Prediction of Water Quality Index of Island Counties Under River Length System—A Case Study of Yuhuan City DOI Creative Commons
Cheng Zhang, Lei Wang, Chuan Lin

et al.

Journal of Marine Science and Engineering, Journal Year: 2025, Volume and Issue: 13(3), P. 539 - 539

Published: March 11, 2025

In order to cope with the extremely difficult challenges of water pollution control, China has widely implemented river chief system. The quality monitoring surface environment, as a solid defense line safeguard human health and ecosystem balance, is great importance in As well-known island county China, Yuhuan City holds even more precious resources. Leveraging machine learning technology develop prediction models significance for enhancing evaluation environment quality. This case study aims evaluate effectiveness six predicting index (CWQI) uses SHAP (Shapley Additive exPlans) an interpretability analysis method deeply analyze contribution each variable model’s results. research results show that all exhibited good performance CWQI, number significantly correlated variables input increased, accuracy also showed gradual improvement trend. Under optimal combination, Extreme Gradient Boosting model demonstrated best performance, root mean square error (RMSE) 0.7081, absolute (MAE) 0.4702, adjusted coefficient determination (Adj.R2) 0.6400. Through analysis, we found concentrations TP (total phosphorus), NH3-N (ammonia nitrogen), CODCr (chemical oxygen demand) have significant impact on CWQI City. implementation system not only enhances pertinence management, but provides richer accurate data support models, further improving reliability models.

Language: Английский

Prediction of Water Quality Index of Island Counties Under River Length System—A Case Study of Yuhuan City DOI Creative Commons
Cheng Zhang, Lei Wang, Chuan Lin

et al.

Journal of Marine Science and Engineering, Journal Year: 2025, Volume and Issue: 13(3), P. 539 - 539

Published: March 11, 2025

In order to cope with the extremely difficult challenges of water pollution control, China has widely implemented river chief system. The quality monitoring surface environment, as a solid defense line safeguard human health and ecosystem balance, is great importance in As well-known island county China, Yuhuan City holds even more precious resources. Leveraging machine learning technology develop prediction models significance for enhancing evaluation environment quality. This case study aims evaluate effectiveness six predicting index (CWQI) uses SHAP (Shapley Additive exPlans) an interpretability analysis method deeply analyze contribution each variable model’s results. research results show that all exhibited good performance CWQI, number significantly correlated variables input increased, accuracy also showed gradual improvement trend. Under optimal combination, Extreme Gradient Boosting model demonstrated best performance, root mean square error (RMSE) 0.7081, absolute (MAE) 0.4702, adjusted coefficient determination (Adj.R2) 0.6400. Through analysis, we found concentrations TP (total phosphorus), NH3-N (ammonia nitrogen), CODCr (chemical oxygen demand) have significant impact on CWQI City. implementation system not only enhances pertinence management, but provides richer accurate data support models, further improving reliability models.

Language: Английский

Citations

0