Application of a QPSO-optimized CNN-LSTM model in water quality prediction DOI Creative Commons
Yue Zhu

Discover Water, Journal Year: 2024, Volume and Issue: 4(1)

Published: Nov. 12, 2024

Globally, over 80% of wastewater is discharged into water bodies without adequate treatment (UNESCO 2017:10–15), making accurate quality prediction essential for safeguarding aquatic ecosystems and public health. This study presents a novel QPSO-CNN-LSTM model that significantly advances by combining Quantum Particle Swarm Optimization (QPSO) with CNN-LSTM architecture. Unlike traditional models, the leverages CNN to capture complex spatial features from data LSTM long-term temporal dependencies. The QPSO algorithm optimizes key hyperparameters, mitigating need manual tuning improving model's adaptability dynamic environmental data. outperforms methods 15–50% improvement in RMSE, MSE, MAE, MAPE dissolved oxygen pH predictions. These enhancements demonstrate superior accuracy robustness, it an invaluable tool real-time monitoring, pollution prevention, cost-effective management strategies. practical implications this offer step forward preserving through data-driven stewardship.

Language: Английский

Understanding dissolved organic matters in stormwater from different urban land uses: implications for reuse safety DOI
Min Ding, Zhifeng Chen, Jun Li

et al.

Frontiers of Environmental Science & Engineering, Journal Year: 2025, Volume and Issue: 19(4)

Published: Feb. 20, 2025

Language: Английский

Citations

0
Rainwater extracting characteristics and its potential impact on DBPs generation: A case study DOI

Yujin Yuan,

Qingsong Li, Jing Deng

et al.

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 906, P. 167282 - 167282

Published: Sept. 26, 2023

Language: Английский

Citations

7
Fluorescence fingerprint as an indicator to identify urban non-point sources in urban river during rainfall period DOI
Qiuran Xiong, Yiming Song, Jian Shen

et al.

Environmental Research, Journal Year: 2023, Volume and Issue: 245, P. 118009 - 118009

Published: Dec. 21, 2023

Language: Английский

Citations

4
Fluorescence spectroscopy for tracking microbiological contamination in urban waterbodies DOI Creative Commons

Natália Angelotti de Ponte Rodrigues,

Rémi Carmigniani,

Arthur Guillot-Le Goff

et al.

Frontiers in Water, Journal Year: 2024, Volume and Issue: 6

Published: May 13, 2024

Dissolved organic matter (DOM) plays a crucial role in freshwater ecosystem function. Monitoring of DOM aquatic environments can be achieved by using fluorescence spectroscopy. Particularly, constitute signature microbiological contamination with potential for high frequency monitoring. However, limited data are available regarding urban waterbodies. This study considers from field campaigns conducted the Paris metropolitan region: two watercourses (La Villette basin and river Marne), stormwater network outlets (SO), wastewater treatment plant effluent (WWTP-O). The objectives were to characterize major components studied sites, investigate impact local rainfall such identify contamination. PARAFAC model (C1-C7), corresponding couple excitation (ex) emission (em) wavelengths, indices HIX BIX used characterization. In parallel, fecal indicator bacteria (FIB) measured selected samples. protein-like components, C6 (ex/em 280/352 nm) C7 305/340 nm), identified as markers microbial sites. La basin, where samplings covered period more than 2 years, which also included similar numbers wet dry weather samples, significantly higher comparison weather. A positive relationship was obtained between FIB. rivers, monitoring levels would support detection rivers. addition, it could help targeting specific collect comprehensive dataset episodes.

Language: Английский

Citations

0
New insight into the spatiotemporal distribution and ecological risk assessment of endocrine-disrupting chemicals in the Minjiang and Tuojiang rivers: perspective of watershed landscape patterns DOI

Weike Zhao,

Peilin Li,

Bo Yang

et al.

Environmental Science Processes & Impacts, Journal Year: 2024, Volume and Issue: 26(8), P. 1360 - 1372

Published: Jan. 1, 2024

This study evaluated the pollution characteristics, spatiotemporal distribution, and ecological risks of eight endocrine-disrupting chemicals (EDCs) in Minjiang Tuojiang rivers. Utilizing 3S technology (ArcGIS, remote sensing, GPS) Fragstats, research calculated landscape pattern indices related to land use types along river established correlations between factors EDC distribution through stepwise multiple regression. The results indicated that bisphenol A (BPA) nonylphenol (NP) were most concerning EDCs, with detection frequencies 97-100% peak concentrations up 63.35 ng L

Language: Английский

Citations

0
Spatial differences of dissolved organic matter composition and humification in an artificial lake DOI Creative Commons
Jin Zhang, Jiajia Tan, Yingjie Wang

et al.

Water Science & Technology, Journal Year: 2024, Volume and Issue: 90(3), P. 995 - 1008

Published: Aug. 1, 2024

The depth-dependent dynamics of dissolved organic matter (DOM) structure and humification in an artificial lake limits the understanding eutrophication carbon cycling. Using fluorescence regional integration (FRI) parallel factor analysis (PARAFAC) models to analyze 3D spectroscopy dataset, we revealed vertical distribution DOM estuarine center regions Lake Hongfeng. percentage response (

Language: Английский

Citations

0
Application of a QPSO-optimized CNN-LSTM model in water quality prediction DOI Creative Commons
Yue Zhu

Discover Water, Journal Year: 2024, Volume and Issue: 4(1)

Published: Nov. 12, 2024

Globally, over 80% of wastewater is discharged into water bodies without adequate treatment (UNESCO 2017:10–15), making accurate quality prediction essential for safeguarding aquatic ecosystems and public health. This study presents a novel QPSO-CNN-LSTM model that significantly advances by combining Quantum Particle Swarm Optimization (QPSO) with CNN-LSTM architecture. Unlike traditional models, the leverages CNN to capture complex spatial features from data LSTM long-term temporal dependencies. The QPSO algorithm optimizes key hyperparameters, mitigating need manual tuning improving model's adaptability dynamic environmental data. outperforms methods 15–50% improvement in RMSE, MSE, MAE, MAPE dissolved oxygen pH predictions. These enhancements demonstrate superior accuracy robustness, it an invaluable tool real-time monitoring, pollution prevention, cost-effective management strategies. practical implications this offer step forward preserving through data-driven stewardship.

Language: Английский

Citations

0