NEURAL NETWORK BASED MODELING AND OPTIMIZATION OF FED-BATCH BIOREACTORS

摘要

Dynamic optimization and optimal control of fed-batch bioreactors depends on the availability of a good model of the process. In this paper, we propose a neural network based modeling scheme that captures the inherently non-linear dynamics of fed-batch bioreactor systems. This neural network based model is then used in conjunction with dynamic programming optimization to generate optimal control policies. The results of applying this method to two fed-batch bioreactor systems are presented and the results obtained are shown to be in good agreement with prior work.