A practical hybrid modelling approach for the prediction of potential fouling parameters in ultrafiltration membrane water treatment plant

Applications of artificial intelligence (AI) in drinking water treatment processes: Possibilities

In water treatment processes (WTPs), artificial intelligence (AI) based techniques, particularly machine learning (ML) models have been increasingly applied in decision-making activities, process control and optimization, and cost management. At least 91 peer-reviewed articles published since 1997 reported the application of AI techniques to coagulation/flocculation (41), membrane filtration (21), disinfection byproducts (DBPs) formation (13), adsorption (16) and other operational management in WTPs. In this paper, these publications were reviewed with the goal of assessing the development and applications of AI techniques in WTPs and determining their limitations and areas for improvement. The applications of the AI techniques have improved the predictive capabilities of coagulant dosages, membrane flux, rejection and fouling, disinfection byproducts (DBPs) formation and pollutants' removal for the WTPs. The deep learning (DL) technology showed excellent extraction capabilities for features and data mining ability, which can develop an image recognition-based DL framework to establish the relationship among the shapes of flocs and dosages of coagulant. Further, the hybrid techniques (e.g., combination of regression and AI; physical/kinetics and AI) have shown better predictive performances. The future research directions to achieve better control for WTPs through improving these techniques were also emphasized.

Regression Metamodel-Based Digital Twin for an Industrial Dynamic Crossflow Filtration Process

Dynamic crossflow filtration (DCF) is the state-of-the-art technology for solid-liquid separation from viscous and sensitive feed streams in the food and biopharma industry. Up to now, the potential of industrial processes is often not fully exploited, because fixed recipes are usually applied to run the processes. In order to take the varying properties of biological feed materials into account, we aim to develop a digital twin of an industrial brownfield DCF plant, allowing to optimize setpoint decisions in almost real time. The core of the digital twin is a mechanistic-empirical process model combining fundamental filtration laws with process expert knowledge. The effect of variation in the selected process and model parameters on plant productivity has been assessed using a model-based design-of-experiments approach, and a regression metamodel has been trained with the data. A cyclic program that bidirectionally communicates with the DCF asset serves as frame of the digital twin. It monitors the process dynamics membrane torque and transmembrane pressure and feeds back the optimum permeate flow rate setpoint to the physical asset in almost real-time during process runs. We considered a total of 24 industrial production batches from the filtration of grape juice from the years 2022 and 2023 in the study. After implementation of the digital twin on site, the campaign mean productivity increased by 15% over the course of the year 2023. The presented digital twin framework is a simple example how an industrial established process can be controlled by a hybrid model-based algorithm. With a digital process dynamics model at hand, the presented metamodel optimization approach can be easily transferred to other (bio)chemical processes.

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.

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.

Existing Filtration Treatment on Drinking Water Process and Concerns Issues

Water is one of the main sources of life's survival. It is mandatory to have good-quality water, especially for drinking. Many types of available filtration treatment can produce high-quality drinking water. As a result, it is intriguing to determine which treatment is the best. This paper provides a review of available filtration technology specifically for drinking water treatment, including both conventional and advanced treatments, while focusing on membrane filtration treatment. This review covers the concerns that usually exist in membrane filtration treatment, namely membrane fouling. Here, the parameters that influence fouling are identified. This paper also discusses the different ways to handle fouling, either based on prevention, prediction, or control automation. According to the findings, the most common treatment for fouling was prevention. However, this treatment required the use of chemical agents, which will eventually affect human health. The prediction process was usually used to circumvent the process of fouling development. Based on our reviews up to now, there are a limited number of researchers who study membrane fouling control based on automation. Frequently, the treatment method and control strategy are determined individually.

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

Membrane fouling is one of major obstacles in the application of membrane technologies. Accurately predicting or simulating membrane fouling behaviours is of great significance to elucidate the fouling mechanisms and develop effective measures to control fouling. Although mechanistic/mathematical models have been widely used for predicting membrane fouling, they still suffer from low accuracy and poor sensitivity. To overcome the limitations of conventional mathematical models, artificial intelligence (AI)-based techniques have been proposed as powerful approaches to predict membrane filtration performance and fouling behaviour. This work aims to present a state-of-the-art review on the advances in AI algorithms (e.g., artificial neural networks, fuzzy logic, genetic programming, support vector machines and search algorithms) for prediction of membrane fouling. The working principles of different AI techniques and their applications for prediction of membrane fouling in different membrane-based processes are discussed in detail. Furthermore, comparisons of the inputs, outputs, and accuracy of different AI approaches for membrane fouling prediction have been conducted based on the literature database. Future research efforts are further highlighted for AI-based techniques aiming for a more accurate prediction of membrane fouling and the optimization of the operation in membrane-based processes.

