Online model-based redesign of experiments for improving parameter precision in continuous flow reactors

摘要

Online model-based redesign of experiments (OMBRE) techniques reduce the experimental effort substantially for achieving high model reliability along with the precise estimation of model parameters. In dynamic systems, OMBRE techniques allow redesigning an experiment while it is still running and information gathered from samples collected at multiple time points is used to update the experimental conditions before the completion of the experiment. For processes evolving through a sequence of steady state experiments, significant time delays may exist when collecting new information from each single run, because measurements can be available only after steady state conditions are reached. In this work an online model-based optimal redesign technique is employed in continuous flow reactors for improving the accuracy of estimation of kinetic parameters with great benefit in terms of time and analytical resources during the model identification task. The proposed approach is applied to a simulated case study and compared with the conventional sequential model-based design of experiments (MBDoE) techniques as well as the offline optimal redesign of experiments.