Model and Controller Order Reduction for Infinite Dimensional Systems

Abstract

This paper presents a reduced order model problem using reciprocal transformation and balanced truncation followed by low order controller design of infinite dimensional systems. The class of systems considered is that of an exponentially stable state linear systems (A, B, C), where operator A has a bounded inverse, and the operator B and C are of finite-rank and bounded. We can connect the system (A, B, C) with its reciprocal system via the solutions of the Lyapunov equations. The realization of the reciprocal system is reduced by balanced truncation. This result is further translated using reciprocal transformation as the reduced-order model for the systems (A, B, C). Then the low order controller is designed based on the reduced order model. The numerical examples are studied using simulations of Euler-Bernoulli beam to show the closed-loop performance.

Keywords

• balanced truncation
• infinite-dimensional systems
• low order controller
• model reduction
• reciprocal transformation