Application of feed-forward neural networks for system identification of a biochemical process

The feasibility of using feedforward neural networks for system identification of a process with highly nonlinear characteristics was studied. A biochemical process was chosen where the microorganism Saccharomyces cerevisiae, a yeast, grows in a chemostat on glucose substrate and produces ethanol as a product of primary energy metabolism. The three state variables considered for the process are microbial concentration, substrate concentration, and product concentration. The Levenberg-Marquardt method was used to train the neural networks by minimizing the sum of squares of the residuals. The inputs to the networks were the three state variables at a given time and the process input variables from that time to the time for which the state variables are to be predicted. The output of each node was calculated by the logistic (sigmoid) or symmetric logarithmoid activation functions on the weighted sum of inputs to that node. In most cases, the symmetric Iogarithmoid resulted in lower error square sum values than the sigmoid