Digital twin for CIP-washing station with SCADA system

Due to stricter sanitary and hygienic standards and the constant increase in prices for energy and water resources, the importance of CIP purification processes is growing in terms of cost for breweries. Optimization by traditional methods is difficult, the processes are nonlinear, depend on the composition of the product and require strict compliance with safety conditions, especially when working with glass containers. Consequently, the task for effective CIP systems is formulated considering the above factors.

Aim. To create a digital twin system for CIP station, capable to generate an optimal operating profile in real-time, taking into account factors like flow, temperature, and time, and to ensure direct industrial SCADA system implementation. The solution is based on a mathematical model that incorporates a system of implicit differential equations describing the processes of heat and mass transfer as well as the hydraulic equilibrium of the pipeline network.

Methods. Optimization is performed using an iterative dynamic programming algorithm that minimizes the total costs of steam, electricity, and chemicals, while maintaining hygienic standards for residual contamination.

Results. To ensure compatibility with the industrial SCADA environment, we have employed automatic code generation in the S-function to transfer the model from MATLAB/Simulink, eliminating the need for additional adaptation. A pilot real-world test was carried out for 60000 bottles per hour. During 30 consecutive days, the digital twin demonstrated stability and high accuracy, deviations from the SCADA logs did not exceed 0.4 °C in temperature and 1.2 % in solution consumption.

Conclusions. The introduction of profile-guided optimization cut the CIP cycle time by 10-15 % and reduce the consumption of caustic and steam by an average of 20 %, while maintaining the microbiological purity of the equipment. The results obtained confirm both the technical and economic feasibility of the proposed approach.