Artificial intelligence-incorporated membrane fouling prediction for membrane-based processes in the past 20 years: A critical review

Influence of graphene oxide additives on the NF separation of triazine-based H2S scavenging compounds using advanced membrane technology

This work proposes an innovative approach for the membrane separation of spent and unspent H2S scavengers (SUS) derived from the application of MEA-triazine in offshore oil and gas production. Modified nanofiltration membranes were fabricated by incorporating graphene oxide (GO) and polyvinyl alcohol (PVA) into a thin film composite (TFC) to obtain a thin film nanocomposite (TFN) with enhanced permeability. In addition, various immobilization strategies for GO were investigated. The performance of the membranes and the effect of the GO loading were evaluated in terms of permeability, fouling propensity, and rejection of key components of the SUS, i.e., MEA-triazine (unspent scavenger), dithiazine (spent scavenger), and monoethanolamine, operating on a sample of SUS wastewater obtained from an offshore oil and gas platform. Various characterization techniques, such as contact angle, FTIR, XRD, SEM, TGA, and AFM, were employed to evaluate the structure, composition, and hydrophilicity of the membrane. The results show a remarkable increase in permeability (from 0.22 Lm-2 h-1 bar-1 for the TFC to 5.8 Lm-2 h-1 bar-1 for the TFN membranes), due to the enhanced hydrophilicity from GO incorporation. The strong interfacial interaction between GO and PVA within the TFN membrane results in negligible nanofiller leaching. The incorporation of GO moderately increases the rejection of the unspent scavenger (63%-73%, 62%-79%, 62%-80%, and 68%-76%), while drastically increasing the rejection of the spent scavenger, which is approximately null for the TFC membrane without GO and increases up to 58% in the TFN membrane with GO. Therefore, while the proposed membranes cannot be used for the selective separation of the unspent form the spent scavenger, they can achieve substantial recovery of all the key components contained in the SUS to avoid their discharge into the sea.

Unveiling the spatiotemporal dynamics of membrane fouling: A focused review on dynamic fouling characterization techniques and future perspectives

Membrane technology has emerged as a crucial method for obtaining clean water from unconventional sources in the face of water scarcity. It finds wide applications in wastewater treatment, advanced treatment, and desalination of seawater and brackish water. However, membrane fouling poses a huge challenge that limits the development of membrane-based water treatment technologies. Characterizing the dynamics of membrane fouling is crucial for understanding its development, mechanisms, and effective mitigation. Instrumental techniques that enable in situ or real-time characterization of the dynamics of membrane fouling provide insights into the temporal and spatial evolution of fouling, which play a crucial role in understanding the fouling mechanism and the formulation of membrane control strategies. This review consolidates existing knowledge about the principal advanced instrumental analysis technologies employed to characterize the dynamics of membrane fouling, in terms of membrane structure, morphology, and intermolecular forces. Working principles, applications, and limitations of each technique are discussed, enabling researchers to select appropriate methods for their specific studies. Furthermore, prospects for the future development of dynamic characterization techniques for membrane fouling are discussed, underscoring the need for continued research and innovation in this field to overcome the challenges posed by membrane fouling.

Modelling and optimization of membrane process for removal of biologics (pathogens) from water and wastewater: Current perspectives and challenges

As one of the 17 sustainable development goals, the United Nations (UN) has prioritized "clean water and sanitation" (Goal 6) to reduce the discharge of emerging pollutants and disease-causing agents into the environment. Contamination of water by pathogenic microorganisms and their existence in treated water is a global public health concern. Under natural conditions, water is frequently prone to contamination by invasive microorganisms, such as bacteria, viruses, and protozoa. This circumstance has therefore highlighted the critical need for research techniques to prevent, treat, and get rid of pathogens in wastewater. Membrane systems have emerged as one of the effective ways of removing contaminants from water and wastewater However, few research studies have examined the synergistic or conflicting effects of operating conditions on newly developing contaminants found in wastewater. Therefore, the efficient, dependable, and expeditious examination of the pathogens in the intricate wastewater matrix remains a significant obstacle. As far as it can be ascertained, much attention has not recently been given to optimizing membrane processes to develop optimal operation design as related to pathogen removal from water and wastewater. Therefore, this state-of-the-art review aims to discuss the current trends in removing pathogens from wastewater by membrane techniques. In addition, conventional techniques of treating pathogenic-containing water and wastewater and their shortcomings were briefly discussed. Furthermore, derived mathematical models suitable for modelling, simulation, and control of membrane technologies for pathogens removal are highlighted. In conclusion, the challenges facing membrane technologies for removing pathogens were extensively discussed, and future outlooks/perspectives on optimizing and modelling membrane processes are recommended.

Understanding Rejection Mechanisms of Trace Organic Contaminants by Polyamide Membranes via Data-Knowledge Codriven Machine Learning

Data-driven machine learning (ML) provides a promising approach to understanding and predicting the rejection of trace organic contaminants (TrOCs) by polyamide (PA). However, various confounding variables, coupled with data scarcity, restrict the direct application of data-driven ML. In this study, we developed a data-knowledge codriven ML model via domain-knowledge embedding and explored its application in comprehending TrOC rejection by PA membranes. Domain-knowledge embedding enhanced both the predictive performance and the interpretability of the ML model. The contribution of key mechanisms, including size exclusion, charge effect, hydrophobic interaction, etc., that dominate the rejections of the three TrOC categories (neutral hydrophilic, neutral hydrophobic, and charged TrOCs) was quantified. Log D and molecular charge emerge as key factors contributing to the discernible variations in the rejection among the three TrOC categories. Furthermore, we quantitatively compared the TrOC rejection mechanisms between nanofiltration (NF) and reverse osmosis (RO) PA membranes. The charge effect and hydrophobic interactions possessed higher weights for NF to reject TrOCs, while the size exclusion in RO played a more important role. This study demonstrated the effectiveness of the data-knowledge codriven ML method in understanding TrOC rejection by PA membranes, providing a methodology to formulate a strategy for targeted TrOC removal.

