Recovery of Whey Protein by Using Microfiltration: Artificial Neural Network–Based Modeling and Effects of Different Operating Parameters

Abstract

Microfiltration is one of the most suitable processes for protein recovery from whey due to its low energy consumption and lack of use of heat and chemicals. However, membrane fouling is one of the limiting factors in the microfiltration process, preventing its commercial use. In this study, an artificial neural network (ANN) based model was employed to study the effects of different operating parameters on membrane fouling in whey concentration. Trans-membrane pressure, Reynolds number, and feed temperature were selected as the input parameters. Experimental data from the available studies were used to train the ANN. The ANN with 23 neurons gave a minimum mean squared error (MSE) for trans-membrane pressure and Reynolds number. The ANN with seven neurons gave the minimum MSE for feed temperature. The predicted values from both ANNs well fitted with the experimental results with R² < 0.99. Simulations showed that membrane fouling increased as flux reduction increased from 36.3% to 76.39% when trans-membrane pressure increased from 0.5 to 2 bar. In contrast, a 19.96% reduction in flux was observed by increasing the Reynolds number from 750 to 2500. An increment of 77.37% of flux reduction was observed with increasing feed temperature from 30°C to 40°C. Simulations confirmed that transmembrane pressure, Reynolds number, and feed temperature strongly influence membrane fouling. An ANN-based approach was the most accurate method to model membrane fouling for whey protein separation.