Abstract
Membrane fouling, which causes water flux decline over time, is the greatest challenge in membrane filtration. The inherent complexity of fouling phenomena, combined with the heterogeneous and fluctuating characteristics of dissolved organic matter, makes predicting fouling a difficult task. This work proposes, develops, and validates a novel approach for predicting fouling and flux decline under varying organic loads. A semi-empirical mechanistic model (Hermia) is enhanced with machine learning to address the complexity of fouling phenomena. The machine learning inputs include initial flux, organic load, and excitation-emission fluorescence spectra. The model was trained using different initial fluxes, organic carbon loads, and compositions. Validation on new experiments (i.e., experiments not used during training) demonstrated good predictive accuracy, with R2 values ranging from 0.87 to 0.99 and an average of 0.95. The proposed approach is expected to have significant potential in the field of membrane processes and water technologies in general.
Supplementary materials
Title
Flux decline prediction in dead-end ultrafiltration combining fluorescence spectroscopy and mechanism-informed machine learning_SI
Description
The Supporting Information contains the following additional materials:
SI 1: Feed solution characteristics. SI 2: Data about SVM training phase. SI 3: Data about SVM validation phase. SI 4: Interpretability analysis. SI 5: Predictors selection, model structure and machine learning algorithms. SI 6: Sensitivity Analysis.
Actions
Supplementary weblinks
Title
Matlab Code
Description
Matlab implementation of the Hermia-informed SVM model to predict flux decline (available soon after publication)
Actions
View