USE OF MODULATING FUNCTIONS FOR REACTION NETWORK IDENTIFICATION

This paper improves a methodology that determines the pseudo-stoichiometric coefficient matrix K in a mass balance based model. The idea consists in using modulating functions in order to filter the data that are often noisy for biotechnological processes. The first stage consists in estimating the number of reactions that must be taken into account to represent the main mass transfer within the bioreactor. This provides the dimension of K. Then a set of modulating functions are used to directly determine the matrix coefficients. This method is illustrated with simulations of a process of lipase production from olive oil by Candida rugosa and compared to the method of Bernard and Bastin (2005a) based on a temporal approach.