A machine learning framework to improve effluent quality control in wastewater treatment plants

Predicting the Occurrence of Substituted and Unsubstituted, Polycyclic Aromatic Compounds in Coking Wastewater Treatment Plant Effluent using Machine Learning Regression

Organic contaminants such as polycyclic aromatic compounds (PACs) occurring in industrial effluents can not only persist in wastewater but transform into more toxic and mobile, substituted heterocyclic products during treatment. Thus, predicting the occurrence of PACs and their heterocyclic derivatives (HPACs) in coking wastewater is of utmost importance to reduce the environmental risks in water bodies that receive industrial effluents. Although HPACs can be monitored through sampling and analysis, the characterisation techniques used in their analyses are costly and time-consuming. In this study, we propose 3 distinct kernel-based machine learning (ML) models for predicting PACs including substituted HPACs and alkylated PACs occurring in coking wastewater. By using routinely measured wastewater quality data as input for our models, we predicted the occurrence of 14 HPACs in the final effluent of a coking wastewater treatment plant. Support Vector Machine based regression model (SVR) used for HPAC prediction showed the highest R2 of 0.83. Performance assessment of SVR model showed a mean absolute logarithmic error (MALE) of 0.46 and root mean square error (RMSE) of 0.073 ng/L. Comparatively, K-Nearest Neighbor and Random Forest models showed lower R2 of 0.75 and 0.76 respectively for HPAC prediction. Feature analysis attributed the superior predictability of SVR model likely to its higher weightage (81%) towards dissolved organic carbon and total ammonia as input variables. Both these variables could capture the underlying secondary PAC transformations likely occurring in the treatment plant. Partial dependence plots predicted that ammonia levels higher than 120 mg/L and DOC levels of 50-60 mg/L were likely linked to higher HPACs occurring in the final effluent. This work highlights the capability of kernel-based ML models in capturing nonlinear wastewater chemistry and offers a tool for monitoring trace organic contaminants released in coking effluents.

Using machine learning to identify environmental factors that collectively determine microbial community structure of activated sludge

Activated sludge (AS) microbial communities are influenced by various environmental variables. However, a comprehensive analysis of how these variables jointly and nonlinearly shape the AS microbial community remains challenging. In this study, we employed advanced machine learning techniques to elucidate the collective effects of environmental variables on the structure and function of AS microbial communities. Applying Dirichlet multinomial mixtures analysis to 311 global AS samples, we identified four distinct microbial community types (AS-types), each characterized by unique microbial compositions and metabolic profiles. We used 14 classical linear and nonlinear machine learning methods to select a baseline model. The extremely randomized trees demonstrated optimal performance in learning the relationship between environmental factors and AS types (with an accuracy of 71.43%). Feature selection identified critical environmental factors and their importance rankings, including latitude (Lat), longitude (Long), precipitation during sampling (Precip), solids retention time (SRT), effluent total nitrogen (Effluent TN), average temperature during sampling month (Avg Temp), mixed liquor temperature (Mixed Temp), influent biochemical oxygen demand (Influent BOD), and annual precipitation (Annual Precip). Significantly, Lat, Long, Precip, Avg Temp, and Annual Precip, influenced metabolic variations among AS types. These findings emphasize the pivotal role of environmental variables in shaping microbial community structures and enhancing metabolic pathways within activated sludge. Our study encourages the application of machine learning techniques to design artificial activated sludge microbial communities for specific environmental purposes.

Genetic programming expressions for effluent quality prediction: Towards AI-driven monitoring and management of wastewater treatment plants

Continuous effluent quality prediction in wastewater treatment processes is crucial to proactively reduce the risks to the environment and human health. However, wastewater treatment is an extremely complex process controlled by several uncertain, interdependent, and sometimes poorly characterized physico-chemical-biological process parameters. In addition, there are substantial spatiotemporal variations, uncertainties, and high non-linear interactions among the water quality parameters and process variables involved in the treatment process. Such complexities hinder efficient monitoring, operation, and management of wastewater treatment plants under normal and abnormal conditions. Typical mathematical and statistical tools most often fail to capture such complex interrelationships, and therefore data-driven techniques offer an attractive solution to effectively quantify the performance of wastewater treatment plants. Although several previous studies focused on applying regression-based data-driven models (e.g., artificial neural network) to predict some wastewater treatment effluent parameters, most of these studies employed a limited number of input variables to predict only one or two parameters characterizing the effluent quality (e.g., chemical oxygen demand (COD) and/or suspended solids (SS)). Harnessing the power of Artificial Intelligence (AI), the current study proposes multi-gene genetic programming (MGGP)-based models, using a dataset obtained from an operational wastewater treatment plant, deploying membrane aerated biofilm reactor, to predict the filtrated COD, ammonia (NH4), and SS concentrations along with the carbon-to-nitrogen ratio (C/N) within the effluent. Input features included a set of process variables characterizing the influent quality (e.g., filtered COD, NH4, and SS concentrations), water physics and chemistry parameters (e.g., temperature and pH), and operation conditions (e.g., applied air pressure). The developed MGGP-based models accurately reproduced the observations of the four output variables with correlation coefficient values that ranged between 0.98 and 0.99 during training and between 0.96 and 0.99 during testing, reflecting the power of the developed models in predicting the quality of the effluent from the treatment system. Interpretability analyses were subsequently deployed to confirm the intuitive understanding of input-output interrelations and to identify the governing parameters of the treatment process. The developed MGGP-based models can facilitate the AI-driven monitoring and management of wastewater treatment plants through devising optimal rapid operation and control schemes and assisting the plants' operators in maintaining proper performance of the plants under various normal and disruptive operational conditions.

