Predicting the Fouling Behavior of Whey Protein Concentrate in Polymeric Heat Exchangers

Abstract

Fouling in heat exchangers, particularly in the dairy industry, presents significant operational challenges, increasing energy consumption and maintenance costs. Polymeric heat exchangers, with their favorable fouling mitigation behavior, offer a potential solution to reduce these impacts. A mechanistic and an empirical fouling model were developed to predict the unique detachment mechanism of whey protein concentrate (WPC) fouling layers on polyetheretherketone (PEEK) heat exchanger surfaces caused by boiling beneath the fouling deposits. Model parameters were estimated using experimental data of the total fouling mass. Fouling experiments were carried out for different process conditions. To identify the dependency of the model parameters on the process condition, symbolic regression was applied. Previously unseen experimental data was used to validate the prediction capabilities of the models, which aim to predict fouling mass and, in case of the mechanistic model, thermal resistance. The results demonstrate that the empirical model predicts the fouling mass with an accuracy of ±20% for untrained operating conditions within the boundaries of the training set. Larger deviations (< 70%) were observed for the mechanistic model. When predicting fouling mass outside the training data set, the empirical model fails to do so when extrapolating. While the mechanistic model provides better results compared to the empirical model when extrapolating, an error of < 130% remains. The calculated thermal resistance shows discrepancies, particularly for high WPC concentrations and high heat flux. The findings suggest that PEEK heat exchangers may significantly reduce fouling-related downtime and energy costs in dairy processing.