Modeling and simulation of the beer fermentation process and temperature control

Beer is the most common alcoholic drink around the world. When talking about the beer quality, flavor of the beer is more concerned. Most of the brewers in Sri Lanka are using traditional methods to brewing beer. Most of brewers using pre identified recipe to produce mass production of beer. Therefore, beer quality i.e. flavor is varying brand by brand. It is important to study the variation of temperature, how will affect to the final alcohol production and the flavor compound formation. Beer manufacturing industry can be used this knowledge to increase the production efficiency and the product quality. It is very important to know about dynamics of forming flavor compounds. In this work the fermentation process is concerned since all the flavor compounds are formed during the fermentation. The mechanistic model is developed by based on the knowledge of biochemical processes in the yeast cell and previously developed mathematical models which are available in the literature. There are many beer models can be found such as Engasser et al.,(1981);Growth kinetic model, Gee at al.,(1988), Phisalaphong et al.(2006);Growth kinetic model and effect on temperature, W.F. Ramirez and J. Maciejowski.,(2007);Optimal beer fermentation, etc. The beer fermentation process is modeled and simulated in MATLAB/Simulink environment. Growth model, nutrient model, and the flavor model are considered and developed. Growth model consists of sugar consumption, biomass growth and ethanol formation models. Those models are developed with temperature dependent parameters to observe the effect of the temperature. Three amino acids which are valine, leucine and iso leucine are considered for the Nutrient model. Consumption of these three amino acids is considered during fermentation. Flavor model is developed based on the growth model and the nutrient model. Flavor compounds are categorized into three groups which are fusel alcohols, esters and vicinal diketones. Altogether nine parameters are considered as flavor compounds and the effects of temperature on those are simulated with MATLAB/Simulink and JModelica. Industrial temperature profile is obtained and applied for the developed model and simulated in MATLAB/Simulink and the results are analyzed. PI controller is applied to get identified temperature profile to obtain flavor formation and the dynamic model is used for find suitable controller parameters for the best control. The controller is simulated in MATLAB/Simulink