Artificial intelligence and machine learning algorithms in the detection of heavy metals in water and wastewater: Methodological and ethical challenges

Heavy metals (HMs) enter waterbodies through various means, which, when exceeding a threshold limit, cause toxic effects both on the environment and in humans upon entering their systems. Recent times have seen an increase in such HM influx incident rates. This requires an instant response in this regard to review the challenges in the available classical methods for HM detection and removal. As well as provide an opportunity to explore the applications of artificial intelligence (AI) and machine learning (ML) for the identification and further redemption of water and wastewater from the HMs. This review of research focuses on such applications in conjunction with the available in-silico models producing worldwide data for HM levels. Furthermore, the effect of HMs on various disease progressions has been provided, along with a brief account of prediction models analysing the health impact of HM intoxication. Also discussing the ethical and other challenges associated with the use of AI and ML in this field is the futuristic approach intended to follow, opening a wide scope of possibilities for improvement in wastewater treatment methodologies.

Predicting effluent quality parameters for wastewater treatment plant: A machine learning-based methodology

Wastewater Treatment Plants (WWTPs) present complex biochemical processes of high variability and difficult prediction. This study presents an innovative approach using Machine Learning (ML) models to predict wastewater quality parameters. In particular, the models are applied to datasets from both a simulated wastewater treatment plant (WWTP), using DHI WEST software (WEST WWTP), and a real-world WWTP database from Santa Catarina Brewery AMBEV, located in Lages/SC - Brazil (AMBEV WWTP). A distinctive aspect is the evaluation of predictive performance in continuous data scenarios and the impact of changes in WWTP operations on predictive model performance, including changes in plant layout. For both plants, three different scenarios were addressed, and the quality of predictions by random forest (RF), support vector machine (SVM), and multilayer perceptron (MLP) models were evaluated. The prediction quality by the MLP model reached an R2 of 0.72 for TN prediction in the WEST WWTP output, and the RF model better adapted to the real data of the AMBEV WWTP, despite the significant discrepancy observed between the real and the predicted data. Techniques such as Partial Dependence Plots (PDP) and Permutation Importance (PI) were used to assess the importance of features, particularly in the simulated WEST tool scenario, showing a strong correlation of prediction results with influent parameters related to nitrogen content. The results of this study highlight the importance of collecting and storing high-quality data and the need for information on changes in WWTP operation for predictive model performance. These contributions advance the understanding of predictive modeling for wastewater quality and provide valuable insights for future practice in wastewater treatment.

Tackling data challenges in forecasting effluent characteristics of wastewater treatment plants

In wastewater treatment plants (WWTPs), the stochastic nature of influent wastewater and operational and weather conditions cause fluctuations in effluent quality. Data-driven models can forecast effluent quality a few hours ahead as a response to the influent characteristics, providing enough time to adjust system operations and avoid undesired consequences. However, existing data for training models are often incomplete and contain missing values. On the other hand, collecting additional data by installing new sensors is costly. The trade-off between using existing incomplete data and collecting costly new data results in three data challenges faced when developing data-driven WWTP effluent forecasters. These challenges are to determine important variables to be measured, the minimum number of required data instances, and the maximum percentage of tolerable missing values that do not impede the development of an accurate model. As these issues are not discussed in previous studies, in this research, for the first time, a comprehensive analysis is done to provide answers to these challenges. Another issue that arises in all data-driven modeling is how to select an appropriate forecasting model. This paper addresses these issues by first testing nine machine learning models on data collected from three wastewater treatment plants located in Iran, Australia, and Spain. The most accurate forecaster, Bayesian network, was then used to address the articulated challenges. Key variables in forecasting effluent characteristics were flow rate, total suspended solids, electrical conductivity, phosphorus compounds, wastewater temperature, and air temperature. A minimum of 250 samples was needed during the model training to achieve a great reduction in the forecasting error. Moreover, a steep increase in the error was observed should the portion of missing values exceed 10%. The results assist plant managers in estimating the necessary data collection effort to obtain an accurate forecaster, contributing to the quality of the effluent.

