Machine Learning-Based Predominant Driving Factors Impacting Urban Industrial Wastewater Discharge in the Yellow River Basin

Abstract

Environmental pollution control has become an important task of ecological protection, which is one of the major strategies for high-quality development of the Yellow River basin (YRB) in China. In this paper, a machine learning model is constructed to explore the driving factors that affect industrial wastewater discharge (IWD) in prefecture-level cities in the YRB. On the basis of statistical data from 2003 to 2018, the relationship between IWD and gross regional product in the YRB obeyed the Environmental Kuznets Curve (EKC) and reached an inflection point in 2010, but not all cities fit the EKC. Therefore, three machine learning algorithms, including weighted k-nearest neighbor (knn), random forest, and support vector machine, are used to construct a regression model of IWD. The knn achieved the best fitting, with determination coefficients (R 2 ) of 0.98 and 0.91 in the training and testing sets, respectively. Variable importance and partial dependence plots explain the machine learning model well. This work provides ideas for the management of IWD in prefecture-level cities in the YRB and references for environmental pollution in other cities.

Applications of Artificial Intelligence, Machine Learning, and Data Analytics in Water Environments

“…Other contributions introduce novel methodologies that combine ML with statistical approaches to address various environmental problems, such as sewer overflow pollution abatement, fault detection in water and wastewater treatment, , an assay for source apportionment of per- and polyfluorinated substances (PFAS), detection of freshwater algae, and influent water data . Beyond water quality, ML has been applied to model water quantity, exploring dominant factors influencing urban industrial wastewater discharges, model energy consumption of wastewater treatment, identify endocrine-active pollutants in the organic Unregulated Contaminant Monitoring Rule (UCMR 1–4) and their toxic potentials, and employ quantitative biodescriptors to predict in vivo toxicity …”