Process modelling of batch enzymatic starch hydrolysis for real-time monitoring applications

Enzymatic starch hydrolysis is a key biotechnological process whose efficient operation requires accurate process modelling, monitoring, and control. This study presents a dynamic mathematical model for batch enzymatic starch hydrolysis based on Michaelis–Menten kinetics and demonstrates its applicability for process analysis and future integration into advanced monitoring and control frameworks. The model describes the interaction between starch substrate, enzyme, enzyme–substrate complex, and glucose product through a system of ordinary differential equations, incorporating enzyme deactivation effects. Numerical integration was performed using an adaptive fourth-order Runge–Kutta algorithm implemented in MATLAB. Model parameters were identified through laboratory-scale hydrolysis experiments, where glucose concentration profiles were monitored by high-performance liquid chromatography. Simulations provided predictions of glucose formation, the results revealing characteristic saturation behaviour of glucose production and highlighting the influence of enzyme deactivation on process duration and final conversion. Model validation against experimental data demonstrated good agreement during the initial and intermediate stages of hydrolysis, while deviations observed near the saturation region suggested a stronger real enzyme deactivation than predicted by the model. From a process control perspective, the developed model provides a valuable foundation for batch duration optimization and conversion prediction. Furthermore, the model structure is suitable for integration with modern machine learning approaches, including hybrid physics-informed models, soft sensors, and predictive control algorithms. Such integration could enhance real-time monitoring and decision support in industrial bioprocesses, contributing to improved productivity and operational efficiency.