Real-Time Feed-Rate Optimization in 3-HP Fermentation Using Digital Twin Technology

In this study, we present the development and validation of a real-time feed-rate control algorithm designed to optimize the production of 3-hydroxypropionic acid (3-HP) in fed-batch fermentation processes. By integrating in situ near-infrared (NIR) spectroscopy with a partial least-squares (PLS) model, we achieved precise real-time monitoring of glycerol and 3-HP concentrations. A kinetic model incorporating the Monod equation and enzyme inhibition kinetics was employed to characterize fermentation dynamics and guide feed-rate adjustments. The algorithm introduced batch-specific parameters to account for variability in cell conditions, enabling accurate prediction and control throughout the fermentation process. The optimized feeding strategy significantly reduced glycerol levels, leading to an increased 3-HP yield and improved downstream purification, resulting in decreased processing time and cost. This strategy enhances both the efficiency and scalability of 3-HP production, offering a sustainable and economically viable solution for industrial bioprocessing.