Prediction of lead (Pb) adsorption on attapulgite clay using the feasibility of data intelligence models.

This study investigates the performance of support vector machine (SVM), multivariate adaptive regression spline (MARS), and random forest (RF) models for predicting the lead (Pb) adsorption by attapulgite clay. Models are constructed using batch stochastic data of heavy metal (HM) concentrations under different physicochemical conditions. Implementation of auto-hyper-parameter tuning using grid-search approach and comparative analysis is performed against the benchmark artificial intelligence (AI) models. Models are constructed based on Pb concentration (IC), the dosage of attapulgite clay (dose), contact time (CT), pH, and NaNO3 (SN). Principle component analysis (PCA) and correlation analysis (CA) methods are integrated to assess the importance of the applied predictors and their relationship with the target. Research findings approved the potential of the grid-RF model as a marginal superior predictive model against the grid-SVM in terms of MAE, i.e., 3.29 and 3.34, respectively; moreover, the md scored the same, i.e., 0.93, which reveals the potential predictability for both. Nonetheless, grid-MARS and standalone MARS models remained likewise in their predictability. IC parameter demonstrated the highest influential among all the predictors with the highest value of importance in the case of all three evaluators. The solution pH and dose stands together with marginal differences in case of PCA method; however, solution pH and CT appeared with similarity impact using the PCA method.

Random forest, support vector machine, and neural networks to modelling suspended sediment in Tigris River-Baghdad.

Suspended sediment is one of the most influential parameters on the water bodies' pollution. It can carry different pollutants with different concentration through the suspension movement in the flow. Therefore, it is of utmost importance to monitoring or modelling these loads so that an accurate sediment reduction strategy can be adopted. However, the monitoring process is laborious and time-consuming task. Thus, modelling is suggested as an alternative method. In this study, three different methods of artificial intelligence (i.e., random forest, support vector machine (Radial Basis Function), and artificial neural network) were employed to model and predict the suspended load at Sarai Station in Baghdad. To this end, observed flow rate (m3/s) and the corresponding suspended sediment concentration (mg/l) measured over the periods 1962-1981 and 2000-2010 were collected. Auto and partial correlation was used to identify the best combinations of input model data. The data was randomly partitioned into 75% for training and 25% for validation. The confidence interval was hypothesized to assess the uncertainty in the observed and predicted data. Whereas, the k-fold cross validation was used to quantify the uncertainty in the modelling results. The predictive modelling results for the three evaluated methods were assessed based on R2, RMSE, and NSE coefficient. Results show that random forest has the superior performance among the others. The total suspended sediment transported was estimated to be 72,734,852 ton during the period 2000-2010.