Hybrid WT–CNN–GRU-based model for the estimation of reservoir water quality variables considering spatio-temporal features

Journal of Environmental Management, Journal Year: 2024, Volume and Issue: 358, P. 120756 - 120756

Published: April 9, 2024

Language: Английский

---

A stacking ANN ensemble model of ML models for stream water quality prediction of Godavari River Basin, India

Ecological Informatics, Journal Year: 2024, Volume and Issue: 80, P. 102500 - 102500

Published: Jan. 28, 2024

The importance of water quality models has increased as their inputs are critical to the development risk assessment framework for environmental management and monitoring rivers. However, with advent a plethora recent advances in ML algorithms better predictions possible. This study proposes causal effect model by considering climatological such temperature precipitation along geospatial information related agricultural land use factor (ALUF), forest (FLUF), grassland usage (GLUF), shrub (SLUF), urban (ULUF). All these factors included input data, whereas four Stream Water Quality parameters (SWQPs) Electrical Conductivity (EC), Biochemical Oxygen Demand (BOD), Nitrate, Dissolved (DO) from 2019 2021 taken outputs predict Godavari River Basin quality. In preliminary investigation, out SWQPs, nitrate's coefficient variation (CV) is high, revealing close association climate practices across sampling stations. authors' earlier study, using single-layer Feed-Forward Neural Network (FFNN) showed improved performance predicting cause linked metrics. To achieve prediction, stacked ANN meta-model nine conventional machine learning (ML) models, including Extreme Gradient Boosting (XGB), Extra Trees (ET), Bagging (BG), Random Forest (RF), AdaBoost or Adaptive (ADB), Decision Tree (DT), Highest (HGB), Light Method (LGBM), (GB), were compared this study. According study's findings, outperformed stand-alone FFNN same dataset superior predictive capabilities terms accuracy forecasting variable interest. For instance, during testing, determination (R2) (BOD) 0.72 0.87. Furthermore, Artificial (ANN) meta that was reinforced (ET) base performed than individual (from R2 = 0.87 0.91 BOD testing). By new framework, effort hyperparameter tuning can be minimized.

Language: Английский

---

Machine-learning-based performance prediction of the energy pile heat pump system

Journal of Building Engineering, Journal Year: 2023, Volume and Issue: 77, P. 107442 - 107442

Published: July 25, 2023

Language: Английский

---

Smarter water quality monitoring in reservoirs using interpretable deep learning models and feature importance analysis

Journal of Water Process Engineering, Journal Year: 2024, Volume and Issue: 60, P. 105187 - 105187

Published: April 1, 2024

Language: Английский

---

Machine learning driven forecasts of agricultural water quality from rainfall ionic characteristics in Central Europe

Agricultural Water Management, Journal Year: 2024, Volume and Issue: 293, P. 108690 - 108690

Published: Jan. 21, 2024

Sodium hazard poses a critical threat to agricultural production globally and regionally which has been previously predicted from ground or surface water. Monitoring rainwater quality in this context is ignored but essential for water management central Europe. Our study focused predict sodium adsorption ratio (SAR) 1985 2021 ten ionic species of (pH, EC, Cl-, SO4−2, NO3-, NH4+, Na+, K+, Mg2+, Ca2+) employing four machine learning (random forest (RF), gaussian process regression (GU), random subspace (RSS), artificial neural network-multilayer perceptron (ANN-MLP)) methods at three stations K-puszta (KP), Farkasfa (FAK), Nyirjes (NYR) Hungary, Exploratory data analysis was performed using the Mann-Kendall test, Pearson correlation, principal component (PCA). Rainwater composition revealed highest percentage SO4−2 ions i.e., 21 31%, followed by 10 15% Na+ ions. test significant (p < 0.05) increasing trend SAR portraying it serious limiting production. Machine results model runs all algorithms prediction KP station proved efficacy ANN-MLP as superior with RMSE range 0.02 0.05, RF 0.14 0.19 scenario 2 (SC-2) (Na+, Ca2+). Validation best-selected algorithm (ANN-MLP) also low 0.08 0.05 both FAK NYR stations, respectively. Hence, efficiency forecasting proves be meticulous tool enhancing practices Central Europe resource crop future.

Language: Английский

---

Stacked hybridization to enhance the performance of artificial neural networks (ANN) for prediction of water quality index in the Bagh river basin, India

Heliyon, Journal Year: 2024, Volume and Issue: 10(10), P. e31085 - e31085

Published: May 1, 2024

Water quality assessment is paramount for environmental monitoring and resource management, particularly in regions experiencing rapid urbanization industrialization. This study introduces Artificial Neural Networks (ANN) its hybrid machine learning models, namely ANN-RF (Random Forest), ANN-SVM (Support Vector Machine), ANN-RSS Subspace), ANN-M5P (M5 Pruned), ANN-AR (Additive Regression) water the rapidly urbanizing industrializing Bagh River Basin, India. The Relief algorithm was employed to select most influential input parameters, including Nitrate (NO

Language: Английский

---

Application of machine learning algorithms and feature selection methods for better prediction of sludge production in a real advanced biological wastewater treatment plant

Journal of Environmental Management, Journal Year: 2023, Volume and Issue: 348, P. 119448 - 119448

Published: Nov. 6, 2023

Language: Английский

---

Harnessing neural network model with optimization for enhanced ciprofloxacin antibiotic adsorption from contaminated water: A transparent and objective framework

Journal of Water Process Engineering, Journal Year: 2024, Volume and Issue: 65, P. 105724 - 105724

Published: July 16, 2024

Language: Английский

---

A critical analysis of parameter choices in water quality assessment

Water Research, Journal Year: 2024, Volume and Issue: 258, P. 121777 - 121777

Published: May 16, 2024

The determination of water quality heavily depends on the selection parameters recorded from samples for index (WQI). Data-driven methods, including machine learning models and statistical approaches, are frequently used to refine parameter set four main reasons: reducing cost uncertainty, addressing eclipsing problem, enhancing performance predicting WQI. Despite their widespread use, there is a noticeable gap in comprehensive reviews that systematically examine previous studies this area. Such essential assess validity these objectives demonstrate effectiveness data-driven methods achieving goals. This paper sets out with two primary aims: first, provide review existing literature selecting parameters. Second, it seeks delineate evaluate principal motivations identified literature. manuscript categorizes into methodological groups refining parameters: one focuses preserving information within dataset, another ensures consistent prediction using full It characterizes each group evaluates how effectively approach meets predefined objectives. study presents minimal WQI approach, common both categories, only has successfully reduced recording costs. Nonetheless, notes simply number does not guarantee savings. Furthermore, classified as dataset demonstrated potential decrease whereas have been able mitigate issue. Additionally, since approaches still rely initial chosen by experts, they do eliminate need expert judgment. further points formula straightforward expedient tool assessing quality. Consequently, argues employing solely reduce enhance standalone solution. Rather, objective should be integrated more research critical analysis characterization lay groundwork future research. will enable subsequent proposed can achieve

Language: Английский

---

Predicting water quality in municipal water management systems using a hybrid deep learning model

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108420 - 108420

Published: April 23, 2024

Language: Английский

---