APPLICATION OF NEURAL NETWORKS TO LYSINE PRODUCTION

Lysine is an essential amino acid in human nutrition and also widely u sed in animal feed formulations. It is produced on a large scale by fe rmentation in stirred tank bioreactors. In the present work lysine was produced by fed-batch fermentation with an industrial Brevibacterium flavum strain grown in a 115 m(3) fermenter on a beet molasses based m edium. The difficulties in on-line monitoring of substrate consumption and of product formation complicate real-time process control. We dem onstrate that well trained backpropagation multilayer neural networks can be employed to overcome such problems without detailed prior knowl edge of the relationships of process variables under investigation. Ne ural network models programmed in MS-Visual C++ for Windows and implem ented on a personal computer were constructed and applied to state est imation and multi-step-ahead prediction of consumed sugar and produced lysine on the basis of on-line measurable variables for process contr ol purposes.