MODEL-BASED PREDICTIVE CONTROL SYSTEM DEVELOPMENT FOR THE BEER PRODUCTION PROCESS

Abstract

This study presents the design and evaluation of a model-based predictive
control (MPC) system for precise temperature regulation during the mashing
stage of beer production. The mashing process requires a strict temperature
profile to ensure optimal enzymatic activity, maximize sugar yield, and maintain
consistent product quality. A dynamic mathematical model of the mash tun,
derived from energy balance equations that account for heat input and thermal
losses, was employed as the predictive core of the controller. The MPC algorithm
forecasts future temperature trajectories over a finite horizon and determines
optimal heating inputs while enforcing operational constraints on temperature
and energy usage. Numerical simulations were carried out in Python with the
NumPy, Matplotlib, and Control libraries, enabling accurate process modeling,
optimization, and visualization of control performance. Results show that MPC
achieves a maximum temperature deviation of ±0.2°C, reduces total heating energy
consumption by approximately 15%, and demonstrates significantly faster recovery
from disturbances. These findings demonstrate that MPC offers a robust and
energy-efficient solution for industrial mashing control, with potential benefits for
improving beer quality and reducing production costs.