Fig. 6.

Control schemes of the regulatory layers used in fed-batch fermentations 3 and 4. Firstly, a partial least squares model (PLS) is used to calculate the process variable (PV) of the glucose concentration. Secondly, the PV is manipulated depending on the pre-defined set-point (SP) and the upper and lower boundaries. The integral term ($\mathit{Ki}$) of the proportional, integral (PI) controller is also manipulated depending on the PV, while the proportional term ($\mathit{Kp}$) is kept constant. The manipulated variable (MV), resulting from the PI controller, is sent to the actuator (controlled pump). The control schemes for the fed-batch fermentations 1 and 2 are shown in Additional file 1