A hybrid deep learning approach to improve real-time effluent quality prediction in wastewater treatment plant

Wastewater treatment plant (WWTP) operation is usually intricate due to large variations in influent characteristics and nonlinear sewage treatment processes. Effective modeling of WWTP effluent water quality can provide valuable decision-making support to facilitate their operations and management. In this study, we developed a novel hybrid deep learning model by combining the temporal convolutional network (TCN) model with the long short-term memory (LSTM) network model to improve the simulation of hourly total nitrogen (TN) concentration in WWTP effluent. The developed model was tested in a WWTP in Jiangsu Province, China, where the prediction results of the hybrid TCN-LSTM model were compared with those of single deep learning models (TCN and LSTM) and traditional machine learning model (feedforward neural network, FFNN). The hybrid TCN-LSTM model could achieve 33.1 % higher accuracy as compared to the single TCN or LSTM model, and its performance could improve by 63.6 % comparing to the traditional FFNN model. The developed hybrid model also exhibited a higher power prediction of WWTP effluent TN for the next multiple time steps within eight hours, as compared to the standalone TCN, LSTM, and FFNN models. Finally, employing model interpretation approach of Shapley additive explanation to identify the key parameters influencing the behavior of WWTP effluent water quality, it was found that removing variables that did not contribute to the model output could further improve modeling efficiency while optimizing monitoring and management strategies.

Enhancing effluent quality prediction in wastewater treatment plants through the integration of factor analysis and machine learning

Precisely predicting the concentration of nitrogen-based pollutants from the wastewater treatment plants (WWTPs) remains a challenging yet crucial task for optimizing operational adjustments in WWTPs. In this study, an integrated approach using factor analysis (FA) and machine learning (ML) models was employed to accurately predict effluent total nitrogen (Ntoteff) and nitrate nitrogen (NO3-Neff) concentrations of the WWTP. The input values for the ML models were honed through FA to optimize factors, thereby significantly enhancing the ML prediction accuracy. The prediction model achieved a highest coefficient of determination (R2) of 97.43 % (Ntoteff) and 99.38 % (NO3-Neff), demonstrating satisfactory generalization ability for predictions up to three days ahead (R2 >80 %). Moreover, the interpretability analysis identified that the denitrification factor, the pollutant load factor, and the meteorological factor were significant. The model framework proposed in this study provides a valuable reference for optimizing the operation and management of wastewater treatment.

Effluent parameters prediction of a biological nutrient removal (BNR) process using different machine learning methods: A case study

This paper proposes a novel targeted blend of machine learning (ML) based approaches for controlling wastewater treatment plant (WWTP) operation by predicting distributions of key effluent parameters of a biological nutrient removal (BNR) process. Two years of data were collected from Plajyolu wastewater treatment plant in Kocaeli, Türkiye and the effluent parameters were predicted using six machine learning algorithms to compare their performances. Based on mean absolute percentage error (MAPE) metric only, support vector regression machine (SVRM) with linear kernel method showed a good agreement for COD and BOD5, with the MAPE values of about 9% and 0.9%, respectively. Random Forest (RF) and EXtreme Gradient Boosting (XGBoost) regression were found to be the best algorithms for TN and TP effluent parameters, with the MAPE values of about 34% and 27%, respectively. Further, when the results were evaluated together according to all the performance metrics, RF, SVRM (with both linear kernel and RBF kernel), and Hybrid Regression algorithms generally made more successful predictions than Light GBM and XGBoost algorithms for all the parameters. Through this case study we demonstrated selective application of ML algorithms can be used to predict different effluent parameters more effectively. Wider implementation of this approach can potentially reduce the resource demands for active monitoring the environmental performance of WWTPs.

Machine learning framework for modeling flocculation kinetics using non-intrusive dynamic image analysis

The implementation of a machine learning (ML) model to improve both the effectiveness and sustainability of the water treatment system is a significant challenge in the water sector, with the optimization of flocculation processes being a major setback. The objective of this study was to develop a ML model for predicting flocs evolution of the flocculation process in water treatment. Furthermore, we have devised a framework for its potential adoption in large-scale water treatment. Therefore, the paper can be split into two parts. In the first one, flocculation evolution has been studied from an experimental setup, using a non-intrusive image acquisition method. Subsequently, the ML framework has been implemented. Batch assay data of two velocity gradients (Gf 20 and 60 s-1) and flocculation time of three hours were partitioned into five groups for flocs length range 0.27-3.5 mm and upscaled using linear method. Multilayer Perceptron (MLP) and Long-Short Term Memory (LSTM) models, and traditional time series model, Auto Regressive Integrated Moving Average (ARIMA) were explored to predict floc length evolution data. The experiments illustrate the kinetics of flocculation, where the initial stage is characterized by a rapid floc growth followed by a plateau during which floc length fluctuates within a narrow range. Results demonstrate that ML is sensitive to flocculation; however, the model should be selected with care. ARIMA model is not suitable for predicting number of flocs with negative test accuracy (R2). In contrast, MLP recorded R2 of 0.86-1.0 for training and 0.92-1.0 for testing, across Gf 20 s-1 and Gf 60 s-1. LSTM model has the best prediction R2 of 0.92-1.00 for Gf 20 s-1 and accurately predicts the number of flocs across all groups and Gfs. Our study has proven that the developed framework could be replicated for water treatment modeling and promotes the application of smart technology in water treatment.

