Generalized minimum variance control design based on the Box-Jenkins model and its application to gas flow control system

Abstract

This paper deals with the problem of controller design based on Box-Jenkins model to achieve both regulating and tracking control. The greatest challenge of this problem lies in embedding the complicated noise term of the Box-Jenkins model into the control law. To solve this problem, this paper establishes a novel control method complemented by the generalized minimum variant control criterion, and introduces an ingenious design of Diophantine equation to deal with the complicated noise term. In addition, this control methodology can be applied to non-minimum-phase system and allows arbitrary pole placement that enables the designer to achieve the desired dynamic response. Also, the corresponding computation algorithm is designed to determine the polynomials in the controller. Finally, an application on gas flow control system is introduced to demonstrate the correctness and effectiveness of this control method.