Reduced-Order Modelling of LTI Systems by Using Routh Approximation and Factor Division Methods

• Published in: Circuits, systems, and signal processing Vol. 38; no. 7; pp. 3340 - 3355
• Main Authors: Prajapati, Arvind Kumar, Prasad, Rajendra
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2019; Springer Nature B.V
• Subjects: Approximation; Circuits and Systems; Continuous time systems; Electrical Engineering; Electronics and Microelectronics; Engineering; Instrumentation; Mathematical models; Model reduction; Reduced order models; Short Paper; Signal,Image and Speech Processing; Routh Hurwitz table; Order reduction; Lower-order modelling; Transfer function; Higher-order system
• ISSN: 0278-081X; 1531-5878
• DOI: 10.1007/s00034-018-1010-6

In this paper, a new model reduction technique for the large-scale continuous time systems is proposed. The proposed technique is a mixed method of Routh approximation and factor division techniques. In this technique, the Routh approximation method is applied for determining the denominator coefficients of the reduced model and the numerator coefficients are calculated by the factor division method. The proposed technique has two main advantages as it gives the stable reduced-order model if the original model is stable and ensures the retention of first “ r ” number of time moments of the actual system in the r th-order reduced system. This method is also applicable for those systems for which Routh approximation method fails. To illustrate the proposed method, a real-time system model is reduced where the reduced model retains the fundamental properties of the actual model. In order to examine the efficiency, accuracy and comparison to other existing standard model reduction methods, the presented technique has been verified on two standard numerical examples taken from the literature.