A Futuristic Approach to Subsurface-Constructed Wetland Design for the South-East Asian Region Using Machine Learning

ACS ES&T Water, Год журнала: 2024, Номер 4(9), С. 4061 - 4074

Опубликована: Авг. 29, 2024

This study investigates the optimized design of horizontal flow constructed wetlands (HFCWs) to enhance pollutant removal efficiency while minimizing surface area requirements, particularly in Southeast Asian region. By refining first-order rate coefficient (k) for organics and nutrients, research aims meet specific performance benchmarks across three scenarios, ensuring compliance with discharge or reuse standards. Utilizing a data set comprising 1680 entries, five machine learning models─multiple linear regression (MLR), eXtreme Gradient Boosting (XGBoost), random forest (RF), artificial neural network (ANN), support vector (SVR)─were employed predict k values. Pearson's correlation, heat maps, ANOVA analysis identified most influential parameters affecting k-value predictions. The values ranged from 0.01 0.52 per day using P–k–C* method, essential effective removal. SVR model demonstrated highest predictive accuracy, R2 0.91 kBOD, 0.90 kTN, 0.82 kTKN, 0.76 kTP. optimization reduced standard deviations significantly, 136.90% 2.28%. Consequently, required wetland was by up 68% biochemical oxygen demand (BOD), 60% TN (total nitrogen), 67% TP phosphorus) larger systems, supporting tailored HFCWs targeted

Язык: Английский

---

Carbon Emission Characteristics Research of Typical Drinking Water Treatment Plants in South China Based on Machine Learning Models

Опубликована: Янв. 1, 2025

Язык: Английский

---

Effective evaluation of greenhouse gases (GHGs) emissions from anoxic/oxic (A/O) process of regenerated papermaking wastewater treatment through hybrid deep learning techniques: Leveraging the critical role of water quality indicators

Journal of Environmental Management, Год журнала: 2025, Номер 380, С. 125094 - 125094

Опубликована: Апрель 1, 2025

Язык: Английский

---

Predicting and Evaluating Different Pretreatment Methods on Methane Production from Sludge Anaerobic Digestion via Automated Machine Learning with Ensembled Semisupervised Learning

ACS ES&T Engineering, Год журнала: 2023, Номер 4(3), С. 525 - 539

Опубликована: Окт. 19, 2023

Accurate prediction of methane production in anaerobic digestion with various pretreatment strategies is the utmost importance for efficient sludge treatment and resource recovery. Traditional machine learning (ML) algorithms have shown limited accuracy due to challenges optimizing complex parameters scarcity data. This work proposed a novel integrated system that employed an ensemble semisupervised (SSL)-automated ML (AutoML) model variable inputs reveal effects different pretreatments on during explainable analysis. Considering direct correlations type substrates, considered as hidden variable. Results demonstrated AutoML outperformed conventional models (i.e., support vector regression (SVR), extreme gradient boosting (XGB), etc.), evidenced by its higher R2 value. Moreover, integration SSL further enhanced effectively leveraging unlabeled data, leading reduction mean squared error from 11.3 9.7. Explainable analysis results revealed significance variables operating time, followed proteins, carbohydrates, chemical oxygen demand, volatile fatty acids. Furthermore, principal component correlation unveiled interconnected relationships among substrate concentration, microbial communities, metabolic functions found increasing concentration promoted enrichment functional functions. These insights shed light advantages SSL-AutoML predicting systems elucidate dependence key variables, offering valuable guidance effective recovery capabilities.