Predicting flocculant dosage in the drinking water treatment process using Elman neural network

Predicting the flocculant dosage in the drinking water treatment process is essential for public health. However, due to the complexity of water quality and flocculation, many difficulties remain. The present study aimed to report on using artificial intelligence, namely, the Elman neural network (ENN), to predict the flocculant dosage and explore the applications of the proposed model in waterworks. The flocculation process of drinking water was introduced in this study, and four typical models were developed based on multiple linear regression (MLR), the radial basis function neural network (RBFNN), the least squares support vector machine (LSSVM), and the ENN. To improve the prediction accuracy, a mixed term including long-term data and short-term data was proposed to capture the periodic and time-varying characteristics of water quality data. The weights of each part are updated adaptively according to the comparison of effluent turbidity and set values. The results demonstrate that the proposed ENN model performed better than the other three models in terms of the prediction performance. With the ENN model of flocculant dosage, the root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R²) of the test data were 1.8917, 5.0067, and 0.8999, which were improved by 36.9%, 41.5%, and 14.0% in comparison with the best one (RBFNN) of the other three models, respectively. The effluent turbidity of sedimentation tank was more stable under the control of proposed ENN model of flocculant dosage than the other three models. Considering its performance, the ENN model can be taken as a preferred data intelligence tool for predicting the drinking water flocculant dosage.

A Review on the Use of Membrane Technology Systems in Developing Countries

Fulfilling the demand of clean potable water to the general public has long been a challenging task in most developing countries due to various reasons. Large-scale membrane water treatment systems have proven to be successful in many advanced countries in the past two decades. This paves the way for developing countries to study the feasibility and adopt the utilization of membrane technology in water treatment. There are still many challenges to overcome, particularly on the much higher capital and operational cost of membrane technology compared to the conventional water treatment system. This review aims to delve into the progress of membrane technology for water treatment systems, particularly in developing countries. It first concentrates on membrane classification and its application in water treatment, including membrane technology progress for large-scale water treatment systems. Then, the fouling issue and ways to mitigate the fouling will be discussed. The feasibility of membrane technologies in developing countries was then evaluated, followed by a discussion on the challenges and opportunities of the membrane technology implementation. Finally, the current trend of membrane research was highlighted to address future perspectives of the membrane technologies for clean water production.

Prediction of lead (Pb) adsorption on attapulgite clay using the feasibility of data intelligence models

This study investigates the performance of support vector machine (SVM), multivariate adaptive regression spline (MARS), and random forest (RF) models for predicting the lead (Pb) adsorption by attapulgite clay. Models are constructed using batch stochastic data of heavy metal (HM) concentrations under different physicochemical conditions. Implementation of auto-hyper-parameter tuning using grid-search approach and comparative analysis is performed against the benchmark artificial intelligence (AI) models. Models are constructed based on Pb concentration (IC), the dosage of attapulgite clay (dose), contact time (CT), pH, and NaNO3 (SN). Principle component analysis (PCA) and correlation analysis (CA) methods are integrated to assess the importance of the applied predictors and their relationship with the target. Research findings approved the potential of the grid-RF model as a marginal superior predictive model against the grid-SVM in terms of MAE, i.e., 3.29 and 3.34, respectively; moreover, the md scored the same, i.e., 0.93, which reveals the potential predictability for both. Nonetheless, grid-MARS and standalone MARS models remained likewise in their predictability. IC parameter demonstrated the highest influential among all the predictors with the highest value of importance in the case of all three evaluators. The solution pH and dose stands together with marginal differences in case of PCA method; however, solution pH and CT appeared with similarity impact using the PCA method.

Hydraulic Resistance and Protein Fouling Resistance of a Zirconia Membrane with a Tethered PVP Layer

The influence of surface modification of zirconia (ZrO2) membrane with tethered poly(vinyl pyrrolidone) (PVP) chains was evaluated with respect to the impact of pH and ionic strength on hydraulic resistance and fouling resistance in the filtration of bovine serum albumin (BSA) and lysozyme (Lys) as model protein foulants. The tethered PVP surface layer led to membrane permeability and fouling propensity that were responsive to both pH and ionic strength. The PVP-modified membrane (PVP-ZrO2) hydraulic resistance increased by up to ~48% over a pH range of 6–11, but with no discernible impact at lower pH. Membrane hydraulic resistance was virtually unaffected by ionic strength over the 0.001–1 M range. However, reversible foulant cake resistance in BSA and Lys solution filtration increased with elevated ionic strength, owing in part to the weakening of protein–protein repulsion. Irreversible BSA and Lys fouling was affected by the operational pH relative to the protein isoelectric point (IEP) and reduced under conditions of chain swelling. Irreversible membrane fouling resistance for both proteins was significantly lower, by ~11–49% and 18–74%, respectively, for the PVP-ZrO2 membrane relative to the unmodified ZrO2 membrane. The present results suggest the merit of further exploration of fouling reduction and improvement of membrane cleaning effectiveness via tuning pH and ionic strength triggered conformational responsiveness of the tethered target polymer layer.