Estimating ammonium changes in pilot and full-scale constructed wetlands using kinetic model, linear regression, and machine learning

Constructed wetlands (CWs) are a widely utilized nature-based wastewater treatment method for various effluents. However, their application has been more focused on pilot and full-scale CWs with substantial surface areas and extended operation times, which hold greater relevance in practical scenarios. This study used kinetics, linear regression (LR), and machine learning (ML) models to estimate effluent ammonium in pilot and full-scale CWs. From screening 1476 papers, 24 pilot and full-scale CW studies were selected to extract data containing 15 features and 975 data points. Nine models were fit to this data, revealing that linear models were less effective in capturing CW effluent compared to nonlinear ML algorithms. For training data, the Monod kinetic model predicted the poorest performance with an RMSE of 41.84 mg/L and R2 of 0.34, followed by simple LR (RMSE 24.29 mg/L and R2 0.77) and multiple LR (RMSE 22.63 mg/L and R2 0.80). In contrast, Cubist and Random Forest achieved high performances, with an average RMSE of 12.01 ± 5.38 and an average R2 of 0.93 ± 0.07 for Cubist, and an average RMSE of 15.94 ± 10.69 and an average R2 of 0.91 ± 0.08 for RF. The trained Random Forest performed the best for new data, with an R2 of 0.93 and RMSE of 13.48 mg/L. This ML-based model is a valuable tool for efficiently estimating effluent ammonium concentration in pilot and full-scale CWs, thereby facilitating the design of systems.

Prediction of biological nutrients removal in full-scale wastewater treatment plants using H2O automated machine learning and back propagation artificial neural network model: Optimization and comparison

The effective control of total nitrogen (ETN) and total phosphorus (ETP) in effluent is challenging for wastewater treatment plants (WWTPs). In this work, automated machine learning (AutoML) (mean square error = 0.4200 ∼ 3.8245, R2 = 0.5699 ∼ 0.6219) and back propagation artificial neural network (BPANN) model (mean square error = 0.0012 ∼ 6.9067, R2 = 0.4326 ∼ 0.8908) were used to predict and analyze biological nutrients removal in full-scale WWTPs. Interestingly, BPANN model presented high prediction performance and general applicability for WWTPs with different biological treatment units. However, the AutoML candidate models were more interpretable, and the results showed that electricity carbon emission dominated the prediction. Meanwhile, increasing data volume and types of WWTP hardly affected the interpretable results, demonstrating its wide applicability. This study demonstrated the validity and the specific advantages of predicting ETN and ETP using H2O AutoML and BPANN model, which provided guidance on the prediction and improvement of biological nutrients removal in WWTPs.

Physics-informed neural network-based serial hybrid model capturing the hidden kinetics for sulfur-driven autotrophic denitrification process

Sulfur-driven autotrophic denitrification (SdAD) is a biological process that can remove nitrate from low carbon/nitrogen (C/N) ratio wastewater. Although this process has been intensively researched, the mechanism whereby its intermediates (i.e., elemental sulfur and nitrite ions) are generated and accumulated remains elusive. Existing mathematical models developed for SdAD cannot accurately predict the intermediates in SdAD because of the incomplete knowledge of process kinetic resulting from changes in the environmental conditions and electron competition during SdAD. To address this limitation, we proposed a novel serial hybrid model structure based on a physics-informed neural network (PINN) to capture the dynamics of the process kinetics and predict the substrate concentrations in SdAD. In this study, we evaluated the model through numerical experiments and applied it to real case studies involving batch and continuous-flow reactor scenarios. By leveraging the PINN approach, the hybrid model yielded accurate predictions at both the state (i.e. substrate concentration) and kinetic levels in the numerical experiments and performed better than both mechanistic and purely data-driven models in the case studies. Furthermore, we used the trained hybrid model to design control strategies for SdAD and a novel integrated process involving SdAD and anammox for energy-efficient nitrogen removal. Finally, we discuss the advantages and application scope of the PINN-based hybrid model.

Artificial neural networks for performance prediction of full-scale wastewater treatment plants: a systematic review

Wastewater treatment plants (WWTPs) are complex systems that must maintain high levels of performance to achieve adequate effluent quality to protect the environment and public health. Artificial intelligence and machine learning methods have gained attention in recent years for modeling complex problems, such as wastewater treatment. Although artificial neural networks (ANNs) have been identified as the most common of these methods, no study has investigated the development and configuration of these models. We conducted a systematic literature review on the use of ANNs to predict the effluent quality and removal efficiencies of full-scale WWTPs. Three databases were searched, and 44 records of the 667 identified were selected based on the eligibility criteria. The data extracted from the papers showed that the majority of studies used the feedforward neural network model with a backpropagation training algorithm to predict the effluent quality of plants, particularly in terms of organic matter indicators. The findings of this research may help in the search for an optimum design modeling process for future studies of similar prediction problems.

