It is important to enhance biopharmaceutical process development to handle more drug candidates, speed up development time and introduce integrated continuous processing. One way to do this is to introduce tools that integrate modelling with an experimental platform for chromatographic separation. The main idea with this work is to present and discuss a modelling procedure composed of a chromatography system for experiments and a software for experiment execution, model simulation and calibration. The control system automatically generates a mathematical model, performs a set of experiments, and do parameter estimation and model calibration of the generated model. The result is a control system that automatically generates a digital twin of a chromatography step with a minimal user input.<br/>This technique has been implemented in the Orbit-controller software. Orbit is an open, flexible, and extendable software mainly for control of laboratory equipment for integrated continuous downstream processes, such as multiple ÄKTA systems. Orbit generates an internal digital representation of the physical system it is connected to, and each object has a set of internal descriptions of how it behaves and interacts with the surroundings. One such internal description is a mechanistic model of the physical object. It means that Orbit has an embedded model generator with an associated simulator that automatically can simulate the physical system, making it possible for Orbit to simulate the experiment before it is performed. To increase the accuracy of the simulation of the process model, its model parameters need to be calibrated against experimental data.<br/>A model calibration procedure based on a set of experiments with associated parameter estimation method is executed in a sequence to generate a complete calibrated model of the system. This procedure will be presented and discussed based on some examples and on the purpose of automated modelling and digital twin generation with a minimum of user interaction.

Integration of modelling in chromatography control system for generation of digital twin

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.