modal Analysis Experiment on a Separation System

This paper discuss identiication of state-space models from impulse response or initial value experiments. We combine Kung's realization algorithm with classical nonlinear parametric optimization in order to further reene the quality of the estimated state-space model. We apply these ideas on real data originating from a modal analysis experiment on a separation system. The results indicate that the parametric optimization step increase the model quality signiicantly compared with the initial model the realization algorithm provides.