A Design of Self-Tuning PID Controllers Based on the Generalized Minimum Variance Control Law

Abstract

Self-tuning control schemes need many control parameters corresponding to the order and the time-delay of the controlled object. Using such many control parameters yields the inferiority in the transient property. Therefore, in order to design practical self-tuning controllers, it is seemed that it is desirable to construct control systems with as few parameters as possible.In this paper, we propose a method how to tune PID parameters adaptively, that is, a self-tuning PID control scheme, based on the relationship between a generalized minimum variance control (GMVC) law and a PID control law. Furthermore, we consider a design scheme of user-specified parameters included in the cost function of the GMVC. Finally, we apply the proposed scheme to an air pressure control system in order to investigate the effectiveness.