Easily scale 3D conductive gradient fiber membrane for contaminants removal and fouling mitigation under electrochemical assistance

An electrochemical membrane filtration system provides an innovative approach to enhance contaminant removal and mitigate membrane fouling. There is an urgent need to develop portable, versatile, and efficient electrochemical membranes for affordable wastewater treatment. Here, a 3D conductive gradient fiber membrane (CC/PVDF) with a gradient porous structure was prepared using a two-step phase inversion method. Methyl orange (MO) was utilized as model organic substance to investigate the electrochemical performance of the CC/PVDF membrane. At applied potentials of +2 V, +3 V, -2 V and -3 V, the removal efficiency of MO was 5.1, 5.3, 4.8, and 5.1 times higher than at 0 V. A dramatic flux loss of 35.02% occurred on the membrane without electrochemistry, interestingly, whereas the flux losses were only 23.59%-10.24% in the applied potential after 30 min of filtration, which were approximately 1.18, 1.28, 1.29 and 1.38 times as high as that without electrochemistry, respectively. The enhanced removal and anti-fouling performances of the membranes were attributed to the functions of electrochemical degradation, electrostatic repulsion, and electrically enhanced wettability. Electrochemical generation of Hydrogen peroxide, along with HO• radicals, was detected and direct electron transfer and HO• were proved to be the dominant oxidants responsible for MO degradation. The intermediate oxidation products were identified by mass spectrometry, and an electrochemical degradation pathway of MO was proposed based on bond-breaking oxidation, ring-opening reactions, and complete oxidation. All the findings emphasize that the ECMF system possesses superior efficiency and creative potential for water purification applications.

Desalting Plasma Protein Solutions by Membrane Capacitive Deionization

Plasma protein therapies are used by millions of people across the globe to treat a litany of diseases and serious medical conditions. One challenge in the manufacture of plasma protein therapies is the removal of salt ions (e.g., sodium, phosphate, and chloride) from the protein solution. The conventional approach to remove salt ions is the use of diafiltration membranes (e.g., tangential flow filtration) and ion-exchange chromatography. However, the ion-exchange resins within the chromatographic column as well as filtration membranes are subject to fouling by the plasma protein. In this work, we investigate the membrane capacitive deionization (MCDI) as an alternative separation platform for removing ions from plasma protein solutions with negligible protein loss. MCDI has been previously deployed for brackish water desalination, nutrient recovery, mineral recovery, and removal of pollutants from water. However, this is the first time this technique has been applied for removing 28% of ions (sodium, chloride, and phosphate) from human serum albumin solutions with less than 3% protein loss from the process stream. Furthermore, the MCDI experiments utilized highly conductive poly(phenylene alkylene)-based ion exchange membranes (IEMs). These IEMs combined with ionomer-coated nylon meshes in the spacer channel ameliorate Ohmic resistances in MCDI improving the energy efficiency. Overall, we envision MCDI as an effective separation platform in biopharmaceutical manufacturing for deionizing plasma protein solutions and other pharmaceutical formulations without a loss of active pharmaceutical ingredients.

Machine learning facilitated the conceptual design of an alum dosing system for phosphorus removal in a wastewater treatment plant

Wastewater treatment plants (WWTPs) face challenges in controlling total phosphorus (TP), given more stringent regulations on TP discharging. In particular, WWTPs that operate at a small scale lack resources for real-time monitoring of effluent quality. This study aimed to develop a conceptual alum dosing system for reducing TP concentration, leveraging machine learning (ML) techniques and data from a full-scale WWTP containing incomplete TP information. The proposed system comprises two ML models in series: an Alert model based on LightGBM with an accuracy of 0.92, and a Dosage model employing a voting algorithm through combining three ML algorithms (LightGBM, SGD, and SVC) with an accuracy of 0.76. The proposed system has demonstrated the potential to ensure that 88.1% of the effluent remains below the TP discharge limit, which outperforms traditional dosing methods and could reduce overdosing from 61.3 to 12.1%. Furthermore, the SHapley Additive exPlanations (SHAP) analysis revealed that incorporating the output features from the previous cycle and utilizing the results of the Alert model as the input features for dosage prediction could be an effective method for data with limited information. The findings of this study have practical applications in improving the efficiency and effectiveness of TP control in small-scale WWTPs, providing a valuable solution for complying with stringent regulations and enhancing environmental sustainability.

Predictive modeling based on artificial neural networks for membrane fouling in a large pilot-scale anaerobic membrane bioreactor for treating real municipal wastewater

Membrane fouling is the primary obstacle to applying anaerobic membrane bioreactors (AnMBRs) in municipal wastewater treatment. This issue holds critical significance as efficient wastewater treatment serves as a cornerstone for achieving environmental sustainability. This study uses machine learning to predict membrane fouling, taking advantage of rapid computational and algorithmic advances. Based on the 525-day operation data of a large pilot-scale AnMBR for treating real municipal wastewater, regression prediction was realized using multilayer perceptron (MLP) and long short-term memory (LSTM) artificial neural networks under substantial variations in operating conditions. The models involved employing the organic loading rate, suspended solids concentration, protein concentration in extracellular polymeric substance (EPSp), polysaccharide concentration in EPS (EPSc), reactor temperature, and flux as input features, and transmembrane pressure as the prediction target output. Hyperparameter optimization enhanced the regression prediction accuracies, and the rationality and utility of the MLP model for predicting large-scale AnMBR membrane fouling were confirmed at global and local levels of interpretability analysis. This work not only provides a methodological advance but also underscores the importance of merging environmental engineering with computational advancements to address pressing environmental challenges.

