Evaluation of machine learning algorithms to predict internal concentration polarization in forward osmosis

Ibrar, Ibrar; Yadav, Sudesh; Braytee, Ali; Altaee, Ali; Hosseinzadeh, Ahmad; Samal, Akshaya K.; Zhou, John L.; Khan, Jamshed Ali; Bartocci, Pietro; Fantozzi, Francesco

doi:10.1016/j.memsci.2022.120257

Cited by 31 publications

(15 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the nonlinear relationship occurs between DS molarity and water flux, particularly at high DS molarity in Figure S5b. It can be observed that the SHAP value of DS molarity rises with the increase of DS molarity and then remains constant, which is due to the effect of ICP in the support layer of the FO membrane. , The feature importance reveals that the DS molecular weight, flow orientation, and DS type could be ignored in the XGBoost model since they have little influence on the concentrative internal concentration polarization (CICP) process …”

Section: Resultsmentioning

confidence: 99%

“…The unimportant input features could decrease the accuracy of XGBoost model due to the introduced noisy data . The SHAP values provide an effective method to select important features, which could enable the ML models to avoid the influence of noisy data and predict water flux with higher accuracy.…”

Section: Resultsmentioning

confidence: 99%

“…Generally, the osmotic pressure difference is the main driving factor in the FO system, and DS CFV has a significant influence on the water flux by affecting the ICP more than FS CFV. 15 Similarly, the FS osmotic pressure is relatively more important than FS molarity on water flux. On this account, the DS osmotic pressure, DS CFV, and FS molarity were dropped from the data set.…”

Section: Methodsmentioning

confidence: 99%

“…That is to say one of them should be removed from the input features. Generally, the osmotic pressure difference is the main driving factor in the FO system, and DS CFV has a significant influence on the water flux by affecting the ICP more than FS CFV . Similarly, the FS osmotic pressure is relatively more important than FS molarity on water flux.…”

Section: Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Modeling and Evaluation of the Permeate Flux in Forward Osmosis Process with Machine Learning

Shi

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Predicting the permeate flux is critical for evaluating and optimizing the performance of the forward osmosis (FO) process. However, the solution diffusion models have poor applicability in accessing the FO process. Recently, the data-driven eXtreme Gradient Boosting (XGBoost) algorithm has been proven to be effective in processing structure data in engineering problems and has not been utilized to assess the FO process. Herein, a combination of the XGBoost model with a genetic algorithm (GA) was first proposed to predict the permeate flux, highlighting its superiority in the FO process through comparison of the support vector regression (SVR) model, the artificial neural network (ANN), and the multiple linear regression (MLR). Moreover, the performance of these models was optimized by tuning hyperparameters with a genetic algorithm (GA) and compared via Taylor Diagram. Among these machine learning (ML) models, the GA-based XGBoost model is superior to the other three models in terms of mean square error (MSE, 2.7326) and coefficient of determination (R 2, 0.9721) on the test data, and its prediction power was compared to that of the solution diffusion (SD) model in the literature. Finally, further insight into the feature importance that affects the permeate flux in the FO process was examined by utilizing the SHapley Additive exPlanations (SHAP) to estimate the contribution value of various variables. The results demonstrated that the XGBoost model could predict the permeate flux in the FO system with high accuracy and good generalization ability for the given data set and even on the unseen data. Furthermore, the findings of the SHAP method show that the osmotic pressure difference, the osmotic pressure difference of draw solution and FS solution, the crossflow velocity of the feed solution and draw solution, and the water permeability coefficient have a significant impact on water flux.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Modeling and Evaluation of the Permeate Flux in Forward Osmosis Process with Machine Learning

Shi

et al. 2022

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

show abstract

“…Water scarcity predictions in the coming decades have led to a surge in research on seawater desalination and wastewater reclamation [ 1 ]. Pressure-driven membrane filtration, such as reverse osmosis (RO) for seawater and wastewater treatment, is the most common method for freshwater supply in arid regions [ 2 , 3 , 4 , 5 , 6 , 7 ].…”

Section: Introductionmentioning

confidence: 99%

Fouling and Performance Investigation of Membrane Distillation at Elevated Recoveries for Seawater Desalination and Wastewater Reclamation

et al. 2022

View full text Add to dashboard Cite

: This study reports on the impact of elevated recovery (i.e., 80%, 85%, and 90%) on the fouling and performance of air gap membrane distillation (AGMD) with real seawater and landfill leachate wastewater samples using polytetrafluoroethylene (PTFE) polymer membranes. Increasing the feed temperature from 55 °C to 65 °C improved the water flux of seawater and wastewater and shortened the operating time by 42.8% for all recoveries. The average water flux in the 80%, 85%, and 90% recovery experiments at the 65 °C feed temperature was 32%, 37.32%, and 36.7% higher than the case of 55 °C for the same recoveries. The water flux decline was more severe at a higher temperature and recovery. The highest flux decline was observed with a 90% recovery at 65 °C feed temperature, followed by an 85% recovery at 65 °C. Close examination of the foulants layer revealed that seawater formed a cake fouling layer made predominantly of metal oxides. In contrast, the landfill leachate fouling was a combination of pore blocking and cake formation, consisting mainly of carbonous and nitrogenous compounds. Physical cleaning with deionized (DI) water at 55 °C and 65 °C and chemical cleaning with hydrogen peroxide (H2O2) were investigated for their efficiency in removing membrane foulants. Analytical results revealed that seawater fouling caused membrane pore blockage while wastewater fouling formed a porous layer on the membrane surface. The results showed that membrane cleaning with hydrogen peroxide restored >97% of the water flux. Interestingly, the fouling factor in seawater tests was 10%, while it was 16% for the wastewater tests.

show abstract

On the evaluating membrane flux of forward osmosis systems: Data assessment and advanced intelligent modeling

Hekmatmehr,

Esmaeili,

Atashrouz

et al. 2024

Water Environment Research

View full text Add to dashboard Cite

As an emerging desalination technology, forward osmosis (FO) can potentially become a reliable method to help remedy the current water crisis. Introducing uncomplicated and precise models could help FO systems' optimization. This paper presents the prediction and evaluation of FO systems' membrane flux using various artificial intelligence‐based models. Detailed data gathering and cleaning were emphasized because appropriate modeling requires precise inputs. Accumulating data from the original sources, followed by duplicate removal, outlier detection, and feature selection, paved the way to begin modeling. Six models were executed for the prediction task, among which two are tree‐based models, two are deep learning models, and two are miscellaneous models. The calculated coefficient of determination (R2) of our best model (XGBoost) was 0.992. In conclusion, tree‐based models (XGBoost and CatBoost) show more accurate performance than neural networks. Furthermore, in the sensitivity analysis, feed solution (FS) and draw solution (DS) concentrations showed a strong correlation with membrane flux.Practitioner Points The FO membrane flux was predicted using a variety of machine‐learning models. Thorough data preprocessing was executed. The XGBoost model showed the best performance, with an R2 of 0.992. Tree‐based models outperformed neural networks and other models.

show abstract

Evaluation of machine learning algorithms to predict internal concentration polarization in forward osmosis

Cited by 31 publications

References 44 publications

Modeling and Evaluation of the Permeate Flux in Forward Osmosis Process with Machine Learning

Modeling and Evaluation of the Permeate Flux in Forward Osmosis Process with Machine Learning

Fouling and Performance Investigation of Membrane Distillation at Elevated Recoveries for Seawater Desalination and Wastewater Reclamation

On the evaluating membrane flux of forward osmosis systems: Data assessment and advanced intelligent modeling

Contact Info

Product

Resources

About