A Grey Box Distributed Parameter Model for a Continuous Vibrated Fluidized Bed Dryer in Pharmaceutical Manufacturing

Abstract

Drying of pharmaceutical mixtures is an essential process in the wet granulation route of the continuous manufacturing of solid dosage form pharmaceutics. The moisture content of the material after the drying process is one of the Critical Quality Attributes (CQAs) of the product. Consequently, modeling of the drying process is critical for the prediction, monitoring and control of the continuous manufacturing process. This paper presents a grey box distributed parameter model for the state of the art continuous Vibrated Fluidized Bed Dryer (VFBD) present in the lab scale unit QbCon®1 developed in the R&D of the company L.B. Bohle. The model is derived using first principles and enhanced with data driven parameterization using Gaussian Process Regression (GPR). The model derivation, formulation, and numerical solution are presented and illustrated. Finally, the model is validated on the real VFBD, where it shows its capability of producing accurate predictions of the drying process variables with reasonable computational cost.

Authors 3

1. Ahmed Elkhashap Aachen Institute of Automatic Control (IRT)

   RWTH Aachen University

   Affiliation as printed

   Institute of Automatic Control, RWTH Aachen University, Aachen, Germany

2. Robin Meier
   Affiliation as printed

   L.B. Bohle Maschinen+Verfahren GmbH, Ennigerloh, Germany

3. Dirk Abel Aachen Institute of Automatic Control (IRT)

   RWTH Aachen University

   Affiliation as printed

   Institute of Automatic Control, RWTH Aachen University, Aachen, Germany