Trajectory tracking of batch processes with varying control interval and incrementally updated models

Abstract

A practical methodology for the tracking control of product quality in batch processes is proposed based on a run-to-run iterative learning control (ILC) strategy. The number of control intervals in a batch run is set to be increased from run to run. A batch-wise linear time varying (LTV) perturbation model is initially built based on only two batches of historical batch runs and control actions for the new batch run are calculated using ILC. The LTV model is updated incrementally by using the new data set from the new batch run while the number of control intervals in a batch run is increased after each batch until reaching a pre-set maximum value. The tracking error can converge and the performance of trajectory tracking is gradually improved by the ILC. The proposed methodology is illustrated on a simulated batch polymerisation process.