The intelligent prediction of membrane fouling during membrane filtration by mathematical models and artificial intelligence models

Recently, membrane separation technology has been widely utilized in filtration process intensification due to its efficient performance and unique advantages, but membrane fouling limits its development and application. Therefore, the research on membrane fouling prediction and control technology is crucial to effectively reduce membrane fouling and improve separation performance. This review first introduces the main factors (operating condition, material characteristics, and membrane structure properties) and the corresponding principles that affect membrane fouling. In addition, mathematical models (Hermia model and Tandem resistance model), artificial intelligence (AI) models (Artificial neural networks model and fuzzy control model), and AI optimization methods (genetic algorithm and particle swarm algorithm), which are widely used for the prediction of membrane fouling, are summarized and analyzed for comparison. The AI models are usually significantly better than the mathematical models in terms of prediction accuracy and applicability of membrane fouling and can monitor membrane fouling in real-time by working in concert with image processing technology, which is crucial for membrane fouling prediction and mechanism studies. Meanwhile, AI models for membrane fouling prediction in the separation process have shown good potential and are expected to be further applied in large-scale industrial applications for separation and filtration process intensification. This review will help researchers understand the challenges and future research directions in membrane fouling prediction, which is expected to provide an effective method to reduce or even solve the bottleneck problem of membrane fouling, and to promote the further application of AI modeling in environmental and food fields.

Role of inorganic foulants in the aging and deterioration of low-pressure membranes during the chemical cleaning process in surface water treatment: A review

Low-pressure membrane (LPM) filtration, including microfiltration (MF) and ultrafiltration (UF), is a promising technology for the treatment of surface water for drinking and other purposes. Various configurations and operational sequences have been developed to ensure the sustainable provision of clean water by overcoming fouling problems. In the literature, various periodic physical and/or chemical approaches to the cleaning of LPMs have been reported, but little data is available on the aging of MF/UF membranes that results from the interaction between the foulants and the cleaning agent. Periodic physical cleaning of the membrane is expected to return the membrane to its original performance capacity, but it only recovers to a certain level because the remaining foulants cause irreversible fouling. Chemical cleaning can then be employed to recover the membrane from this irreversible fouling but, in the process, it can cause irrecoverable damage to the membrane. In this review, the foulants responsible for irrecoverable damage to MF/UF membranes are summarized, and their interaction with cleaning agents and other foulants is described. The impact of these foulants on various membrane parameters, including filtration efficiency, flux decline, permeability, membrane characterization, and membrane integrity are also summarized and discussed in detail. In addition, mitigation options and future prospects are also discussed with regard to increasing the operational life span of a membrane in a cost-effective manner. Ultimately, this review suggests an advanced control system based on membrane-foulant interactions under the impact of various operational parameters to mitigate the integrity loss of membranes.

Application of Machine Learning to Characterize the Permeate Quality in Pilot-Scale Vacuum-Assisted Air Gap Membrane Distillation Operation

Membrane distillation (MD) is a thermal desalination technique proposed for the valorization of residual brines that other operations such as reverse osmosis cannot treat. Previous studies have shown that vacuum-assisted air gap (V-AGMD) operation in commercial multi-envelope modules improves the performance of MD noticeably. However, the permeate quality at pilot scale has not been thoroughly characterized so far. The aim of this study is, therefore, to assess and model the effect of the main operating conditions (feed flow rate, inlet temperatures, and feed salinity) on the permeate quality. Results from different steady-state experiments allowed to estimate descriptive metrics such as the salt rejection factor (SRF) and the membrane leak ratio (MLR). Given their non-linear behavior, these metrics were subsequently modeled using artificial neural networks (ANN) to estimate the permeate quality in the whole scope of operating conditions. Acceptable SRF results with MLR values lower than 0.2% confirmed the validity of MD as an operation for the treatment of concentrated brines, although the salinity of the resulting permeate does not comply in all cases with that permitted for human consumption.

Thermodynamics and kinetic analysis of membrane: Challenges and perspectives

The ultimate structure of the membrane is determined using two important effects: (i) thermodynamic effect and (ii) kinetic effect. Controlling the mechanism of kinetic and thermodynamic processes in phase separation is essential for enhancing membrane performance. However, the relationship between system parameters and the ultimate membrane morphology is still largely empirical. This review focuses on the fundamental ideas behind thermally induced phase separation (TIPS) and nonsolvent-induced phase separation (NIPS) methods, including both kinetic and thermodynamic elements. The thermodynamic approach to understanding phase separation and the effect of different interaction parameters on membrane morphology has been discussed in detail. Furthermore, this review explores the capabilities and limitations of different macroscopic transport models used for the last four decades to explore the phase inversion process. The application of molecular simulations and phase field to understand phase separation has also been briefly examined. Finally, it discusses the thermodynamic approach to understanding phase separation and the consequence of different interaction parameters on membrane morphology, as well as possible directions for artificial intelligence to fill the gaps in the literature. This review aims to provide comprehensive knowledge and motivation for future modeling work for membrane fabrication via new techniques such as nonsolvent-TIPS, complex-TIPS, non-solvent assisted TIPS, combined NIPS-TIPS method, and mixed solvent phase separation.

Biological processes modelling for MBR systems: A review of the state-of-the-art focusing on SMP and EPS

A mathematical correlation between biomass kinetic and membrane fouling can improve the understanding and spread of Membrane Bioreactor (MBR) technology, especially in solving the membrane fouling issues. On this behalf, this paper, produced by the International Water Association (IWA) Task Group on Membrane modelling and control, reviews the current state-of-the-art regarding the modelling of kinetic processes of biomass, focusing on modelling production and utilization of soluble microbial products (SMP) and extracellular polymeric substances (EPS). The key findings of this work show that the new conceptual approaches focus on the role of different bacterial groups in the formation and degradation of SMP/EPS. Even though several studies have been published regarding SMP modelling, there still needs to be more information due to the highly complicated SMP nature to facilitate the accurate modelling of membrane fouling. The EPS group has seldom been addressed in the literature, probably due to the knowledge deficiency concerning the triggers for production and degradation pathways in MBR systems, which require further efforts. Finally, the successful model applications showed that proper estimation of SMP and EPS by modelling approaches could optimise membrane fouling, which can influence the MBR energy consumption, operating costs, and greenhouse gas emissions.

