Optimization of water quality index models using machine learning approaches

Fei Ding; Wenjie Zhang; Shaohua Cao; Shilong Hao; Liangyao Chen; Xin Xie; Wenpan Li; Mingcen Jiang

doi:10.1016/j.watres.2023.120337

Optimization of water quality index models using machine learning approaches

Water Res. 2023 Sep 1:243:120337. doi: 10.1016/j.watres.2023.120337. Epub 2023 Jul 11.

Authors

Fei Ding¹, Wenjie Zhang², Shaohua Cao³, Shilong Hao², Liangyao Chen², Xin Xie⁴, Wenpan Li⁴, Mingcen Jiang⁵

Affiliations

¹ Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering, College of Architecture and Civil Engineering, Beijing University of Technology, Beijing 100124, China. Electronic address: dingfei@bjut.edu.cn.
² Key Laboratory of Beijing for Water Quality Science and Water Environment Recovery Engineering, College of Architecture and Civil Engineering, Beijing University of Technology, Beijing 100124, China.
³ State Environmental Protection Key Laboratory of Soil Environmental Management and Pollution Control, Nanjing Institute of Environmental Sciences, Ministry of Ecology and Environment of China, Nanjing 210042, Jiangsu, China.
⁴ State Environmental Protection Key Laboratory of Quality Control in Environmental Monitoring, China National Environmental Monitoring Centre, Beijing 100012, China.
⁵ State Environmental Protection Key Laboratory of Quality Control in Environmental Monitoring, China National Environmental Monitoring Centre, Beijing 100012, China. Electronic address: jiangmc@cnemc.cn.

PMID: 37473509
DOI: 10.1016/j.watres.2023.120337

Abstract

To optimize the water quality index (WQI) assessment model, this study upgraded the parameter weight values and aggregation functions. We determined the combined weights based on machine learning and game theory to improve the accuracy of the models, and proposed new aggregation functions to reduce the uncertainty of the model. A new water quality assessment system was established, and took the Chaobai River Basin as a case study. To optimize the weight, two combined weights were established based on game theory. The weight CW_AE was combined by the Analytic Hierarchy Process (AHP) and Entropy Weight Method (EWM). The weight CW_AL was combined by AHP and machine learning (LightGBM). CW_AL was judged to be an optimal composite weight by comparing the coefficient of variation (CV) values and the Kaiser-Meyer-Olkin (KMO) extracted values. To reduce the uncertainty of the model, we proposed two aggregation functions, the Sinusoidal Weighted Mean (SWM) and the Log-weighted Quadratic Mean (LQM). The three water quality assessment models (WQI_S, WQI_L and WQI_W) were established based on the optimal weights besides. All three models had good reliability. Both WQI_S and WQI_W models had low eclipsing problems (25.49% and 18.63%). The accuracy of the models was ranked as WQI_S > WQI_W > WQI_L. The uncertainty of WQIs (0.000) in assessing poor water quality was low, and so was WQI_W (0.259) in assessing good water quality. Overall, the WQI_S model was recommended for assessing poor water quality and the WQI_W model was recommended for assessing good water quality. The assessment results of WQI_S showed that the Chaobai River Basin was "slightly polluted", and the water quality upstream was better than that downstream. TN was the main pollutant in the basin, and there was slight pollution with COD_Mn, COD_Cr, BOD₅, etc. There was little metal contamination, only a few months exceeded Class I. The model established in this study can provide a reference for the same type work of water quality assessment. The assessment results can provide a scientific basis for the protection of the regional water environment.

Keywords: Chaobai River Basin; Combined weight; Game theory; LightGBM; Machine learning; Water quality assessment.

MeSH terms

China
Environmental Monitoring / methods
Machine Learning
Reproducibility of Results
Rivers
Water Pollutants, Chemical* / analysis
Water Quality*

Substances

Water Pollutants, Chemical