Dual-objective optimization for energy-saving and fouling mitigation in MBR plants using AI-based influent prediction and an integrated biological-physical model

Predicting Membrane Fouling of Submerged Membrane Bioreactor Wastewater Treatment Plants Using Machine Learning

Membrane fouling remains a significant challenge in the operation of membrane bioreactors (MBRs). Plant operators rely heavily on observations of filtration performance from noisy sensor data to assess membrane fouling conditions and lab-based protocols for plant maintenance, often leading to inaccurate estimations of future performance and delayed membrane cleaning. This challenge is further compounded by the difficulty in integrating existing complex mechanistic models with the Internet of Things (IoT) systems of wastewater treatment plants (WWTPs). By harnessing data obtained from WWTPs, along with innovative data denoising and model training strategies, we developed a machine learning application (MBR-Net) that is capable of forecasting membrane fouling, as indicated by permeability, for a full-scale submerged MBR plant in real time. We show that the trained model can effectively predict one-day-ahead changes in irreversible fouling under different desired fluxes, cleaning conditions and feedwater conditions (with MAPE < 6.45%, MAE < 3.71 LMH bar-1, and R2 > 0.87 on two independent testing sets). Although data availability presented certain limitations in the model development process, the current results demonstrate the significant value of machine learning in membrane fouling predictions and in providing decision support for fouling mitigation strategies in full-scale WWTPs.

Enhancing process monitoring and control in novel carbon capture and utilization biotechnology through artificial intelligence modeling: An advanced approach toward sustainable and carbon-neutral wastewater treatment

Integrating carbon capture and utilization (CCU) technologies into wastewater treatment plants (WWTPs) is essential for mitigating greenhouse gas (GHG) emissions and enhancing environmental sustainability, but further advancements in process monitoring and control are critical to optimizing treatment performance. This study investigates the application of artificial intelligence (AI) modeling to enhance process monitoring and control in a novel integrated CCU biotechnology with a moving bed biofilm reactor (MBBR) sequenced with an algal photobioreactor (aPBR). This system reduces GHG and odour emissions simultaneously. Several machine learning (ML) models, including artificial neural networks (ANNs), support vector machines (SVM), random forest (RF), and least-squares boosting (LSBoost), were tested. The LSBoost was the most suitable for modeling the MBBR + aPBR system, exhibiting the highest accuracy in predicting CO2 (R2 = 0.97) and H2S (R2 = 0.95) emissions from the MBBR. LSBoost also achieved the highest accuracy for predicting CO2 (R2 = 0.85) and H2S (R2 = 0.97) outlet concentrations from the aPBR. These findings underscore the importance of aligning AI algorithms to the characteristics of the treatment technology. The proposed AI models outperformed conventional statistical methods, demonstrating their ability to capture the complex, nonlinear dynamics typical of processes in environmental technologies. This study highlights the potential of AI-driven monitoring and control systems to significantly improve the efficiency of CCU biotechnologies in WWTPs for climate change mitigation and sustainable wastewater management.

Optimization of MBRs through integrated modelling: A state of the art

The optimization of integrated membrane bioreactors (MBRs) models is of paramount importance in view of reducing the costs, greenhouse gas emissions or enhancing the water quality. 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 control and optimization of integrated MBR models. Whether aerobic or anaerobic, such modelling allows the consideration of specific functioning conditions and optimization problems together with the estimation and monitoring of Performance Index (PIs). This paper reviews the diversity of those problems criteria used in performance assessment. Dividing issues that can be addressed either off-line or online, it is shown that integrated models have attained an important degree of maturity. Several recommendations for mainstreaming the optimization of MBRs using such integrated models. The key findings of this work show that there is room for improving and optimizing the functioning of MBRs using integrated modelling and that this integrated modelling approach is necessary to link functioning conditions together with PI estimation and monitoring.

