Pole preserving model reduction using hankel-norm approximation

Abstract

In this paper, a model reduction for linear time invariant system whose pole are located in a specified disk on the complex left half plane is considered. This method is based on the Hankel-norm approximation using balanced realization and the linear fractional trasformation technique which transforms any disk region to the complex left half plane. And the error bounds is considered when pole preserving model reduction have done. In this case the error bound is larger than direct the Hankel model reducton. Hence there exist trade-off between pole preserving and reduction error bound.