Applications of machine learning algorithms (MLAs) to modeling the adsorption efficiencies of different heavy metals have been limited by the adsorbate–adsorbent pair and the selection of specific MLAs. In the current study, adsorption efficiencies of fourteen heavy metal–adsorbent (HM-AD) pairs were modeled with a variety of ML models such as support vector regression with polynomial and radial basis function kernels, random forest (RF), stochastic gradient boosting, and bayesian additive regression tree (BART). The wet experiment-based actual measurements were supplemented with synthetic data samples. The first batch of dry experiments was performed to model the removal efficiency of an HM with a specific AD. The ML modeling was then implemented on the whole dataset to develop a generalized model. A ten-fold cross-validation method was used for the model selection, while the comparative performance of the MLAs was evaluated with statistical metrics comprising Spearman’s rank correlation coefficient, coefficient of determination (R²), mean absolute error, and root-mean-squared-error. The regression tree methods, BART, and RF demonstrated the most robust and optimum performance with 0.96 ⫹ R² ⫹ 0.99. The current study provides a generalized methodology to implement ML in modeling the efficiency of not only a specific adsorption process but also a group of comparable processes involving multiple HM-AD pairs.

Accuracy of prediction of pressure losses plays a vital role in the design of multiphase flow systems. The present study focused on the development of a computational fluid dynamics model to predict these parameters conveniently and accurately. The main objective was to validate the developed model through comparison of its simulation results with existing experimental data and empirical correlations. Both pipeline and annular geometries were considered for validation. Several data sets that involved a significant variation in process conditions were used for the validation. All three phases—liquid (water), gas (air), and solid (sand)—were taken into account. The simulations were conducted using the workbench platform of ANSYS Fluent 16.2. The Eulerian model of multiphase flow and the Reynolds stress model of turbulence closure available in Fluent were used for the present study. The average velocities and volumetric concentrations of involved phases were specified as the inlet boundary conditions. The stationary surfaces of the flow channels were hydrodynamically considered as either smooth or rough walls, and the outlets were regarded as being open to the atmosphere. The simulation results of pressure loss showed a good agreement with the corresponding measured values as well as with the predictions of well-established correlations.

The traditional procedure of predicting the settling velocity of a spherical particle is inconvenient as it involves iterations, complex correlations, and an unpredictable degree of uncertainty. The limitations can be addressed efficiently with artificial intelligence-based machine-learning algorithms (MLAs). The limited number of isolated studies conducted to date were constricted to specific fluid rheology, a particular MLA, and insufficient data. In the current study, the generalized application of ML was comprehensively investigated for Newtonian and three varieties of non-Newtonian fluids such as Power-law, Bingham, and Herschel Bulkley. A diverse set of nine MLAs were trained and tested using a large dataset of 967 samples. The ranges of generalized particle Reynolds number (Re_G) and drag coefficient (C_D) for the dataset were 10⁻³ < Re_G (-) < 10⁴ and 10⁻¹ < C_D (-) < 10⁵, respectively. The performances of the models were statistically evaluated using an evaluation metric of the coefficient-of-determination (R²), root-mean-square-error (RMSE), mean-squared-error (MSE), and mean-absolute-error (MAE). The support vector regression with polynomial kernel demonstrated the optimum performance with R² = 0.92, RMSE = 0.066, MSE = 0.0044, and MAE = 0.044. Its generalization capability was validated using the ten-fold-cross-validation technique, leave-one-feature-out experiment, and leave-one-data-set-out validation. The outcome of the current investigation was a generalized approach to modeling the settling velocity.

The continuous growth in population, urbanization, and industrial development has been increasing the generation of solid waste (SW) in the Kingdom of Saudi Arabia. Consequently, the associated greenhouse gas (GHG) emission is also following an increasing trend. The collection and use of greenhouse gases emitted from solid waste management practices are still limited. A causality analysis examined the driving factors of the emissions from solid waste management. The methane (CH₄) emissions from municipal solid waste (MSW) increased with an increase in gross domestic product (GDP) per capita and urban population, and an increase in foreign direct investment (FDI) inflows and literacy rate was likely to reduce CH₄ emissions from municipal solid waste and vice versa. The CH₄ emission generated from industrial solid wastes was found to be positively related to GDP per capita, urban population, and FDI inflows. However, a decrease in the unemployment rate was likely to increase CH₄ emissions from industrial solid wastes. The future greenhouse gas emissions were projected under different possible socio-economic conditions. The scenario analysis based on different variations of population and GDP growth revealed that methane emission from total waste would increase at an average annual rate of 5.13% between 2020 and 2050, and is projected to reach about 4000 Gg by the end of the year 2050. Although the Kingdom has been taking some initiatives towards climate change mitigation, it has significant opportunities to adopt some of the best practices in solid waste management including reduction, recycling, composting and waste-to-energy, and carbon capture and utilization. This study also put emphasis on developing appropriate policy approaches for climate change mitigation based on the circular economy which is gaining momentum in the Kingdom.

Arsenic is a deleterious heavy metal that is usually removed from polluted water based on adsorption processes. The latest mode of modeling such a process is to implement artificial intelligence (AI). In the current work, a new artificial neural network (ANN) model was developed to predict the adsorption efficiency of arsenate (As(III)) from contaminated water by analyzing different architectures of an adaptive network-based fuzzy inference system (ANFIS). The database for the current study consisted of the experimental data of the adsorption of As(III) by different adsorbents/biosorbents. The data were randomly divided into two sets: 70% for the training phase and 30% for the testing phase. Four statistical evaluation metrics, namely, mean square error (MSE), root-mean-square error (RMSE), Pearson’s correlation coefficient (R%), and the determination coefficient (R²) were used for the analysis. The best performing ANFIS model was characterized with the average values of 97.72%, 0.9333, 0.137, and 0.274 of R%, R², MSE, and RMSE, respectively. In addition, a parametric investigation revealed that the most dominating parameters on the adsorption process efficiency were in the following order: pH, As initial concentration, contact time, adsorbent dosage, inoculum size, and temperature. The results of the current study would be useful in the adsorption process scale-up and optimization.

The accurate prediction of terminal settling velocity of solid spheres in non-Newtonian liquids is important for various fluid-particle systems such as slurry pipelines, separation processes, hole-cleaning in drilling operations, and mineral processing. The standard practice for the prediction involves an implicit procedure that requires repeated iterations using Newtonian correlations. Wilson et al. developed an explicit method that allows direct (noniterative) prediction of the velocity in non-Newtonian liquids. Although very useful, the original Wilson model has an empirical constraint that limits its application. In this study, experiments are performed to measure the terminal settling velocity of precision spheres in Newtonian liquid (water) and non-Newtonian drilling fluids (Flowzan solutions). The Herschel–Bulkley three parameter model satisfactorily modeled the non-Newtonian rheology. Experimental data and similar measurements available in the literature are presented in this paper. The data exhibited the standard relationship between the drag coefficient and the Reynolds number. The original Wilson model was tested for these data points and was modified in this study to address its limitations. Consequently, it was observed that the modified version yielded more accurate results than the original model. Its prediction was especially better when the value of corresponding Reynolds number was more than 10.