Modeling and control by artificial neural networks in biotechnology

Abstract

The use of artificial neural networks (ANN) for modeling and control in biotechnology is presented. In this work are reviewed models of general classes of ANN structures. Modeling of baker's yeast production is studied by use of classical feedforward structure with multilayer of static neurons. For adaptation of parameters is applied the algorithm of Ribiera-Polak with Powell modification. Learning sets of iput-output patterns are obtained by a computer model and by data from an industrial plant. Proposed is a modular Structure of ANN MISO systems for on-line estimation of pH, ethanol partial pressure, biomass concentration and the cell (metabolic) respiratory quotient (RQ) based on observed RQ in gas phase. Discussed is application of ANN modules for integration into a control structure.