Event software sensor and adaptive extremum seeking alternatives for optimizing a class of fed-batch bioreactors

Optimization and control in spite of plant uncertainties is always a challenge, especially if additional constraints about measurements are present. Here, two different approaches to solve this problem are compared using simulation. They aim at reducing the operation time of bioreactors with inhibitory behavior where measuring the reaction rate is not feasible. The "Adaptive Extremum Seeking" proposed version relies on the structure information of the kinetic model and requires the measurements of the substrate and one other related variable. The "Event Driven Time Optimal Controller" strategy avoids the substrate measurement and, using an event software sensor, provides a nearly optimal solution without requiring a complete model.