Язык: Английский

---

The Optuna–LightGBM–XGBoost Model: A Novel Approach for Estimating Carbon Emissions Based on the Electricity–Carbon Nexus

Applied Sciences, Год журнала: 2024, Номер 14(11), С. 4632 - 4632

Опубликована: Май 28, 2024

With the challenge posed by global warming, accurately estimating and managing carbon emissions becomes a key step for businesses, especially power generation companies, to reduce their environmental impact. Optuna–LightGBM–XGBoost, novel emission relationship model that aims improve efficiency of monitoring estimation is proposed in this paper. Deeply exploring intrinsic link between production data emissions, paves new path “measuring through electricity”, contrast factor method commonly used China. Unit from companies are processed into structured tabular data, parallel processing framework constructed with LightGBM XGBoost, optimized Optuna algorithm. The multilayer perceptron (MLP) fuse features enhance prediction accuracy capturing characters individual models cannot detect. Simulation results show Optuna–LightGBM–XGBoost can achieve better performance compared existing methods. mean absolute error (MAE), squared (MSE), percentage (MAPE), coefficient determination (R2) 0.652, 0.939, 0.136, 0.994, respectively. This not only helps governments enterprises develop more scientific reasonable reduction strategies policies, but also lays solid foundation achieving neutrality goals.

Язык: Английский

---

Machine Learning Accelerating the Condition Screening of Ceftriaxone Sodium Anaerobic Co-Metabolic Degradation

ACS ES&T Engineering, Год журнала: 2024, Номер 4(4), С. 947 - 955

Опубликована: Март 19, 2024

Anaerobic co-metabolism is a biotechnological process that improves biodegradation efficiency of refractory organics. By adding cosubstrates, it provides additional carbon sources and energy for the microbial metabolic degradation organic matter. However, large number repeated experiments are required to screen out suitable conditions. It typically lengthy carries significant uncertainty. In this study, machine learning (ML) was used drive screening anaerobic conditions ceftriaxone sodium (CTX) in wastewater treatment. The results showed XGBoost algorithm able effectively predict decomposition with an accuracy up 95%. A Shapley additive explanation (SHAP) analysis temperature, pH, CTX/glucose ratio had greatest impacts on removal CTX, thus highlighting remarkable ability ML accelerate optimal high-throughput proved dominant genera structures presented under two environmental largest difference (temperature, ratio) were significantly different bacterial such as Fastidiosipila, norank_f_Prolixibacteraceae, norank_f_Bacteroidetes_vadinHA17, Georgenia affected hydrolysis acidification process. This work stands by integrating advanced techniques into engineering, thereby enhancing providing richer analytical insights compared traditional methods.

Язык: Английский

---

A new scheme for low-carbon recycling of urban and rural organic waste based on carbon footprint assessment: A case study in China

npj Sustainable Agriculture, Год журнала: 2024, Номер 2(1)

Опубликована: Июнь 7, 2024

Abstract Organic waste treatment is a major driver of global carbon emissions, thus its low-carbon utilization essential yet unclear. Through life cycle assessment organic data from 34 provincial-level regions in China, we have determined that the synergistic and integrated scheme (URIRP) with fertilizer biochar as primary products can reduce annual emissions 6.9 Mt CO 2 e to 2.83 e. This reduction offset 6% electricity industry mainly through sequestration by application biochar-based fertilizer, fossil fuel displacement bio-energy. Moreover, URIRP promote recycling N P, emission air pollutants 866 Mt, increase topsoil matter content 0.25‰ economic efficiency 135%. These findings indicate could realize sustainable management UROSW significant environmental benefits, contribute realization China’s neutrality goal.

Язык: Английский

---

A SHAP machine learning-based study of factors influencing urban residents' electricity consumption - evidence from chinese provincial data

Environment Development and Sustainability, Год журнала: 2024, Номер unknown

Опубликована: Авг. 2, 2024

Язык: Английский

---

Integrating automated machine learning and metabolic reprogramming for the identification of microplastic in soil: A case study on soybean

Journal of Hazardous Materials, Год журнала: 2024, Номер 478, С. 135555 - 135555

Опубликована: Авг. 23, 2024

Язык: Английский

---

Modeling processes and sensitivity analysis of machine learning methods for environmental data

Elsevier eBooks, Год журнала: 2024, Номер unknown, С. 511 - 522

Опубликована: Янв. 1, 2024

Язык: Английский

---