Integrating artificial intelligence modeling and membrane technologies for advanced wastewater treatment: Research progress and future perspectives

Membrane technologies have become proficient alternatives for advanced wastewater treatment, ensuring high contaminant removal and sustainable resource recovery. Despite significant progress, ongoing research efforts aim to further optimize treatment performance. Among the challenges faced, membrane fouling persists as a relevant obstacle in membrane technologies, necessitating the development of more effective mitigation strategies. Mathematical models, widely employed for predicting treatment performance, generally exhibit low accuracy and suffer from uncertainties due to the complex and variable nature of wastewater. To overcome these limitations, numerous studies have proposed artificial intelligence (AI) modeling to accurately predict membrane technologies' performance and fouling mechanisms. This approach aims to provide advanced simulations and predictions, thereby enhancing process control, optimization, and intensification. This literature review explores recent advancements in modeling membrane-based wastewater treatment processes through AI models. The analysis highlights the enormous potential of this research field in enhancing the efficiency of membrane technologies. The role of AI modeling in defining optimal operating conditions, developing effective strategies for membrane fouling mitigation, enhancing the performance of novel membrane-based technologies, and improving membrane fabrication techniques is discussed. These enhanced process optimization and control strategies driven by AI modeling ensure improved effluent quality, optimized resource consumption, and minimized operating costs. The potential contribution of this cutting-edge approach to a paradigm shift toward sustainable wastewater treatment is examined. Finally, this review outlines future perspectives, emphasizing the research challenges that require attention to overcome the current limitations hindering the integration of AI modeling in wastewater treatment plants.

Membrane reciprocation and quorum quenching: An innovative combination for fouling control and energy saving in membrane bioreactors

Membrane bioreactors (MBRs) play a crucial role in wastewater treatment, but they face considerable challenges due to fouling. To tackle this issue, innovative strategies are needed. This study investigated the effectiveness of membrane reciprocation and quorum quenching (QQ) to control fouling in MBRs. The study compared MBRs using membrane reciprocation (30 rpm) and QQ (injecting media containing 100 or 200 mg/L BH4) with conventional MBRs employing different air-scouring intensities. The results demonstrated that combining membrane reciprocation (30 rpm) with QQ (200 mg/L BH4) significantly extended the service time of MBRs, making it approximately six times longer than conventional methods. Moreover, this approach reduced physically reversible resistance. The reduction in signal molecules related to biofouling due to QQ showcased its critical role in controlling biofouling, even under high shear caused by membrane reciprocation. However, the impact of QQ on microbial community structure appeared relatively insignificant when compared to factors such as operation time, aeration intensity, and membrane reciprocation. By combining membrane reciprocation and QQ, the study achieved a remarkable 81 % energy saving compared to extensive aeration (103 s-1 in velocity gradient), in addition to the extended service time. Importantly, this combined antifouling approach did not negatively affect microbial characteristics and wastewater treatment, emphasizing its effectiveness in MBRs. Overall, the findings of this study offer valuable insights for developing synergistic fouling control strategies in MBRs, significantly improving the energy efficiency of the wastewater treatment process.

Optimizing coagulant dosage using deep learning models with large-scale data

Water treatment plants are facing challenges that necessitate transition to automated processes using advanced technologies. This study introduces a novel approach to optimize coagulant dosage in water treatment processes by employing a deep learning model. The study utilized minute-by-minute data monitored in real time over a span of five years, marking the first attempt in drinking water process modeling to leverage such a comprehensive dataset. The deep learning model integrates a one-dimensional convolutional neural network (Conv1D) and gated recurrent unit (GRU) to effectively extract features and model complex time-series data. Initially, the model predicted coagulant dosage and sedimentation basin turbidity, validated against a physicochemical model. Subsequently, the model optimized coagulant dosage in two ways: 1) maintaining sedimentation basin turbidity below the 1.0 NTU guideline, and 2) analyzing changes in sedimentation basin turbidity resulting from reduced coagulant dosage (5-20%). The findings of the study highlight the effectiveness of the deep learning model in optimizing coagulant dosage with substantial reductions in coagulant dosage (approximately 22% reduction and 21 million KRW/year). The results demonstrate the potential of deep learning models in enhancing the efficiency and cost-effectiveness of water treatment processes, ultimately facilitating process automation.