Predicting rejection of emerging contaminants through RO membrane filtration based on ANN-QSAR modeling approach: trends in molecular descriptors and structures towards rejections

In this work, a quantitative structure-activity relationship (QSAR) study was performed on a set of emerging contaminants (ECs) to predict their rejections by reverse osmosis membrane (RO). A wide range of molecular descriptors was calculated by Dragon software for 72 ECs. The QSAR data was analyzed by an artificial neural network method (ANN), in which four out of 3000 theoretical molecular descriptors were chosen and their significance was computed based on the Garson method. The significance trends of descriptors were as follows in descending order: ESpm14u > R2e > SIC1 > EEig03d. The selected descriptors were ranked based on their importance and then an explorative study was conducted on the QSAR data to show the trends in molecular descriptors and structures toward the rejections values of ECs. The MLR algorithm was used to make a linear model and the results were compared with those of the nonlinear ANN algorithm. The comparison results revealed it is necessary to apply the ANN model to this data with non-linear properties. For the whole dataset, the correlation coefficient (R2) and residual mean squared error (RMSE) of the ANN and MLR methods were 0.9528, 6.4224; and 0.8753, 11.3400, respectively. The comparison results showed the superiority of ANN modeling in the analysis of ECs' QSAR data.

A Review on Membrane Fouling Prediction Using Artificial Neural Networks (ANNs)

Membrane fouling is a major hurdle to effective pressure-driven membrane processes, such as microfiltration (MF), ultrafiltration (UF), nanofiltration (NF), and reverse osmosis (RO). Fouling refers to the accumulation of particles, organic and inorganic matter, and microbial cells on the membrane's external and internal surface, which reduces the permeate flux and increases the needed transmembrane pressure. Various factors affect membrane fouling, including feed water quality, membrane characteristics, operating conditions, and cleaning protocols. Several models have been developed to predict membrane fouling in pressure-driven processes. These models can be divided into traditional empirical, mechanistic, and artificial intelligence (AI)-based models. Artificial neural networks (ANNs) are powerful tools for nonlinear mapping and prediction, and they can capture complex relationships between input and output variables. In membrane fouling prediction, ANNs can be trained using historical data to predict the fouling rate or other fouling-related parameters based on the process parameters. This review addresses the pertinent literature about using ANNs for membrane fouling prediction. Specifically, complementing other existing reviews that focus on mathematical models or broad AI-based simulations, the present review focuses on the use of AI-based fouling prediction models, namely, artificial neural networks (ANNs) and their derivatives, to provide deeper insights into the strengths, weaknesses, potential, and areas of improvement associated with such models for membrane fouling prediction.

Impact of SWMM Fouling and Position on the Performance of SWRO Systems in Operating Conditions of Minimum SEC

Due to water stress in the world in general desalination technologies are becoming increasingly important. Among the available technologies, reverse osmosis (RO) is the most widespread due to its reliability and efficiency compared to other technologies. The main weakness of RO is the loss of performance due to membrane fouling, which usually affects the water permeability coefficient (A), causing it to decrease. In RO desalination plants, fouling does not affect all spiral wound membrane modules (SWMMs) in the pressure vessels (PVs) in the same way. This will depend on the type of fouling and the position of the SWMM inside the PV. In this study, the impact of A and the position of the SWMM on the performance of the RO system is analyzed. For this purpose, decrements of up to 50% have been assumed for the seven SWMMs in series considering nine commercial SWMM models. The operating point analyzed is that which minimizes the specific energy consumption (SEC), a point obtained in a previous work carried out by the authors. The results show how the impact of A on the SWMM in the first position is more significant than the impact on modules that are in another position for the nine SWRO models studied. A drop of 50% in the coefficient A of the first element produces a permeate loss in the pressure pipe between 0.67 and 1.35 m3 d-1. Furthermore, it was observed that the models with the lowest coefficient A exhibited the highest performance losses in terms of permeate production when A was decreased.

Recent Advances in Membranes Used for Nanofiltration to Remove Heavy Metals from Wastewater: A Review

The presence of heavy metal ions in polluted wastewater represents a serious threat to human health, making proper disposal extremely important. The utilization of nanofiltration (NF) membranes has emerged as one of the most effective methods of heavy metal ion removal from wastewater due to their efficient operation, adaptable design, and affordability. NF membranes created from advanced materials are becoming increasingly popular due to their ability to depollute wastewater in a variety of circumstances. Tailoring the NF membrane's properties to efficiently remove heavy metal ions from wastewater, interfacial polymerization, and grafting techniques, along with the addition of nano-fillers, have proven to be the most effective modification methods. This paper presents a review of the modification processes and NF membrane performances for the removal of heavy metals from wastewater, as well as the application of these membranes for heavy metal ion wastewater treatment. Very high treatment efficiencies, such as 99.90%, have been achieved using membranes composed of polyvinyl amine (PVAM) and glutaraldehyde (GA) for Cr³⁺ removal from wastewater. However, nanofiltration membranes have certain drawbacks, such as fouling of the NF membrane. Repeated cleaning of the membrane influences its lifetime.

