Development of a high-fidelity digital twin using the discrete element method for a continuous direct compression...

Area of Science:

• Chemical Engineering
• Pharmaceutical Manufacturing
• Computational Modeling

Background:

• Continuous powder mixing is crucial in pharmaceutical manufacturing.
• Predicting the impact of inlet concentration variations on outlet concentration is essential for process control.
• Previous work established reliable Residence Time Distribution (RTD) models using DEM and small-scale tests.

Purpose of the Study:

• To evaluate the predictive capability of a reduced-order model for a continuous powder mixer across an extended design space.
• To demonstrate the application of calibrated Discrete Element Method (DEM) models for predicting process responses to inlet concentration changes.
• To develop a computational tool for control strategy evaluation in powder mixing.

Main Methods:

• Utilizing a calibrated Discrete Element Method (DEM) workflow for simulations.
• Extrapolating DEM simulations using a relay race method.
• Parameterizing cumulative RTD curves with the n-CSTR model to develop reduced-order models.
• Validating predictions through experiments and simulations of step changes in API concentration.

Main Results:

• Calibrated DEM models accurately predicted the response of a continuous powder mixer to step changes in API inlet concentration.
• The n-CSTR model effectively parameterized predicted RTD curves.
• Reduced-order models enabled process simulation in seconds, significantly reducing computational effort.
• The relay race method proved effective for extrapolating DEM simulation results.

Conclusions:

• Calibrated DEM models serve as a powerful tool to guide and minimize experimental work in powder mixer development.
• The developed approach facilitates the establishment of adequate control strategies for continuous powder manufacturing.
• High-fidelity DEM simulations, combined with material-sparing characterization, offer a robust platform for control strategy evaluation.