Abstract

The present study focuses on building a model of a laboratory-scale bioreactor using genetic programming (GP) and optimizing
it for profit maximization. The glucose to gluconic acid bioprocess was employed as a case study. It is challenging to create
a reliable first principle-based model for the fermenter since it is a multiphase enzymatic bioreactor. On the other hand, data-
driven models lack explicability. Consequently, a general methodology has been developed in this work, in which a data-
driven approach, such as multi-gene GP, was used as a modeling tool, and the model was then post-processed to increase the
explainability of the model. The model was used because it could effectively represent the underlying physics of the system.
An acceptable model was constructed, and then it underwent optimization. This study looked at how to increase gluconic acid
yield, which has a big influence on how profitable the process is. By applying an evolutionary algorithm to the produced model,
an ideal solution was also discovered.