Less cooling energy in wine fermentation – A case study in mathematical modeling, simulation and optimization

Abstract In this paper, we study model predictive control (MPC) of the cooling process during wine fermentation. A strategy to solve nonlinear control problems with changing model parameters and changes in the states is presented. The parameters and states determining the fermentation dynamics are regularly estimated from measurements and the optimal cooling profile is computed and if necessary adjusted. The process of wine fermentation is described by a modified version of the novel model including a death phase for yeast and the influence of oxygen on the process published in Borzi et al. (2014) . This modification guarantees that nutrients can still be consumed in an anaerobic environment. The numerical results regarding the control inputs and the development of the substrates and the product for an industrial controller and for this MPC controller are compared. It arises that the usage of this MPC cooling strategy results in considerable savings for the energy consumption in the process of wine fermentation.