Novel approach to simplification of large-scale systems

A novel hybrid approach to the simplification of large-scale systems is introduced. It is suggested that a high-order linear system can be approximated by a low-order model with time delay on the basis of the proposed mixed method. The denominator of the reduced model is determined by the Routh stability method, which gives rise to the desirable stability of the model produced. The numerator parameters are estimated by using identification techniques, and the time delay is optimally chosen according to an output-error criterion. It is shown that the reduced-order time-delay model produces a sufficiently good approximation as compared to models from other methods. The reduction procedure is fully computer-oriented. The performance of the approach is verified by numerical examples.