Artificial intelligence in wastewater treatment: A data-driven analysis of status and trends

Wastewater treatment is a complex process that involves many uncertainties, leading to fluctuations in effluent quality and costs, and environmental risks. Artificial intelligence (AI) can handle complex nonlinear problems and has become a powerful tool for exploring and managing wastewater treatment systems. This study provides a summary of the current status and trends in AI research as applied to wastewater treatment, based on published papers and patents. Our results indicate that, at present, AI is primarily used to evaluate removal of pollutants (conventional, typical, and emerging contaminants), optimize models and process parameters, and control membrane fouling. Future research will likely continue to focus on removal of phosphorus, organic pollutants, and emerging contaminants. Moreover, analyzing microbial community dynamics and achieving multi-objective optimization are promising directions of research. The knowledge map shows that there may be future technological innovation related to predicting water quality under specific conditions, integrating AI with other information technologies and utilizing image-based AI and other algorithms in wastewater treatment. In addition, we briefly review development of artificial neural networks (ANNs) and explore the evolutionary path of AI in wastewater treatment. Our findings provide valuable insights into potential opportunities and challenges for researchers applying AI to wastewater treatment.

Patterned membranes for improving hydrodynamic properties and mitigating membrane fouling in water treatment: A review

Membrane technologies have been widely applied in water treatment over the past few decades. However, membrane fouling remains a hinderance for the widespread use of membrane processes because it decreases effluent quality and increases operating costs. To mitigate membrane fouling, researchers have been exploring effective anti-fouling strategies. Recently, patterned membranes are gaining attention as a novel non-chemical membrane modification for membrane fouling control. In this paper, we review the research on patterned membranes used in water treatment over the past 20 years. In general, patterned membranes show superior anti-fouling performances, which mainly results from two aspects: hydrodynamic effects and interaction effects. Due to the introduction of diversified topographies onto the membrane surface, patterned membranes yield dramatic improvements on hydrodynamic properties, e.g., shear stress, velocity field and local turbulence, restraining concentration polarization and foulants' deposition on the membrane surface. Besides, the membrane-foulant and foulant-foulant interactions play an important role in the mitigation of membrane fouling. Due to the existence of surface patterns, the hydrodynamic boundary layer is destroyed and the interaction force as well as the contact area between foulants and surface are decreased, which contributes to the fouling suppression. However, there are still some limitations in the research and application of patterned membranes. Future research is suggested to focus on the development of patterned membranes appropriate for different water treatment scenarios, the insights into the interaction forces affected by surface patterns, and the pilot-scale and long-term studies to verify the anti-fouling performances of patterned membranes in practical applications.

The roles of artificial intelligence techniques for increasing the prediction performance of important parameters and their optimization in membrane processes: A systematic review

Membrane-based separation processes has been recently of significant global interest compared to other conventional separation approaches due to possessing undeniable advantages like superior performance, environmentally-benign nature and simplicity of application. Computational simulation of fluids has shown its undeniable role in modeling and simulation of numerous physical/chemical phenomena including chemical engineering, chemical reaction, aerodynamics, drug delivery and plasma physics. Definition of fluids can be occurred using the Navier-Stokes equations, but solving the equations remains an important challenge. In membrane-based separation processes, true perception of fluid's manner through disparate membrane modules is an important concern, which has been significantly limited applying numerical/computational procedures such s computational fluid dynamics (CFD). Despite this noteworthy advantage, the optimization of membrane processes using CFD is time-consuming and expensive. Therefore, combination of artificial intelligence (AI) and CFD can result in the creation of a promising hybrid model to accurately predict the model results and appropriately optimize membrane processes and phase separation. This paper aims to provide a comprehensive overview about the advantages of commonly-employed ML-based techniques in combination with the CFD to intelligently increase the optimization accuracy and predict mass transfer and the unfavorable events (i.e., fouling) in various membrane processes. To reach this objective, four principal strategies of AI including SL, USL, SSL and ANN were explained and their advantages/disadvantages were discussed. Then after, prevalent ML-based algorithm for membrane-based separation processes. Finally, the application potential of AI techniques in different membrane processes (i.e., fouling control, desalination and wastewater treatment) were presented.

Operational parameter prediction of electrocoagulation system in a rural decentralized water treatment plant by interpretable machine learning model

Electrocoagulation (EC) is a promising alternative for decentralized drinking water treatment in rural areas as a chemical-free technology. However, seasonal fluctuations of water quality in influent remain a significant challenge for rural decentralized water supply, which was a potential threat to water safety. The frequent operation was required to ensure the effluent water quality by the experienced technicians, who were in shortage in rural areas. If the operational parameters prediction model based on water quality could be established, it might reduce the dependence on technicians. Therefore, an artificial neural network (ANN) model combined with genetic algorithm (GA) was used to establish a prediction model for unattended intelligent operation. Data on water quality and operational parameters were collected from a practical EC system in a decentralized water treatment plant. Seven water quality parameters (e.g., turbidity, temperature, pH and conductivity) were selected as input variables and the operational current was employed as the output. A non-linear relationship between water quality parameters and the operational current was verified by correlation analysis and principal component analysis (PCA). The mean squared error (MSE) and coefficient of determination (R²) were used as evaluation indexes to optimize the structure of the GA-ANN model. Influent turbidity was identified to be crucial in the GA-ANN model by model interpretation using sensitivity analysis and scenario analysis. The Garson weight of turbidity in the seven input variables achieved 45.4%. The predictive accuracy of the GA-ANN model sharply declined from 90% to 67.1% when influent turbidity data were absent. In addition, it was estimated that energy consumption savings of the GA-ANN method declined by 14.2% in comparison with the gradient control method. This study verifies the feasibility and stability of machine learning strategy for unattended operation in the rural decentralized water treatment plant.