Data mining from process monitoring of typical polluting enterprise

With the increasing volume of environmental monitoring data, extracting valuable insights from multivariate time series sensor data can facilitate comprehensive information utilization and support informed decision-making in environmental management. However, there is a dearth of comprehensive research on multivariate data analysis for process monitoring in typical polluting enterprises. In this study, an artificial neural network model based on back-propagation algorithm (BP-ANN) was developed to predict the wastewater and exhaust gas emissions using IoT data obtained from process monitoring of a typical polluting enterprise located in Taizhou, Zhejiang Province, China. The results indicate that the model constructed has a high predictive coefficient of determination (R2) with values of 0.8510, 0.9565, 0.9561, 0.9677, and 0.9061 for chemical oxygen demand (COD), potential of hydrogen (pH), electrical conductivity (EC), flue gas emission (FGE), and non-methane hydrocarbon concentration (NMHC) respectively. For the first time, the variable importance measure (VIM)-assisted BP-ANN was employed to investigate the internal and external correlations between wastewater and exhaust gas treatment, thereby enhancing the interpretability of mapping features in the BP-ANN model. The predicted errors for pH and FGE have been demonstrated to fall within the range of - 0.62 ~ 0.30 and - 0.21 ~ 0.15 m3/s, respectively, with average relative errors of 1.05% and 9.60%, which is advantageous in detecting anomalous data and forecasting pollution indicator values. Our approach successfully addresses the challenge of segregating data analysis for wastewater disposal and exhaust gas disposal in the process monitoring of polluting enterprises, while also unearthing potential variables that significantly contribute to the BP-ANN model, thereby facilitating the selection and extraction of characteristic variables.

Quantitative prediction of the hydraulic performance of free water surface constructed wetlands by integrating numerical simulation and machine learning

Quantitative prediction of the design parameter-influenced hydraulic performance is significant for optimizing free water surface constructed wetlands (FWS CWs) to reduce point and non-point source pollution and improve land utilization. However, owing to limitations of the test conditions and data scale, a quantitative prediction model of the hydraulic performance under multiple design parameters has not yet been established. In this study, we integrated field test data, mechanism model, statistical regression, and machine learning (ML) to construct such quantitative prediction models. A FWS CW numerical model was established by integrating 13 groups of trace data from field tests. Subsequently, training, test and extension datasets comprising 125 (5^3), 25 (L₂₅(5⁶)) and 16 (L₁₆(4⁴)) data points, respectively, were generated via numerical simulation of multi-level value combination of three quantitative design parameters, namely, water depth, hydraulic loading rate (HLR), and aspect ratio. The short circuit index (φ₁₀), Morrill dispersion index (MDI), hydraulic efficiency (λ) and moment index (MI) were used as representative hydraulic performance indicators. Training set with large samples were analyzed to determine the variation rules of different hydraulic indicators. Based on the control variable method, φ₁₀, λ, and MI grew exponentially with increasing aspect ratio whereas MDI showed a decreasing trend; with increasing water depth, φ₁₀, λ, and MI showed polynomial decreases whereas MDI increased; with increasing HLR, φ₁₀, λ, and MI slowly increased linearly whereas MDI showed the opposite trend. Finally, we constructed models based on multivariate nonlinear regression (MNLR) and ML (random forest (RF), multilayer perceptron (MLP), and support vector regression. The coefficients of determination (R²) of the MNLR and ML models fitting the training and test sets were all greater than 0.9; however, the generalization abilities of different models in the extension set were different. The most robust MLP, MNLR without interaction term, and RF models were recommended as the preferred models to hydraulic performance prediction. The extreme importance of aspect ratio in hydraulic performance was revealed. Thus, gaps in the current understanding of multivariate quantitative prediction of the hydraulic performance of FWS CWs are addressed while providing an avenue for researching FWS CWs in different regions according to local conditions.

Opportunities and challenges of machine learning in bioprocesses: Categorization from different perspectives and future direction

Recent advances in machine learning (ML) have revolutionized an extensive range of research and industry fields by successfully addressing intricate problems that cannot be resolved with conventional approaches. However, low interpretability and incompatibility make it challenging to apply ML to complicated bioprocesses, which rely on the delicate metabolic interplay among living cells. This overview attempts to delineate ML applications to bioprocess from different perspectives, and their inherent limitations (i.e., uncertainties in prediction) were then discussed with unique attempts to supplement the ML models. A clear classification can be made depending on the purpose of the ML (supervised vs unsupervised) per application, as well as on their system boundaries (engineered vs natural). Although a limited number of hybrid approaches with meaningful outcomes (e.g., improved accuracy) are available, there is still a need to further enhance the interpretability, compatibility, and user-friendliness of ML models.

Artificial intelligence and machine learning-based monitoring and design of biological wastewater treatment systems

Artificial intelligence (AI) and machine learning (ML) are currently used in several areas. The applications of AI and ML based models are also reported for monitoring and design of biological wastewater treatment systems (WWTS). The available information is reviewed and presented in terms of bibliometric analysis, model's description, specific applications, and major findings for investigated WWTS. Among the applied models, artificial neural network (ANN), fuzzy logic (FL) algorithms, random forest (RF), and long short-term memory (LSTM) were predominantly used in the biological wastewater treatment. These models are tested by predictive control of effluent parameters such as biological oxygen demand (BOD), chemical oxygen demand (COD), nutrient parameters, solids, and metallic substances. Following model performance indicators were mainly used for the accuracy analysis in most of the studies: root mean squared error (RMSE), mean square error (MSE), and determination coefficient (DC). Besides, outcomes of various models are also summarized in this study.

