Stepwise Parameter Fitting to Combine Industrial and Pilot Plant Datasets

• Published in: Computer Aided Chemical Engineering Vol. 53; pp. 901 - 906
• Main Authors: Becker, Per Julian, Celse, Benoit
• Format: Book Chapter Journal Article
• Language: English
• Published: Elsevier 2024
• Series: Computer Aided Chemical Engineering
• Subjects: Catalyst Deactivation; Chemical engineering; Chemical Sciences; Hydrotreatment; Industrial Data; Modeling; Catalyst Deactivation; Industrial Data; Hydrotreatment; Modeling
• ISBN: 9780443288241; 0443288240
• ISSN: 1570-7946
• DOI: 10.1016/B978-0-443-28824-1.50151-4

In kinetic model development for hydrotreatment processes industrial data are generally not used because deactivation must be taken into account, which is very difficult due to the high complexity of chemical phenomena. This is unfortunate because industrial data contains far larger feedstocks variation compared to pilot plant tests. The aim of this work is to propose an innovative method to include industrial data in kinetic model fitting. A stepwise parameter fitting method is proposed to use both pilot plant and industrial data. Pilot plant experiments provide robust data but with small feed variations, on the contrary industrial plant data provide huge feed variation. To obtain more robust models, a combined modeling framework for the kinetic reactor and deactivation models, solved simultaneously, is proposed for the HDN reaction in a hydrotreatment reactor. The kinetic parameters (reaction orders, activation energy) are calibrated on pilot plant points, while the empirical feedstock parameters as well as the deactivation model is calibrated on industrial points. This methodology leverages the strengths of each of the two datasets which results in more robust predictive models.