Exploratory Analysis of Biochemical Processes Using Hybrid Modeling Methods

A heuristic process data based procedure has been developed that allows a universal time-efficient exploratory analysis of biochemical processes and predictive information extraction from large data sets. It uses artificial neural networks in combination with mass balance equations to represent unknown process relationships between process variables. An efficient algorithm for training of this hybrid system has been developed. The result of the procedure is a numerical representation of the important process relationships that immediately allows to determine improved set points and/or profiles for the manipulated variables with respect to process performance. For illustration, the procedure is applied for extraction of the important relationships in fed-batch bacterial cultivation process.