Potential of predictive model-based dissolved oxygen control for intermittent fed-batch processes

Philipp Pably*, Jakob K. Huusom, Julian Kager

*Corresponding author for this work

Research output: Contribution to conferenceConference abstract for conferenceResearchpeer-review

6 Downloads (Pure)

Abstract

Ensuring sufficient dissolved oxygen (DO) levels during a cultivation is one of the most essential tasks in aerobic bioprocesses [1]. Therefore, control strategies are employed that act on the stirring speed, aeration rate and oxygen partial pressure in the reactor, often by utilizing PID algorithms of varying complexity [2]. While these controllers perform well in batch or continuously fed systems, their purely reactive nature flashes its limitations when challenged with abrupt changes in nutrient addition. This can occur during the transition between phases or with intermittent feeding profiles in high-throughput small scale multi-reactor systems, where a robot adds the substrate in intervals through bolus shots. The resulting sudden drops in the DO signal can lead to the system being oxygen limited due to the response time of the control loop and consequently lead to the organism changing its metabolism and physiological state [3].

This work investigates the opportunities that model-based predictive DO control algorithms offer to avoid limitations in oxygen supply during the process. The combination of elemental balances, simple growth kinetics and the Van’t Riet equation present enough tools to create a process model for the oxygen demand. Provided with the scheduled feeding profile it can predict the resulting metabolic activity and needed changes in stirring speed and aeration within a selected bioprocess. The impact and potential benefits over basic PID control are shown through in-silico simulations for a microbial intermittent fed-batch process.
Original languageEnglish
Publication date2024
Number of pages1
Publication statusPublished - 2024
Event8th BioProScale Symposium - Berlin, Germany
Duration: 9 Apr 202411 Apr 2024

Conference

Conference8th BioProScale Symposium
Country/TerritoryGermany
CityBerlin
Period09/04/202411/04/2024

Fingerprint

Dive into the research topics of 'Potential of predictive model-based dissolved oxygen control for intermittent fed-batch processes'. Together they form a unique fingerprint.

Cite this