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.

Automated machine learning-based prediction of microplastics induced impacts on methane production in anaerobic digestion

Microplastics as emerging pollutants have been heavily accumulated in the waste activated sludge (WAS) during biological wastewater treatment, which showed significantly diverse impacts on the subsequent anaerobic sludge digestion for methane production. However, a robust modeling approach for predicting and unveiling the complex effects of accumulated microplastics within WAS on methane production is still missing. In this study, four automated machine learning (AutoML) approach was applied to model the effects of microplastics on anaerobic digestion processes, and integrated explainable analysis was explored to reveal the relationships between key variables (e.g., concentration, type, and size of microplastics) and methane production. The results showed that the gradient boosting machine had better prediction performance (mean squared error (MSE) = 17.0) than common neural networks models (MSE = 58.0), demonstrating that the AutoML algorithms succeeded in predicting the methane production and could select the best machine learning model without human intervention. Explainable analysis results indicated that the variable of microplastic types was more important than the variable of microplastic diameter and concentration. The existence of polystyrene was associated with higher methane production, whereas increasing microplastic diameter and concentration both inhibited methane production. This work also provided a novel modeling approach for comprehensively understanding the complex effects of microplastics on methane production, which revealed the dependence relationships between methane production and key variables and may be served as a reference for optimizing operational adjustments in anaerobic digestion processes.

Applying machine learning to construct braking emission model for real-world road driving

Brake emissions from vehicles are increasing as the number of vehicles increases. However, current research on brake emissions, particularly the intensity and characteristics of emissions under real road conditions, is significantly inadequate compared to exhaust emissions. To this end, a dataset of 600 (200 unique real-world braking events simulated using three types of brake pads) real-world braking events (called brake pad segments) was constructed and a mapping function between the average brake emission intensity of PM2.5 from the segments and the segment features was established by five algorithms (multiple linear regression (MLR) and four machine learning algorithms). Based on the five algorithms, the importance of the different features of the fragments was discussed and brake energy intensity (BEI) and metal content (MC) of the brake pad emissions were identified as the most significant factors affecting brake emissions and used as the final modeling features. Among the five algorithms, categorical boosting (CatBoost) had the best prediction performance, with a mean R2 and RMSE of 0.83 and 0.039 respectively for the tenfold cross-validation. In addition, the CatBoost-based model was further compared with the MOVES model to demonstrate its applicability. The CatBoost-based model has better prediction performance than the MOVES model. The MOVES model overpredicts brake fragment emissions for urban roads and underpredicts brake fragment emissions for motorways. Furthermore, the CatBoost-based model was interpreted and visualized by an individual conditional expectation (ICE) plot to break the machine learning "black box", with BEI and MC showing nonlinear monotonic increasing relationships with braking emissions. ICE plot also provides viable technical solutions for controlling brake emissions in the future. Both avoiding aggressive braking driving behavior (e.g., the application of smart transportation technologies) and using brake pads with less metal content (e.g., using ceramic brake pads) can effectively reduce brake emissions. The construction of a machine learning-based brake emission model and the white-boxing of its model provide excellent insights for the future detailed assessment and control of brake emissions.

Disentangling the Factors That Contribute to the Growth of Betula spp. and Cunninghami lanceolata in China Based on Machine Learning Algorithms

Forests are indispensable materials and spiritual foundations for promoting ecosystem circulation and human survival. Exploring the environmental impact mechanism on individual-tree growth is of great significance. In this study, the effects of biogeoclimate, competition, and topography on the growth of Betula spp. and Cunninghamia lanceolata (Lamb.) Hook., two tree species with high importance value in China, were explored by gradient boosting regression tree (GBRT), k-nearest neighbor (KNN), and random forest (RF) machine learning (ML) algorithms. The results showed that the accuracy of RF was better than KNN, which was better than GBRT. All ML algorithms performed well for future diameter at breast height (DBH) predictions; the Willmott’s indexes of agreement (WIA) of each ML algorithm in predicting the future DBH were all higher than 0.97, and the R2 was higher than 0.98 and 0.90, respectively. The individual tree annual growth rate is mainly affected by the single-tree size, and the external environment can promote or inhibit tree growth. Climate and stand structure variables were relatively more important for tree growth than the topographic factors. Lower temperature and precipitation, higher stand density, and canopy closure were more unfavorable for their growth. In afforestation, the following factors should be considered in order: geographic location, meteorological climate, stand structure, and topography.

