Implementation of Stir-Speed Adopted Controllers Onto a Batch Bioreactor for Improved Fermentation

Abstract

Quality, quantity, and economy of the fermentation product depend on the transient response and the steady-state of the fermentation process. Due to its simple construction, an industrial batch bioreactor cannot be equipped with a closed-loop control system that applies adding or removing substances during fermentation. This theoretical and practical study displays how fermentation can be controlled, in this case by altering the rotational speed of a stirrer system of such a bioreactor. A numerical analysis was employed to set up a mathematical model that describes the impact of the stirrer speed on the fermentation. The model consists of linear and non-linear parts. To determine their parameters, a least square identification and a particle swarm optimization method were utilized. The model was fitted and tested experimentally in a laboratory on the operating batch bioreactor. The design and synthesis of a conventional linear control system and an advanced model reference adaptive control system, which is also convenient for control of non-linear controlled plants, were carried out based on the determined mathematical model. Both control concepts were first implemented into the dSpace 1103 control system, convenient for faster development and prototyping. Finally, they were realized with the Siemens SIMATIC S7 programmable logic controller, which was ideal for industrial environment operation. During the operation of the bioreactor, the control systems were tested along with the acquisition of measured data involving CO2 concentration of a fermentation product. The findings of successfully applying linear and adaptive controllers of the milk fermentation in the batch bioreactor are analyzed and presented.

Adaptive Control of CO2 Production during Milk Fermentation in a Batch Bioreactor

“…The new bioreactors enable an easy, fast and wide range of changes in the mechanical (by mixing) and thermal (by heating and cooling) conditions of fermentation processes, also during operation. It turns out that the course of the fermentation process can be influenced by changes in these fermentation quantities (stirrer speed and temperature), and it is not always necessary to control the fermentation process by feeding substances into the bioreactor [7]. To develop the control system for the fermentation process, which would use stirrer speed or heater temperature as an input quantity, we need a mathematical model that describes the influence of these quantities on the fermentation process.…”

“…During the performing the experiments, the sampling time was changed and adjusted to the dynamics of the measured signal. The measured data were saved into Microsoft Office 365 Excel documents, transferred into MathWorks MATLAB, and processed using MATLAB with its Optimization toolbox functions [7].…”

“…For the implementation of the control system, a dSpace 1103 PPC controller board was utilized. The controller is equipped with 16-bit A/D and D/A converters as well as serial and CAN interfaces [7]. An analogue output module of the basic device SevenMulti was used for the transfer of the measured signal of the CO 2 concentration from the bioreactor system to the controller's analogue input.…”

“…Fermentation is described as a process in which an agent causes an organic substance to break down into simpler substances. The agents are mainly microorganisms, the source substance is named a substrate, and the final substance is named a product [7].…”

“…The profiles are most easily influenced by the bioreactor's heating/cooling system and the stirrer system. In [6,30], an analysis of the influence of temperature change on the course of the fermentation process is made, and the influence of stirrer speed change is discussed in [6,7,31]. The fundamental mathematic model does not enable the evaluation of the impact of temperature changes on the time courses of concentrations of individual substances of the fermentation process.…”