Batch-to-Batch Iterative Learning Control of a Batch Polymerization Process Based on Online Sequential Extreme Learning Machine

Abstract

This paper develops a batch-to-batch iterative learning control (ILC) strategy based on online sequential extreme learning machine (OS-ELM) for batch optimal control. On the basis of extreme learning machine (ELM), a data-based nonlinear model is first adopted to capture the batch process characteristics aiming to obtain superior predictive accuracy. Subsequently, due to the model−plant mismatch in real batch processes, an ILC algorithm with adjusting input trajectory by means of error feedback is employed focusing on the improvement of the final product quality. In order to cope with the problems of the unknown disturbances and process variations from batch to batch, when a batch run is completed, OS-ELM is utilized to update the model weights so as to guarantee the precision of the model for optimal control, which corresponds to a nonlinear updating procedure. The feasibility and effectiveness of the proposed method are demonstrated via the application to a simulated bulk polymerization of the styrene b...