A review of the application of machine learning in water quality evaluation

With the rapid increase in the volume of data on the aquatic environment, machine learning has become an important tool for data analysis, classification, and prediction. Unlike traditional models used in water-related research, data-driven models based on machine learning can efficiently solve more complex nonlinear problems. In water environment research, models and conclusions derived from machine learning have been applied to the construction, monitoring, simulation, evaluation, and optimization of various water treatment and management systems. Additionally, machine learning can provide solutions for water pollution control, water quality improvement, and watershed ecosystem security management. In this review, we describe the cases in which machine learning algorithms have been applied to evaluate the water quality in different water environments, such as surface water, groundwater, drinking water, sewage, and seawater. Furthermore, we propose possible future applications of machine learning approaches to water environments.

Real-Time Optimization of Wastewater Treatment Plants via Constraint Adaptation

An important requirement of wastewater treatment plants (WWTPs) is compliance with the local regulations on effluent discharge, which are going to become more stringent in the future. The operation of WWTPs exhibits a trade-off between operational cost and effluent quality, which provides a scope for optimization. Process optimization is usually done by optimizing a model of the process. However, due to inevitable plant–model mismatch, the computed optimal solution is usually not optimal for the plant. This study represents the first attempt to handle plant–model mismatch via constraint adaptation (CA) for the real-time optimization of WWTPs. In this simulation study, the “plant” is a model adopted from the BSM1 benchmark, while a reduced-order “model” is used for making predictions and computing the optimal inputs. A first implementation uses steady-state measurements of the plant constraints to adjust the model in the optimization framework. A fast CA technique is also proposed, which adjusts the model using transient measurements. It is observed that, even in the presence of significant plant–model mismatch, the two proposed techniques are able to meet the active plant constraints. These techniques are found to reduce the pumping and aeration energy by 20%, as compared to that adopted in BSM1.

Machine learning exhibited excellent advantages in the performance simulation and prediction of free water surface constructed wetlands

Optimizing the design and operation parameters of free water surface constructed wetlands (FWS CWs) in runoff regulation and wastewater treatment is necessary to improve the comprehensive performance. In this study, nine machine learning (ML) algorithms were successfully developed to optimize the parameter combinations for FWS CWs. The scale effect of surface area on wetland performance was determined based on consistently smaller predictions (-6.2% to -28.9%) of the nine well-established ML algorithms. The models most suitable for FWS CW performance simulation and prediction were random forest and extra trees algorithms because of their high R² values (0.818 in both) with the training set and low mean absolute relative errors (4.7% and 3.8%, respectively) with the test set. Results from feature analysis of the six tree-based algorithms emphasized the importance of water depth and layout of inlet and outlet, and revealed the negligible effect of the aspect ratio. Feature importance and partial dependence analysis enhanced the interpretability of the tree-based algorithms. The proposed ML algorithms enabled the implementation of an extended scenario at a low cost in real time. Therefore, ML algorithms are suitable for expressing the complex and uncertain effects of the design and operation parameters on the performance of FWS CWs. Acquiring datasets consisting of more extensive, uniform, and unbiased parameter combinations is crucial for developing more robust and practical ML algorithms for the optimal design of FWS CWs.

Machine Learning Approach to Predict Quality Parameters for Bacterial Consortium-Treated Hospital Wastewater and Phytotoxicity Assessment on Radish, Cauliflower, Hot Pepper, Rice and Wheat Crops

Raw hospital wastewater is a source of excessive heavy metals and pharmaceutical pollutants. In water-stressed countries such as Pakistan, the practice of unsafe reuse by local farmers for crop irrigation is of major concern. In our previous work, we developed a low-cost bacterial consortium wastewater treatment method. Here, in a two-part study, we first aimed to find what physico-chemical parameters were the most important for differentiating consortium-treated and untreated wastewater for its safe reuse. This was achieved using a Kruskal–Wallis test on a suite of physico-chemical measurements to find those parameters which were differentially abundant between consortium-treated and untreated wastewater. The differentially abundant parameters were then input to a Random Forest classifier. The classifier showed that ‘turbidity’ was the most influential parameter for predicting biotreatment. In the second part of our study, we wanted to know if the consortium-treated wastewater was safe for crop irrigation. We therefore carried out a plant growth experiment using a range of popular crop plants in Pakistan (Radish, Cauliflower, Hot pepper, Rice and Wheat), which were grown using irrigation from consortium-treated and untreated hospital wastewater at a range of dilutions (turbidity levels) and performed a phytotoxicity assessment. Our results showed an increasing trend in germination indices and a decreasing one in phytotoxicity indices in plants after irrigation with consortium-treated hospital wastewater (at each dilution/turbidity measure). The comparative study of growth between plants showed the following trend: Cauliflower > Radish > Wheat > Rice > Hot pepper. Cauliflower was the most adaptive plant (PI: −0.28, −0.13, −0.16, −0.06) for the treated hospital wastewater, while hot pepper was susceptible for reuse; hence, we conclude that bacterial consortium-treated hospital wastewater is safe for reuse for the irrigation of cauliflower, radish, wheat and rice. We further conclude that turbidity is the most influential parameter for predicting bio-treatment efficiency prior to water reuse. This method, therefore, could represent a low-cost, low-tech and safe means for farmers to grow crops in water stressed areas.

