A model discrimination based approach to the determination of operating regimes for chemical reactors

The operation of a chemical process that involves the interplay between chemical reaction(s) and transport phenomena can materialise in the occurrence of a number of different operating regimes. When developing a new process, to achieve successful scale-up, the operating regime which yields ideal performance in smaller scale experiments should be identified and retained in the full scale realization. In the past, experimental procedures have been proposed for identifying operating regimes based on the qualitative trends of the response of a process to the change in operating conditions. In this work, a quantitative approach is proposed, in which the problem of determining operating regimes is formulated as one of model discrimination. The proposed approach makes use of hybrid models to handle missing mechanistic knowledge and an optimal experimental design technique was applied to generate the most discriminative data. A simulated case study on the nitration of toluene demonstrates that, compared with existing qualitative methods, this approach has the potential to achieve sharper discrimination, impose fewer requirements on experimental facilities, and complement existing methods.