Batch-to-Batch Iterative Learning Control for End-Point Qualities Based on Kernel Principal Component Regression Model

A batch-to-batch model-based iterative learning control (ILC) strategy for the end-point product quality control in batch processes is proposed in this paper. A nonlinear model for end-point product quality is developed from process operating data using kernel principal component regression (KPCR). The ILC algorithm is derived to calculate the control policy by linearizing the KPCR model around the nominal trajectories and minimising a quadratic objective function concerning the end-point product quality. To overcome the detrimental effects of unknown process variations or disturbances, it is proposed in the paper that the KPCR model should be updated in a batchwise manner by removing the earliest batch data from the training data set and adding the latest batch data to the training data set. The ILC based on updated KPCR model shows adaptability for process variations or disturbances when applied to a simulated batch polymerization process. Comparisons between KPCR model and principal component regression (PCR) model based ILCs are also made in the simulations.