Towards better process management in wastewater treatment plants: Process analytics based on SHAP values for tree-based machine learning methods

Understanding the mechanisms of pollutant removal in Wastewater Treatment Plants (WWTPs) is crucial for controlling effluent quality efficiently. However, the numerous treatment units, operational factors, and the underlying interactions between these units and factors usually obfuscate the comprehensive and precise understanding of the processes. We have previously proposed a machine learning (ML) framework to uncover complex cause-and-effect relationships in WWTPs. However, only one interpretable ML model, Random forest (RF), was studied and the interpretation method was not granular enough to reveal very detailed relationships between operational factors and effluent parameters. Thus, in this paper, we present an upgraded framework involving three interpretable tree-based models (RF, XGboost and LightGBM), three metrics (R², Root mean squared error (RMSE), and Mean absolute error (MAE)) and a more advanced interpretation system SHapley Additive exPlanations (SHAP). Details of the framework are provided along with a demonstration of its practical applicability based on a case study of the Umeå WWTP in Sweden. Results show that, for both labels TSS_e (Total suspended solids in effluent) and PO4_e (Phosphate in effluent), the XGBoost models are optimal whereas the RF models are the least optimal, due to overfitting and polarized fitting. This study has yielded multiple new and significant findings with respect to the control of TSS_e and PO4_e in the Umeå WWTP and other similarly configured WWTPs. Additionally, this study has produced two important generic findings relating to ML applications for WWTPs (or even other process industries) in terms of cause-and-effect investigations. First, the model comparison should be carried out from multiple perspectives to ensure that underlying details are fully revealed and examined. Second, using a precise, robust, and granular (feature attribution available for individual instances) explanation method can bring extra insight into both model comparison and model interpretation. SHAP is recommended as we found it to be of great value in this study.

Application oriented bioaugmentation processes: Mechanism, performance improvement and scale-up

Bioaugmentation is an optimization method with great potential to improve the treatment effect by introducing specific strains into the biological treatment system. In this study, a comprehensive review of the mechanism of bioaugmentation from the aspect of microbial community structure, the optimization methods facilitating application as well as feasible approaches of scale-up application has been provided. The different contribution of indigenous and exogenous strains was critically analyzed, the relationship between microbial community variation and system performance was clarified. Operation regulation and immobilization technologies are effective methods to deal with the possible failure of bioaugmentation. The gradual expansion from lab-scale, pilot scale to full-scale, the transformation and upgrading of wastewater treatment plants through the combination of direct dosing and biofilm, and the application of side-stream reactors are feasible ways to realize the full-scale application. The future challenges and prospects in this field were also proposed.

Comparison of random forest and multiple linear regression to model the mass balance of biosolids from a complex biosolids management area

The use of biosolids as a soil amendment provides an important alternative to disposal and can improve soil health; however, distribution for water resource recovery facilities (WRRFs) in the United States can be challenging due to decreasing cropland, increased precipitation, variable plant operations, and financial constraints. Although statistical modeling is commonly used in the water sector, machine learning is still an emerging tool and can provide insights to optimize operations. Random forest (RF), a machine learning model, and multiple linear regression (MLR) were used in this study to model the mass balance of biosolids from a complex biosolids management area. The RF model outperformed (R²  = 0.89) the MLR model (R²  = 0.49) and showed that rainfall was a major factor impacting distribution. Storage for dried biosolids would help decouple drying operations from wet weather and increase distribution. This study demonstrated how machine learning can assist in decision-making processes for long-term planning at WRRFs. PRACTITIONER POINTS: Random forest predicted the 7-day average mass balance of biosolids from a complex biosolids management area. Decoupling biosolids drying operations from wet weather was identified as the highest operational priority. Machine learning outperformed multiple linear regression and can be an important tool for the water sector.

Optimal Bands Combination Selection for Extracting Garlic Planting Area with Multi-Temporal Sentinel-2 Imagery

Garlic is one of the main economic crops in China. Accurate and timely extraction of the garlic planting area is critical for adjusting the agricultural planting structure and implementing rural policy actions. Crop extraction methods based on remote sensing usually use spectral-temporal features. Still, for garlic extraction, most methods simply combine all multi-temporal images. There has been a lack of research on each band's function in each multi-temporal image and optimal bands combination. To systematically explore the potential of the multi-temporal method for garlic extraction, we obtained a series of Sentinel-2 images in the whole garlic growth cycle. The importance of each band in all these images was ranked by the random forest (RF) method. According to the importance score of each band, eight different multi-temporal combination schemes were designed. The RF classifier was employed to extract garlic planting area, and the accuracy of the eight schemes was compared. The results show that (1) the Scheme VI (the top 39 bands in importance score) achieved the best accuracy of 98.65%, which is 6% higher than the optimal mono-temporal (February, wintering period) result, and (2) the red-edge band and the shortwave-infrared band played an essential role in accurate garlic extraction. This study gives inspiration in selecting the remotely sensed data source, the band, and phenology for accurately extracting garlic planting area, which could be transferred to other sites with larger areas and similar agriculture structures.