Modification of a plant-scale semi-centralized wastewater treatment system to enhance nitrogen and phosphorus removal from black water

Owing to the low ratio of chemical oxygen demand to total nitrogen (SCOD/TN), effective removal of nutrient pollutants from black water is difficult. In this study, to enhance nitrogen and phosphorus removal from such wastewater, a series of operational modification strategies was investigated and applied to a plant-scale semi-centralized system used for black water treatment. The results showed that 21 mg Fe³⁺/L was the optimal dosage for the chemical-enhanced pretreatment process, achieving average removal efficiencies of 51.1 and 74.1% for organics and phosphorus, respectively, with a slight enhancement in nitrogen removal by 2.3%. However, nitrogen and phosphorus removal could be further enhanced to 88 and 96%, by the addition of carbon sources in the post-anoxic zone of the reversed anaerobic-anoxic-aerobic process. Contrastingly, neither the addition of carbon sources in the pre-anoxic zone nor the prolongation of the time for pre-denitrification could significantly improve the nitrogen and phosphorus removal efficiencies. Furthermore, reducing the aeration intensity promoted simultaneous nitrification and denitrification in aerobic reactors, thereby making it a potential energy-saving method for system operation.

Materials discovery of ion-selective membranes using artificial intelligence

Significant attempts have been made to improve the production of ion-selective membranes (ISMs) with higher efficiency and lower prices, while the traditional methods have drawbacks of limitations, high cost of experiments, and time-consuming computations. One of the best approaches to remove the experimental limitations is artificial intelligence (AI). This review discusses the role of AI in materials discovery and ISMs engineering. The AI can minimize the need for experimental tests by data analysis to accelerate computational methods based on models using the results of ISMs simulations. The coupling with computational chemistry makes it possible for the AI to consider atomic features in the output models since AI acts as a bridge between the experimental data and computational chemistry to develop models that can use experimental data and atomic properties. This hybrid method can be used in materials discovery of the membranes for ion extraction to investigate capabilities, challenges, and future perspectives of the AI-based materials discovery, which can pave the path for ISMs engineering.

Optimising the Flux Enhancer Dosing Strategy in a Pilot-Scale Anaerobic Membrane Bioreactor by Mathematical Modelling

Flux enhancers (FEs) have been successfully applied for fouling mitigation in membrane bioreactors. However, more research is needed to compare and optimise different dosing strategies to improve the filtration performance, while minimising the use of FEs and preventing overdosing. Therefore, the goal of this research is to develop an optimised control strategy for FE dosing into an AnMBR by developing a comprehensive integrated mathematical model. The integrated model includes filtration, flocculation, and biochemical processes to predict the effect of FE dosing on sludge filterability and membrane fouling rate in an AnMBR. The biochemical model was based on an ADM1, modified to include FEs and colloidal material. We developed an empirical model for the FE-induced flocculation of colloidal material. Various alternate filtration models from the literature and our own empirical models were implemented, calibrated, and validated; the best alternatives were selected based on model accuracy and capacity of the model to predict the effect of varying sludge characteristics on the corresponding output, that is fouling rate or sludge filterability. The results showed that fouling rate and sludge filterability were satisfactorily predicted by the selected filtration models. The best integrated model was successfully applied in the simulation environment to compare three feedback and two feedforward control tools to manipulate FE dosing to an AnMBR. The modelling results revealed that the most appropriate control tool was a feedback sludge filterability controller that dosed FEs continuously, referred to as ∆R20_10. Compared to the other control tools, application of the ∆R20_10 controller resulted in a more stable sludge filterability and steady fouling rate, when the AnMBR was subject to specific disturbances. The simulation environment developed in this research was shown to be a useful tool to test strategies for dosing flux enhancer into AnMBRs.

