Adaptive on-line model for aerobic Saccharomyces cerevisiae fermentation

Area of Science:

• Biochemical Engineering
• Process Control
• Mathematical Modeling

Background:

• Fermentation processes require effective study and control methods.
• Indirect on-line measurements and mathematical models are crucial for process monitoring.
• Existing models may lack adaptivity and real-time predictive capabilities.

Purpose of the Study:

• To develop an adaptive mathematical on-line model for fermentation processes.
• To enable real-time estimation of process states (biomass, substrate, products).
• To facilitate model-based automation and validation of measurement variables.

Main Methods:

• Development of a model based on atom and partial mass balances and acid-base system equations.
• Incorporation of transport equations for mass transfer and unstructured expressions for fermentation kinetics to create an adaptive model.
• On-line state estimation using balance equations and measurement of input/output flows.
• Recursive regression analysis for on-line parameter estimation in transport and kinetic expressions.
• Application to Saccharomyces cerevisiae cultivation in a laboratory fermentor.

Main Results:

• The adaptive model successfully estimated the state of the fermentation process in real-time.
• Model-based predictions closely correlated with experimental data, validated by High-Performance Liquid Chromatography (HPLC) analyses.
• The model enabled effective prediction of the fermentation process.

Conclusions:

• The developed adaptive mathematical model provides a robust tool for studying and controlling fermentation processes.
• On-line state estimation and prediction capabilities facilitate process automation and improve monitoring accuracy.
• The model's effectiveness is demonstrated through successful application to Saccharomyces cerevisiae cultivation.