Semi Optimal Feed Profiles for Inorganic Phosphate Addition in Tetracycline Fermentation

Inorganic phosphates are added to antibiotic fermentation to keep pH within a favourable range. However, phosphate itself has very interesting influence upon the fermentation. Very low dose of phosphate is detrimental to cell growth. Low cell concentration shall result in low volumetric productivity in fermentation. On the other hand, phosphate is inhibitory to tetracycline formation. Phosphate concentration should come down to very low levels before tetracycline production begins. Therefore, this pose a nice optimization problem. The purpose of the present paper is to find semi-optimal phosphate addition profile by a hybrid chemotaxis bubblesort (H.C.B.) algorithm.

We simulated the fermentation by a modified model from the model developed by Votruba by differentiating the model O.D.E s by a Runge-Kutta method and developed the relevant profit function. Then optimizations of feed profiles were done by an improved chemotaxis algorithm. Chemotaxis algorithms are frequently used to train artificial neural nets, it has been modified for current bioprocess optimization task.