Neural network software sensors design for lysine fermentation process

Abstract

The main goal of this article is to study the possibility for neural network software sensors design for Lysine fermentation process on the basis of only one on-line measurable process variable, in this case - the dissolved oxygen concentration. Software sensors for biomass, lysine, glucose, and ammonium concentrations were designed. It is shown that past values of the dissolved oxygen must also be included in order to get successful results, or to enhance the quality of the estimation. The applicability of the sensors in the framework of a closed loop control system was also investigated.