A hybrid neural network algorithm for on-line state inference that accounts for differences in inoculum of Cephalosporium acremonium in fed-batch fermentors

One serious difficulty in modeling a fermentative process is the forecastin g of the duration of the lag phase. The usual approach to model biochemical reactors relies on first-principles, unstructured mathematical models. The se models are not able to take into account changes in the process response caused by different incubation times or by repeated fedbatches. To overcom e this problem, we have proposed a hybrid neural network algorithm. Feedfor ward neural networks were used to estimate rates of cell growth, substrate consumption, and product formation from on-line measurements during cephalo sporin C production. These rates were included in the mass balance equation s to estimate key process variables: concentrations of cells, substrate, an d product. Data from fed-batch fermentation runs in a stirred aerated biore actor employing the microorganism Cephalosporium acremonium ATCC 48272 were used. On-line measurements strongly related to the mass and activity of th e cells used. They include carbon dioxide and oxygen concentrations in the exhausted gas. Good results were obtained using this approach.