Deep Study on Fouling Modelling of Ultrafiltration Membranes Used for OMW Treatment: Comparison Between Semi-empirical Models, Response Surface, and Artificial Neural Networks

Olive oil production generates a large amount of wastewater called olive mill wastewater. This paper presents the study of the effect of transmembrane pressure and cross flow velocity on the decrease in permeate flux of different ultrafiltration membranes (material and pore size) when treating a two-phase olive mill wastewater (olive oil washing wastewater). Both semi-empirical models (Hermia models adapted to tangential filtration, combined model, and series resistance model), as well as statistical and machine learning methods (response surface methodology and artificial neural networks), were studied. Regarding the Hermia model, despite the good fit, the main drawback is that it does not consider the possibility that these mechanisms occur simultaneously in the same process. According to the accuracy of the fit of the models, in terms of R2 and SD, both the series resistance model and the combined model were able to represent the experimental data well. This indicates that both cake layer formation and pore blockage contributed to membrane fouling. The inorganic membranes showed a greater tendency to irreversible fouling, with higher values of the Ra/RT (adsorption/total resistance) ratio. Response surface methodology ANOVA showed that both cross flow velocity and transmembrane pressure are significant variables with respect to permeate flux for all membranes studied. Regarding artificial neural networks, the tansig function presented better results than the selu function, all presenting high R2, ranging from 0.96 to 0.99. However, the comparison of all the analyzed models showed that depending on the membrane, one model fits better than the others. Finally, through this work, it was possible to provide a better understanding of the data modelling of different ultrafiltration membranes used for the treatment of olive mill wastewater.

Modeling of Organic Fouling in an Ultrafiltration Cell Using Different Three-Dimensional Printed Turbulence Promoters

Designing turbulence promoters with optimal geometry and using them for ultrafiltration systems has been a key challenge in mitigating membrane fouling. In this study, six different turbulence promoters were created using three-dimensional printing technology and applied in dead-end ultrafiltration. Three-dimensional-printed (3DP) turbulence promoter configurations were integrated into a classical batch ultrafiltration cell. The effects of these configurations and the stirring speeds on the permeate filtration flux, organic rejections, and membrane resistances were investigated. The fouling control efficiency of the 3DP promoters was evaluated using two polyethersulfone membranes in a stirred ultrafiltration cell with model dairy wastewater. The Hermia and resistance-in-series models were studied to further investigate the membrane fouling mechanism. Of the Hermia models, the cake layer model best described the fouling in this membrane filtration system. It can be concluded that the 3DP turbulence promoters, combined with intense mechanical stirring, show great promise in terms of permeate flux enhancement and membrane fouling mitigation. Using a well-designed 3DP turbulence promoter improves the hydrodynamic flow conditions on the surface of the stirred membrane separation cells based on computational fluid dynamics modeling. Therefore, the factors effecting the fabrication of 3DP turbulence promoters are important, and further research should be devoted to revealing them.

Quantification of interfacial interaction related with adhesive membrane fouling by genetic algorithm back propagation (GABP) neural network

Since adhesive membrane fouling is critically determined by the interfacial interaction between a foulant and a rough membrane surface, efficient quantification of the interfacial interaction is critically important for adhesive membrane fouling mitigation. As a current available method, the advanced extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) theory involves complicated rigorous thermodynamic equations and massive amounts of computation, restricting its application. To solve this problem, artificial intelligence (AI) visualization technology was used to analyze the existing literature, and the genetic algorithm back propagation (GABP) artificial neural network (ANN) was employed to simplify thermodynamic calculation. The results showed that GABP ANN with 5 neurons could obtain reliable prediction performance in seconds, versus several hours or even days time-consuming by the advanced XDLVO theory. Moreover, the regression coefficient (R) of GABP reached 0.9999, and the error between the prediction results and the simulation results was less than 0.01%, indicating feasibility of the GABP ANN technique for quantification of interfacial interaction related with adhesive membrane fouling. This work provided a novel strategy to efficiently optimize the thermodynamic prediction of adhesive membrane fouling, beneficial for better understanding and control of adhesive membrane fouling.

Simultaneous degradation and separation of antibiotics in sewage effluent by photocatalytic nanofiltration membrane in a continuous dynamic process

Bifunctional photocatalytic nanofiltration (PNF) membrane is increasingly concerned in practical micro-polluted water purification, but there are still several bottlenecks that inhibit its practicality. In this context, the feasibility of a novel metal-free and visible light-responsive surface-anchored PNF membrane for simultaneously removing target antibiotics in real sewage effluent in a continuous dynamic process was explored. The results showed that the optimal PNF-4 membrane was expectedly consisted of an inside tight sub-nanopore structured separation layer and an outside thinner, smoother, super hydrophilic mesoporous degradation layer, respectively. Consequently, the activated PNF-4 membrane could synergistically reduce trimethoprim and sulfamethoxazole concentrations to below two orders of magnitude, accompanying with almost constant high water permeability, suggesting that the hydrophilic modification of the mesoporous degradation layer basically offsets its inherent hydraulic resistance. Also, after repeating the fouling-physical rinsing process three times lasted for 78 h, only sporadic adherent contaminants remained onto the top surface, together with the minimal total and irreversible fouling ratios (as low as 7.2% and 1.2%, respectively), strongly demonstrated that PNF-4 membrane displayed good self-cleaning performance. Undoubtedly, this will significantly reduce its potential cleaning frequency and maintenance cost in long-term operation. Meanwhile, the acute and chronic biotoxicities of its permeate to Virbrio qinghaiensis sp. -67 were also reduced sharply to 2.22% and 0.45%, respectively. All of these evidences suggest that the dual functions of PNF-4 membrane are synergetic in an uninterrupted permeating process. It will provide useful insights for continuously enhancing the practicality and effectiveness of PNF membrane in actual micro-polluted water purification scenarios.

