Turbidity prediction of lake-type raw water using random forest model based on meteorological data: A case study of Tai lake, China

Zhang, Yiping; Yao, Xinyu; Wu, Qiang; Huang, Yongbin; Zhou, Zhixu; Yang, Jin-Ling; Liu, Xiaowei

doi:10.1016/j.jenvman.2021.112657

Cited by 32 publications

(10 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RF is a ML algorithm that generates repeated predictions for a given phenomenon by constructing a large number of decision trees and is widely used in various regression prediction models. 48 Fig. 2(a) shows the schematic diagram of RF generation.…”

Section: Model and Theorymentioning

confidence: 99%

Design of an ultra-broadband terahertz absorber based on a patterned graphene metasurface with machine learning

Ding

Luo

et al. 2023

J. Mater. Chem. C

View full text Add to dashboard Cite

show abstract

Section: Model and Theorymentioning

confidence: 99%

Design of an ultra-broadband terahertz absorber based on a patterned graphene metasurface with machine learning

Ding

Luo

et al. 2023

J. Mater. Chem. C

View full text Add to dashboard Cite

show abstract

“…Although those nine features were highly correlated with the target variable, two features among them had a multi-collinearity problem. This could be solved by eventually selecting the features with low inter-correlation coefficients (0.4 or less [48,49]). As shown in Figure 5, we could recognize that WT and DO should be eliminated because they were highly correlated with the other features.…”

Section: Determination Of the Modeling Casesmentioning

confidence: 99%

“…To address the multi-collinearity problem, the correlation analysis was performed among the features selected in the first step. As the final process for the feature selection, the paired features with low inter-correlation coefficients (0.4 or less [48,49]) were selected. Here, Pearson's correlation analysis was performed with only 241 instances by excluding the zero values of Cyano(t) in 378 instances, as the zero values were able to distort the analysis result.…”

mentioning

confidence: 99%

A Classification-Based Machine Learning Approach to the Prediction of Cyanobacterial Blooms in Chilgok Weir, South Korea

Kim

Jonoski

Solomatine

2022

Water

View full text Add to dashboard Cite

Cyanobacterial blooms appear by complex causes such as water quality, climate, and hydrological factors. This study aims to present the machine learning models to predict occurrences of these complicated cyanobacterial blooms efficiently and effectively. The dataset was classified into groups consisting of two, three, or four classes based on cyanobacterial cell density after a week, which was used as the target variable. We developed 96 machine learning models for Chilgok weir using four classification algorithms: k-Nearest Neighbor, Decision Tree, Logistic Regression, and Support Vector Machine. In the modeling methodology, we first selected input features by applying ANOVA (Analysis of Variance) and solving a multi-collinearity problem as a process of feature selection, which is a method of removing irrelevant features to a target variable. Next, we adopted an oversampling method to resolve the problem of having an imbalanced dataset. Consequently, the best performance was achieved for models using datasets divided into two classes, with an accuracy of 80% or more. Comparatively, we confirmed low accuracy of approximately 60% for models using datasets divided into three classes. Moreover, while we produced models with overall high accuracy when using logCyano (logarithm of cyanobacterial cell density) as a feature, several models in combination with air temperature and NO3-N (nitrate nitrogen) using two classes also demonstrated more than 80% accuracy. It can be concluded that it is possible to develop very accurate classification-based machine learning models with two features related to cyanobacterial blooms. This proved that we could make efficient and effective models with a low number of inputs.

show abstract

“…Turbidity indicates the concentration of suspended substances in water samples, such as clay, organic matter, and inorganic matter, which often correlates with other pollutants in the water (Zhang et al, 2021). In municipal water supplies, turbidity is considered a surrogate measurement for both microbiological contamination and water clarity (Tinker et al, 2010).…”

Section: Impacts On Water Qualitymentioning

confidence: 99%

The influence of recent bushfires on water quality and the operation of water purification systems in regional NSW

Jackson

Krishna

Pillai

et al. 2022

Preprint

View full text Add to dashboard Cite

The occurrence of extreme weather events aggravated by global warming has increased exponentially in the past decade. In recent years, New South Wales (NSW) has experienced catastrophic, unprecedented drought followed by extreme fire conditions and widespread, record-breaking floods. However, the impact of the drought and bushfires on regional water quality is poorly investigated. This study investigates immediate changes in regional environmental water quality and specific issues reported from water purification unit operations after the 2019 bushfires in NSW. Water samples were collected from the waterways in bushfire affected catchments, water intakes, unit operations of water purification systems, and treated water reservoirs. The performance of water purification processes and operational issues of regional water purification plants were extracted from daily maintenance records. Elevated levels of suspended sediment and algae blooms were of most concern in terms of aesthetic and recreation aspects of environmental water use after the drought and bushfires. Compared to the control samples, extremely high turbidity and low dissolved oxygen concentrations were reported in bushfire affected water samples. Shock loads of sediment and ash detected in waterways posed the greatest threat to aquatic ecosystems following fires. Elevated turbidity levels, suspended sediments, pH, and alkalinity were the parameters of most concern for water purification. The maximum turbidity level in raw water was reported as more than 195 NTU. Flocculation and sedimentation were the most affected unit operations in the water purification process. Immediate measures to handle shock loads due to bushfires included sediment traps, aeration on the intake lines, pre-chlorination, and inline monitoring for pH and turbidity. These results can be used to develop management strategies to minimise the impacts of bushfires on water quality and modify and operate water purification plants within bushfire prone catchments.

show abstract

Turbidity prediction of lake-type raw water using random forest model based on meteorological data: A case study of Tai lake, China

Cited by 32 publications

References 15 publications

Design of an ultra-broadband terahertz absorber based on a patterned graphene metasurface with machine learning

Design of an ultra-broadband terahertz absorber based on a patterned graphene metasurface with machine learning

A Classification-Based Machine Learning Approach to the Prediction of Cyanobacterial Blooms in Chilgok Weir, South Korea

The influence of recent bushfires on water quality and the operation of water purification systems in regional NSW

Contact Info

Product

Resources

About