Review of New Approaches for Fouling Mitigation in Membrane Separation Processes in Water Treatment Applications

This review investigates antifouling agents used in the process of membrane separation (MS), in reverse osmosis (RO), ultrafiltration (UF), nanofiltration (NF), microfiltration (MF), membrane distillation (MD), and membrane bioreactors (MBR), and clarifies the fouling mechanism. Membrane fouling is an incomplete substance formed on the membrane surface, which will quickly reduce the permeation flux and damage the membrane. Foulant is colloidal matter: organic matter (humic acid, protein, carbohydrate, nano/microplastics), inorganic matter (clay such as potassium montmorillonite, silica salt, metal oxide, etc.), and biological matter (viruses, bacteria and microorganisms adhering to the surface of the membrane in the case of nutrients) The stability and performance of the tested nanometric membranes, as well as the mitigation of pollution assisted by electricity and the cleaning and repair of membranes, are reported. Physical, chemical, physico-chemical, and biological methods for cleaning membranes. Biologically induced biofilm dispersion effectively controls fouling. Dynamic changes in membrane foulants during long-term operation are critical to the development and implementation of fouling control methods. Membrane fouling control strategies show that improving membrane performance is not only the end goal, but new ideas and new technologies for membrane cleaning and repair need to be explored and developed in order to develop future applications.

Effect of Operating Conditions on Membrane Fouling in Pilot-Scale MBRs; Filaments Growth, Diminishing Dissolved Oxygen and Recirculation Rate of the Activated Sludge

This is the first study that examines the effect of operating conditions on fouling of Membrane Bio-Reactors (MBRs), which treat municipal wastewater in field conditions, with specific regard to the controlled development of filamentous microorganisms (or filaments). The novelty of the present work is extended to minimize the dissolved oxygen (DO) in recirculated activated sludge for improving the process of denitrification. For this purpose, two pilot-scale MBRs were constructed and operated in parallel: (i) Filament-MBR, where an attempt was made to regulate the growth of filaments by adjustment of DO, the Food-to-Microorganisms (F/M) ratio and temperature, and (ii) Control-MBR, where a gentle stirring tank was employed for the purpose of zeroing the DO in the recycled sludge. Results showed that low temperature (<15 °C) slightly increased the number of filaments in the Filament-MBR which, in turn, decreased the Trans-Membrane Pressure (TMP). As the Soluble Microbial Products (SMP) and the colloids are considered to be the basic foulants of membranes in MBR systems, specific attention was directed to keep their concentration at low values in the mixed liquor. The low F/M ratio in the aeration tanks which preceded the membrane tank was achieved to keep the SMP proteins and carbohydrates at very low values in the mixed liquor, i.e., less than 6 mg/L. Moreover, as a result of the low recirculation rate (2.6∙Q_in), good aggregation of the produced excess sludge was achieved, and low concentration of colloids with a size ≤50 nm (nearly the membranes’ pore size used for filtration/separation) was measured, accounted for maximum 15% of the total colloids. Additionally, the increase in filamentous population at the Filament-MBR contributed to the further reduction of colloids in the mixed liquor at 7.9%, contributing beneficially to the reduction of TMP and of membrane fouling. The diminishing of DO in the recirculated sludge improved denitrification, and resulted in lower concentrations of Ν-NO₃⁻ and TN in the effluent of the Control-MBR. Furthermore, the recirculation rate of Q_r = 2.6∙Q_in, in comparison with Q_r = 4.3∙Q_in, resulted in improved performance regarding the removal of N-NH₄⁺. Finally, high organics removal and ammonium nitrification was observed in the effluent of both pilots, since COD and Ν-ΝH₄⁺ concentrations were generally in the range of 10–25 mg/L and <0.1 mg/L, respectively.