3D printing titanium dioxide-acrylonitrile-butadiene-styrene (TiO2-ABS) composite membrane for efficient oil/water separation

The oily water treatment is becoming one of the hottest topics due to that increase of offshore oil transportation and the various accident oil leakages. In this study, a functional TiO₂-ABS composite membrane was generated through the three-dimensional (3D) printing strategy for the first time and was conducted to simulated oily water treatment. The TiO₂-ABS composite membrane demonstrated a significant promotion in hydrophilicity and oleophobicity which were evidenced by the water contact angle of 14.8° and the underwater oil contact angle of 144.7°, respectively. The optimal modified membrane had both exceedingly high flux (1.8 × 10⁵ L m^-2·h^-1) and oil rejection rate (99.5%). Moreover, the results of filtration cycles of 10 days and extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) theory demonstrated that the modified membranes took possession of excellent stability and antifouling property. What was more, the TiO₂-ABS composite membrane revealed over 99% rejection to all five types of oil/water systems. The interestingly experimental results indicated that the prepared membrane possessed a broad development trend and application prospect in the field of oily water treatment.

Advances in the Removal of Cr(III) from Spent Industrial Effluents-A Review

The review presents advances in the removal of Cr(III) from the industrial effluents published in the last ten years. Although Cr(III) has low solubility and is less dangerous for the aquatic environment than Cr(VI), it cannot be released into the aquatic environment without limitations and its content in water should be restricted. The development of efficient techniques for the removal of Cr(III) is also a response to the problem of chromium wastewater containing Cr(VI) ions. Very often the first step in dealing with such wastewater is the reduction in chromium content. In some cases, removal of Cr(III) from wastewaters is an important step for pretreatment of solutions to prepare them for subsequent recovery of other metals. In the review, hydrometallurgical operations for Cr(III) removal are presented, including examples of Cr(III) recovery from real industrial effluents with precipitation, adsorption, ion exchange, extraction, membrane techniques, microbial-enhanced techniques, electrochemical methods. The advantages and disadvantages of the operations mentioned are also presented. Finally, perspectives for the future in line with circular economy and low-environmental impact are briefly discussed.

An Overview of the Modification Strategies in Developing Antifouling Nanofiltration Membranes

Freshwater deficiency has become a significant issue affecting many nations' social and economic development because of the fast-growing demand for water resources. Nanofiltration (NF) is one of the promising technologies for water reclamation application, particularly in desalination, water, and wastewater treatment fields. Nevertheless, membrane fouling remains a significant concern since it can reduce the NF membrane performance and increase operating expenses. Consequently, numerous studies have focused on improving the NF membrane's resistance to fouling. This review highlights the recent progress in NF modification strategies using three types of antifouling modifiers, i.e., nanoparticles, polymers, and composite polymer/nanoparticles. The correlation between antifouling performance and membrane properties such as hydrophilicity, surface chemistry, surface charge, and morphology are discussed. The challenges and perspectives regarding antifouling modifiers and modification strategies conclude this review.

A Review on Membrane Biofouling: Prediction, Characterization, and Mitigation

Water scarcity is an increasing problem on every continent, which instigated the search for novel ways to provide clean water suitable for human use; one such way is desalination. Desalination refers to the process of purifying salts and contaminants to produce water suitable for domestic and industrial applications. Due to the high costs and energy consumption associated with some desalination techniques, membrane-based technologies have emerged as a promising alternative water treatment, due to their high energy efficiency, operational simplicity, and lower cost. However, membrane fouling is a major challenge to membrane-based separation as it has detrimental effects on the membrane's performance and integrity. Based on the type of accumulated foulants, fouling can be classified into particulate, organic, inorganic, and biofouling. Biofouling is considered the most problematic among the four fouling categories. Therefore, proper characterization and prediction of biofouling are essential for creating efficient control and mitigation strategies to minimize the damage associated with biofouling. Moreover, the use of artificial intelligence (AI) in predicting membrane fouling has garnered a great deal of attention due to its adaptive capability and prediction accuracy. This paper presents an overview of the membrane biofouling mechanisms, characterization techniques, and predictive methods with a focus on AI-based techniques, and mitigation strategies.

Machine learning modelling of a membrane capacitive deionization (MCDI) system for prediction of long-term system performance and optimization of process control parameters in remote brackish water desalination

Membrane Capacitive Deionization (MCDI) is a promising electrochemical technique for water desalination. Previous studies have confirrmed the effectiveness of MCDI in removing contaminants from brackish groundwaters, especially in remote areas where electricity is scarce. However, as with other water treatment technologies, performance deterioration of the MCDI system still occurs, hindering the stability of long-term operation. Herein, a machine learning (ML) modelling framework and various ML models were developed to (i) investigate the performance deterioration due particularly to insufficient charging/discharging of the electrode caused by accumulation of ions and electrode scaling and (ii) optimise MCDI operating parameters such that the impacts of these deleterious effects on unit performance were minimized. The ML models developed in this work exhibited a prediction accuracy of cycle time with average mean absolute percentage error (MAPE) values of 16.82% and 16.09% after 30-fold cross validation for Random Forest (RF) and Multilayer Perceptron (MLP) models respectively. The pre-trained ML model predicted different declining trends of water production for two different operating conditions and provided corresponding recommendations on frequencies of chemical cleaning. A case study on the adjustment of operating parameters using the results suggested by the optimization ML model was conducted. The model validation results showed that the overall water production and water recovery of the system using the cycle-based optimized process control parameters (SCN 1) exceeds the MCDI system performance under three fixed parameter settings that were used at each stage of SCN 1 by 1.78% to 4.48% and 2.95% to 9.46%, respectively. Permutation-based and Shapley additive explanation (SHAP) coefficients were also employed for variable importance (VIMP) analysis to uncover the "black-box" nature of the ML models and to better understand the various features' contributions to overall MCDI system performance.

