Rejection of unknown periodic load disturbances in continuous steel casting process using learning repetitive control approach

This paper describes the design and implementation of a discrete-time repetitive controller for the continuous steel casting process that is used for the rejection of periodic load disturbances with an unknown period. A discrete, recursive scheme is applied to identify the unknown disturbance period and to adaptively change the period setting of the repetitive control algorithm. An experiment was conducted on a full-scale industrial-grade water model. A comparison of the disturbance rejection properties of the repetitive controller with those of PI control is presented. Results that indicate the ease and effectiveness of manual tuning for the rejection of periodic disturbances are then given. Finally, an online adaptation of the repetitive controller for the rejection of a disturbance with an unknown period is demonstrated.