Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach

Wastewater treatment and reuse are being regarded as the most effective strategy for combating water scarcity threats. This study examined and reported the applications of the Internet of Things (IoT) and artificial intelligence in the phytoremediation of wastewater using Salvinia molesta plants. Wa...

Full description

Saved in:
Bibliographic Details
Main Authors: Mustafa H.M., Hayder G., Abba S.I., Algarni A.D., Mnzool M., Nour A.H.
Other Authors: 57217195204
Format: Article
Published: Multidisciplinary Digital Publishing Institute (MDPI) 2024
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.uniten.dspace-34281
record_format dspace
spelling my.uniten.dspace-342812024-10-14T11:18:48Z Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach Mustafa H.M. Hayder G. Abba S.I. Algarni A.D. Mnzool M. Nour A.H. 57217195204 56239664100 57208942739 57204971671 57852200500 14719696000 computational analysis energy error ensemble methods total dissolved solids water quality forecasting Biochemical oxygen demand Biological water treatment Bioremediation Dissolved oxygen Errors Forecasting Internet of things Learning systems Mean square error Potable water Quality control Redox reactions Wastewater reclamation Wastewater treatment Water conservation After-treatment Computational analysis Energy Ensemble methods Error ensemble method Means square errors Root mean square errors Square-root Total dissolved solids Water quality forecasting Water quality Wastewater treatment and reuse are being regarded as the most effective strategy for combating water scarcity threats. This study examined and reported the applications of the Internet of Things (IoT) and artificial intelligence in the phytoremediation of wastewater using Salvinia molesta plants. Water quality (WQ) indicators (total dissolved solids (TDS), temperature, oxidation-reduction potential (ORP), and turbidity) of the S. molesta treatment system at a retention time of 24 h were measured using an Arduino IoT device. Finally, four machine learning tools (ML) were employed in modeling and evaluating the predicted concentration of the total dissolved solids after treatment (TDSt) of the water samples. Additionally, three nonlinear error ensemble methods were used to enhance the prediction accuracy of the TDSt models. The outcome obtained from the modeling and prediction of the TDSt depicted that the best results were observed at SVM-M1 with 0.9999, 0.0139, 1.0000, and 0.1177 for R2, MSE, R, and RMSE, respectively, at the training stage. While at the validation stage, the R2, MSE, R, and RMSE were recorded as 0.9986, 0.0356, 0.993, and 0.1887, respectively. Furthermore, the error ensemble techniques employed significantly outperformed the single models in terms of mean square error (MSE) and root mean square error (RMSE) for both training and validation, with 0.0014 and 0.0379, respectively. � 2023 by the authors. Final 2024-10-14T03:18:48Z 2024-10-14T03:18:48Z 2023 Article 10.3390/pr11020478 2-s2.0-85149251802 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85149251802&doi=10.3390%2fpr11020478&partnerID=40&md5=33311389956c428039d39c23b4d3a797 https://irepository.uniten.edu.my/handle/123456789/34281 11 2 478 All Open Access Gold Open Access Multidisciplinary Digital Publishing Institute (MDPI) Scopus
institution Universiti Tenaga Nasional
building UNITEN Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Tenaga Nasional
content_source UNITEN Institutional Repository
url_provider http://dspace.uniten.edu.my/
topic computational analysis
energy
error ensemble methods
total dissolved solids
water quality forecasting
Biochemical oxygen demand
Biological water treatment
Bioremediation
Dissolved oxygen
Errors
Forecasting
Internet of things
Learning systems
Mean square error
Potable water
Quality control
Redox reactions
Wastewater reclamation
Wastewater treatment
Water conservation
After-treatment
Computational analysis
Energy
Ensemble methods
Error ensemble method
Means square errors
Root mean square errors
Square-root
Total dissolved solids
Water quality forecasting
Water quality
spellingShingle computational analysis
energy
error ensemble methods
total dissolved solids
water quality forecasting
Biochemical oxygen demand
Biological water treatment
Bioremediation
Dissolved oxygen
Errors
Forecasting
Internet of things
Learning systems
Mean square error
Potable water
Quality control
Redox reactions
Wastewater reclamation
Wastewater treatment
Water conservation
After-treatment
Computational analysis
Energy
Ensemble methods
Error ensemble method
Means square errors
Root mean square errors
Square-root
Total dissolved solids
Water quality forecasting
Water quality
Mustafa H.M.
Hayder G.
Abba S.I.
Algarni A.D.
Mnzool M.
Nour A.H.
Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
description Wastewater treatment and reuse are being regarded as the most effective strategy for combating water scarcity threats. This study examined and reported the applications of the Internet of Things (IoT) and artificial intelligence in the phytoremediation of wastewater using Salvinia molesta plants. Water quality (WQ) indicators (total dissolved solids (TDS), temperature, oxidation-reduction potential (ORP), and turbidity) of the S. molesta treatment system at a retention time of 24 h were measured using an Arduino IoT device. Finally, four machine learning tools (ML) were employed in modeling and evaluating the predicted concentration of the total dissolved solids after treatment (TDSt) of the water samples. Additionally, three nonlinear error ensemble methods were used to enhance the prediction accuracy of the TDSt models. The outcome obtained from the modeling and prediction of the TDSt depicted that the best results were observed at SVM-M1 with 0.9999, 0.0139, 1.0000, and 0.1177 for R2, MSE, R, and RMSE, respectively, at the training stage. While at the validation stage, the R2, MSE, R, and RMSE were recorded as 0.9986, 0.0356, 0.993, and 0.1887, respectively. Furthermore, the error ensemble techniques employed significantly outperformed the single models in terms of mean square error (MSE) and root mean square error (RMSE) for both training and validation, with 0.0014 and 0.0379, respectively. � 2023 by the authors.
author2 57217195204
author_facet 57217195204
Mustafa H.M.
Hayder G.
Abba S.I.
Algarni A.D.
Mnzool M.
Nour A.H.
format Article
author Mustafa H.M.
Hayder G.
Abba S.I.
Algarni A.D.
Mnzool M.
Nour A.H.
author_sort Mustafa H.M.
title Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
title_short Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
title_full Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
title_fullStr Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
title_full_unstemmed Performance Evaluation of Hydroponic Wastewater Treatment Plant Integrated with Ensemble Learning Techniques: A Feature Selection Approach
title_sort performance evaluation of hydroponic wastewater treatment plant integrated with ensemble learning techniques: a feature selection approach
publisher Multidisciplinary Digital Publishing Institute (MDPI)
publishDate 2024
_version_ 1814061114008272896
score 13.214268