Model based predictive control for bioprocesses, using a feedforward neural network

摘要

This paper deals with a neural network based GPC structure for a bioprocess control. Comparing to IMC structure, this method offers two advantages: the neural inverting operation of the process model is eliminated and there are various possibilities to adjust the control law properties. The GPC method is applied to a biomass production process and to a lipase production process. In both cases many simulation results are presented which illustrate the validity of the method.