Evolutionary Algorithms for Optimal Control in Fed-Batch Fermentation Processes

In this work, Evolutionary Algorithms (EAs) are used to achieve optimal feedforward control in a recombinant bacterial fed-batch fermentation process, that aims at producing a bio-pharmaceutical product. Three different aspects are the target of the optimization procedure: the feeding trajectory (the amount of substrate introduced in a bioreactor per time unit), the duration of the fermentation and the initial conditions of the process. A novel EA with variable size chromosomes and using real-valued representations is proposed that is capable of simultaneously optimizing the aforementioned aspects. Outstanding productivity levels were achieved and the results are validated by practice.