Molecular insights into membrane fouling caused by polysaccharides with different structures in polyaluminum chloride coagulation-ultrafiltration process

In this study, mechanisms of membrane fouling caused by polysaccharides with different molecular structures in polyaluminum chloride (PACl) coagulation-ultrafiltration (C-UF) process were explored. Carrageenan and xanthan gum were chosen for model foulants of straight chain and branched chain polysaccharides, respectively. Filtration experiments showed that, with PACl dosage of 0-5 mM, specific filtration resistance (SFR) of carrageenan and xanthan solution showed a unimodal pattern and a continuous decrease pattern, respectively. A series of experimental characterizations indicated that the different SFR pattern was closely related to structure of foulants layer. Density functional theory (DFT) calculation suggested that Al³⁺ preferentially coordinating with the terminal sulfonyl groups of carrageenan chains to promote gel layer formation at low PACl concentration (0.15 mM). There existed a chemical potential gap between bound water in gel layer and free water in the permeate, so that, filtration through gel layer corresponded to rather high SFR for overcoming this gap. In contrast, Al³⁺ coordinating with the non-terminal sulfonyl groups of carrageenan at high PACl concentration caused transition from gel layer to cake layer, leading to SFR decrease. However, xanthan gum itself can form a dense gel layer with a complex polymer network by virtue of the interlacing of main chains and branches. Al³⁺ coordinating with the carboxyl groups on branched chains of xanthan gum resulted in clusters of polymer chains and flocculation, corresponding to the reduced SFR. This proposed molecular-level mechanism well explained membrane fouling behaviors of polysaccharides with different molecular structure, and also facilitated to optimize C-UF process for water treatment.

Mechanistic insights into Ca-alginate gel-associated membrane fouling affected by ethylene diamine tetraacetic acid (EDTA)

While transparent exopolymer particles (TEP) is a major foulant, and ethylene diamine tetraacetic acid (EDTA) is a strong chelating agent frequently used for fouling mitigation in membrane-based water treatment processes, little has been known about TEP-associated membrane fouling affected by EDTA. This work was performed to investigate roles of EDTA addition in TEP (Ca-alginate gel was used as a TEP model) associated fouling. It was interestingly found that, TEP had rather high specific filtration resistance (SFR) of 2.49 × 10¹⁵ m^-1·kg^-1, and SFR of TEP solution firstly decreased and then increased rapidly with EDTA concentration increase (0-1 mM). A series of characterizations suggested that EDTA took roles in SFR of TEP solution by means of changing TEP microstructure. The rather high SFR of TEP layer can be attributed to the big chemical potential gap during filtration described by the extended Flory-Huggins lattice theory. Initial EDTA addition disintegrated TEP structure by EDTA chelating calcium in TEP, inducing reduced SFR. Continuous EDTA addition decreased solution pH, resulting into no effective chelating and accumulation of EDTA on membrane surface, increasing SFR. It was suggested that factors increasing homogeneity of TEP gel will increase SFR, and vice versa. This study revealed the thermodynamic mechanism of TEP fouling behaviors affected by EDTA, and also demonstrated the importance of EDTA dosage and pH adjustment for TEP-associated fouling control.

Comparison of Artificial Intelligence Control Strategies for a Peristaltically Pumped Low-Pressure Driven Membrane Process

Peristaltic pumping is used in membrane applications where high and sterile sealing is required. However, control is difficult due to the pulsating pump characteristics and the time-varying properties of the system. In this work, three artificial intelligence control strategies (artificial neural networks (ANN), fuzzy logic expert systems, and fuzzy-integrated local models) were used to regulate transmembrane pressure and crossflow velocity in a microfiltration system under high fouling conditions. A pilot plant was used to obtain the necessary data to identify the AI models and to test the controllers. Humic acid was employed as a foulant, and cleaning-in-place with NaOH was used to restore the membrane state. Several starting operating points were studied and setpoint changes were performed to study the plant dynamics under different control strategies. The results showed that the control approaches were able to control the membrane system, but significant differences in the dynamics were observed. The ANN control was able to achieve the specifications but showed poor dynamics. Expert control was fast but showed problems in different working areas. Local models required less data than ANN, achieving high accuracy and robustness. Therefore, the technique to be used will depend on the available information and the application dynamics requirements.

Purifying water with silver nanoparticles (AgNPs)-incorporated membranes: Recent advancements and critical challenges

In the face of the growing global water crisis, membrane technology is a promising means of purifying water and wastewater. Silver nanoparticles (AgNPs) have been widely used to improve membrane performance, for antibiofouling, and to aid in photocatalytic degradation, thermal response, and electro-conductivity. However, several critical issues such as short antimicrobial periods, trade-off effects and silver inactivation seriously restrict the engineering application of AgNPs-incorporated membranes. In addition, there is controversy around the use of AgNPs given the toxic preparation process and environmental/biological risks. Hence, it is of great significance to summarize and analyze the recent developments and critical challenges in the use of AgNPs-incorporated membranes in water and wastewater treatment, and to propose potential solutions. We reviewed the different properties and functions of AgNPs and their corresponding applications in AgNPs-incorporated membranes. Recently, multifunctional, novel AgNP-incorporated membranes combined with other functional materials have been developed with high-performance. We further clarified the synergistic mechanisms between AgNPs and these novel nanomaterials and/or polymers, and elucidated their functions and roles in membrane separation. Finally, the critical challenges of AgNPs-incorporated membranes and the proposed solutions were outlined: i) Prolonging the antimicrobial cycle through long-term and controlled AgNPs release; ii) Overcoming the trade-off effect and organic fouling of the AgNPs-incorporated membranes; iii) Preparation of sustainable AgNPs-incorporated membranes; iv) Addressing biotoxicity induced by AgNPs; and v) Deactivation of AgNPs-incorporated membrane. Overall, this review provides a comprehensive discussion of the advancements and challenges of AgNPs-incorporated membranes and guides the development of more robust, multi-functional and sustainable AgNPs-incorporated membranes.