Predicting the concentration of sulfate using machine learning methods

Comparison of individual and ensemble machine learning models for prediction of sulphate levels in untreated and treated Acid Mine Drainage

Machine learning was used to provide data for further evaluation of potential extraction of octathiocane (S8), a commercially useful by-product, from Acid Mine Drainage (AMD) by predicting sulphate levels in an AMD water quality dataset. Individual ML regressor models, namely: Linear Regression (LR), Least Absolute Shrinkage and Selection Operator (LASSO), Ridge (RD), Elastic Net (EN), K-Nearest Neighbours (KNN), Support Vector Regression (SVR), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multi-Layer Perceptron Artificial Neural Network (MLP) and Stacking Ensemble (SE-ML) combinations of these models were successfully used to predict sulphate levels. A SE-ML regressor trained on untreated AMD which stacked seven of the best-performing individual models and fed them to a LR meta-learner model was found to be the best-performing model with a Mean Squared Error (MSE) of 0.000011, Mean Absolute Error (MAE) of 0.002617 and R2 of 0.9997. Temperature (°C), Total Dissolved Solids (mg/L) and, importantly, iron (mg/L) were highly correlated to sulphate (mg/L) with iron showing a strong positive linear correlation that indicated dissolved products from pyrite oxidation. Ensemble learning (bagging, boosting and stacking) outperformed individual methods due to their combined predictive accuracies. Surprisingly, when comparing SE-ML that combined all models with SE-ML that combined only the best-performing models, there was only a slight difference in model accuracies which indicated that including bad-performing models in the stack had no adverse effect on its predictive performance.

Groundwater quality assessment by multi-model comparison: a comprehensive study during dry and wet periods in semi-arid regions

With the impact of human engineering activities, groundwater pollution has seriously threatened the health of human life. Accurate water quality assessment is the basis of controlling groundwater pollution and improving groundwater management, especially in specific regions. A typical semi-arid city in Fuxin Province of China is taken as an example. We use remote sensing and GIS to compile four environmental factors, such as rainfall, temperature, LULC, and NDVI, to analyze and screen the correlation of indicators. The differences among the four algorithms were compared by using hyperparameters and model interpretability, including random forest (RF), support vector machine support vector machine (SVM), decision tree (DT), and K-nearest neighbor (KNN). The groundwater quality of the city during the dry and wet periods was comprehensively evaluated. The results show that the RF model has higher integrated precision (MSE = 0.11, 0.035; RMSE = 0.19,0.188; R² = 0.829,0.811; ROC = 0.98, 0.98). The quality of shallow groundwater is poor in general, 29%, 38%, 33% of the groundwater quality in low-water period is III, IV, V water. Thirty-three percent and 67% of the groundwater quality in the high-water period were IV and V water. The proportion of poor water quality in high-water period was higher than that in low-water period, which was consistent with the actual investigation. This study provides a kind of machine learning method for the semi-arid area, which cannot only promote the sustainable development of groundwater in this area, but also provide reference for the management policy of related departments.

Experimental Analysis and Neural Network Modeling of the Rheological Behavior of Xanthan Gum and Its Derivatives

The main objective of this study was to create a mathematical tool that could be used with experimental data to predict the rheological flow behavior of functionalized xanthan gum according to the types of chemical groups grafted onto its backbone. Different rheological and physicochemical analyses were applied to assess six derivatives synthesized via the etherification of xanthan gum by hydrophobic benzylation with benzyl chloride and carboxymethylation with monochloroacetic acid at three (regent/polymer) ratios R equal to 2.4 and 6. Results from the FTIR study verified that xanthan gum had been modified. The degree of substitution (DS) values varying between 0.2 and 2.9 for carboxymethylxanthan gum derivatives were found to be higher than that of hydrophobically modified benzyl xanthan gum for which the DS ranged from 0.5 to 1. The molecular weights of all the derivatives were found to be less than that of xanthan gum for the two types of derivatives, decreasing further as the degree of substitution (DS) increased. However, the benzyl xanthan gum derivatives presented higher molecular weights varying between 1,373,146 (g/mol) and 1,262,227 (g/mol) than carboxymethylxanthan gum derivatives (1,326,722-1,015,544) (g/mol). A shear-thinning behavior was observed in the derivatives, and the derivatives' viscosity was found to decrease with increasing DS. The second objective of this research was to create an ANN model to predict one of the rheological properties (the apparent viscosity). The significance of the ANN model (R² = 0.99998 and MSE = 5.95 × 10^-3) was validated by comparing experimental results with the predicted ones. The results showed that the model was an efficient tool for predicting rheological flow behavior.

Application of Walnut Shell Biowaste as an Inexpensive Adsorbent for Methylene Blue Dye: Isotherms, Kinetics, Thermodynamics, and Modeling

This research aimed to assess the adsorption properties of raw walnut shell powder (WNSp) for the elimination of methylene blue (MB) from an aqueous medium. The initial MB concentration (2–50 mg/L), the mass of the biomaterial (0.1–1 g/L), the contact time (10–120 min), the medium’s pH (2–12), and the temperature (25–55 °C) were optimized as experimental conditions. A maximum adsorption capacity of 19.99 mg/g was obtained at an MB concentration of 50 mg/L, a medium pH of 6.93 and a temperature of 25 °C, using 0.2 g/L of WNSp. These conditions showed that the MB dye elimination process occurred spontaneously. Different analytical approaches were used to characterize the WNSp biomaterial, including functional groups involved in MB adsorption, the surface characteristics and morphological features of the WNSp before and after MB uptake, and identification of WNSp based on their diffraction pattern. The experimental isotherm data were analyzed by the Langmuir and Freundlich models for the adsorption of MB dye. The corresponding values of parameter RL of Langmuir were between 0.51 and 0.172, which confirmed the WNSp’s favorable MB dye adsorption. The experimental kinetic data were examined, and the pseudo-second-order model was shown to be more suitable for describing the adsorption process, with an excellent determination coefficient (R2 = 0.999). The exchanged standard enthalpy (H° = −22.456 KJ.mol−1) was calculated using the van ‘t Hoff equation, and it was proven that the adsorption process was exothermic. The spontaneous nature and feasibility of the MB dye adsorption process on WNSp were validated by negative standard enthalpy values (G°) ranging from −2.580 to −0.469 at different temperatures. It was established that WNSp may be employed as a novel, effective, low-cost adsorbent for the elimination of methylene blue in aqueous solutions.