Virus removal by membrane bioreactors: A review of mechanism investigation and modeling efforts

The increasing pressure on the global water supply calls for more advanced solutions with higher efficiency and better sustainability, leading to the promptly developing water reclamation and reuse schemes including treatment technologies and risk management strategies where microbial safety is becoming a crucial aspect in the interest of public health. Backed up by the development of membrane technology, membrane bioreactors (MBR) have received substantial attention for their superiority over conventional treatment methods in many ways and are considered promising in the water reclamation realm. This review paper provides an overview of the efforts made to manage and control the potential waterborne viral disease risks raised by the use of effluent from MBR treatment processes, including the mechanisms involved in the virus removal process and the attempts to model the dynamics of the removal process. In principle, generalized and integrated virus removal models that provide insight into real-time monitoring are urgently needed for advanced real-time control purpose. Future studies of approaches that can well handle the inherent uncertainty and nonlinearity of the complex removal process are crucial to the development and promotion of related technologies.

Permeate Flux Control in SMBR System by Using Neural Network Internal Model Control

This paper presents a design of a data-driven-based neural network internal model control for a submerged membrane bioreactor (SMBR) with hollow fiber for microfiltration. The experiment design is performed for measurement of physical parameters from an actuator input (permeate pump voltage), which gives the information (outputs) of permeate flux and trans-membrane pressure (TMP). The palm oil mill effluent is used as an influent preparation to depict fouling phenomenon in the membrane filtration process. From the experiment, membrane fouling potential is observed from flux decline pattern, with a rapid increment of TMP (above 200 mbar). Membrane fouling is a complex process and the available models in literature are not designed for control system (filtration performance). Therefore, this work proposes an aeration fouling control strategy to measure the filtration performance. The artificial neural networks (Feed-Forward Neural Network—FFNN, Radial Basis Function Neural Network—RBFNN and Nonlinear Autoregressive Exogenous Neural Network—NARXNN) are used to model dynamic behaviour of flux and TMP. In this case, only flux is used in closed loop control application, whereby the TMP effect is used for monitoring. The simulation results show that reliable prediction of membrane fouling potential is obtained. It can be observed that almost all the artificial neural network (ANN) models have similar shape with the actual data set, with the highest accuracy of more than 90% for both RBFNN and NARXN. The RBFNN is preferable due to simple structure of the network. In the control system, the RBFNN IMC depicts the highest closed loop performance with only 3.75 s (settling time) for setpoint changes when compared with other controllers. In addition, it showed fast performance in disturbance rejection with less overshoot. In conclusion, among the different neural network tested configurations the one based on radial basis function provides the best performance with respect to prediction as well as control performance.

Hybrid Modeling for Simultaneous Prediction of Flux, Rejection Factor and Concentration in Two-Component Crossflow Ultrafiltration

Ultrafiltration is a powerful method used in virtually every pharmaceutical bioprocess. Depending on the process stage, the product-to-impurity ratio differs. The impact of impurities on the process depends on various factors. Solely mechanistic models are currently not sufficient to entirely describe these complex interactions. We have established two hybrid models for predicting the flux evolution, the protein rejection factor and two components’ concentration during crossflow ultrafiltration. The hybrid models were compared to the standard mechanistic modeling approach based on the stagnant film theory. The hybrid models accurately predicted the flux and concentration over a wide range of process parameters and product-to-impurity ratios based on a minimum set of training experiments. Incorporating both components into the modeling approach was essential to yielding precise results. The stagnant film model exhibited larger errors and no predictions regarding the impurity could be made, since it is based on the main product only. Further, the developed hybrid models exhibit excellent interpolation properties and enable both multi-step ahead flux predictions as well as time-resolved impurity forecasts, which is considered to be a critical quality attribute in many bioprocesses. Therefore, the developed hybrid models present the basis for next generation bioprocessing when implemented as soft sensors for real-time monitoring of processes.

Membrane processes

The membrane processes have played important role in the industrial separation process. These technologies can be found in all industrial areas such as food, beverages, metallurgy, pulp and paper, textile, pharmaceutical, automotive, biotechnology and chemical industry, as well as in water treatment for domestic and industrial application. Although these processes are known since twentieth century, there are still many studies that focus on the testing of new membranes’ materials and determining of conditions for optimal selectivity, i. e. the optimum transmembrane pressure (TMP) or permeate flux to minimize fouling. Moreover the researchers proposed some calculation methods to predict the membrane processes properties. In this article, the laboratory scale experiments of membrane separation techniques, as well their validation by calculation methods are presented. Because membrane is the “heart” of the process, experimental and computational methods for its characterization are also described.