Comparative Modeling and Analysis of the Mass Transfer Coefficient in a Turbulent Bed Contactor using Artificial Neural Network and Adaptive Neuro-Fuzzy Inference Systems

This study deals with predicting the gas film volumetric mass transfer coefficient (K_oga) in a turbulent bed contactor (TBC), using both artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) techniques. The networks have been trained and evaluated with the experimental data available in the literature. Input variables to the networks are process variables such as gas and liquid phase concentration, gas and liquid superficial velocities and also specific area of packings. The results obtained the ability of developed ANN and ANFIS for prediction of K_oga. Although it was observed that both ANN and ANFIS models provided a good statistical prediction in terms of coefficient of determination (R²), mean relative error (MRE) and root mean square error (RMSE), the accuracies of ANFIS predictions were better than those of ANN predictions.     

Modelling of adsorption of nickel (II) by blend hydrogels (cellulose nanocrystals and corn starch) from aqueous solution using adaptive neuro-fuzzy inference systems (ANFIS) and artificial neural networks (ANN)

The current study looks at the effectiveness of the removal of nickel (II) from aqueous solution using an adsorption method in a laboratory-size reactor. An artificial neural network (ANN) and an adaptive neuro-fuzzy inference system (ANFIS) were used in this study to predict blend hydrogels adsorption potential in the removal of nickel (II) from aqueous solution. Four operational variables, including initial Ni (II) concentration (mg/L), pH, contact duration (min), and adsorbent dose (mg/L), were used as an input with removal percentage (%) as the only output; they were studied to assess their impact on the adsorption study in the ANFIS model. In contrast, 70% of the data was used for training, while 15% of the data was used in testing and validation to build the ANN model. Feedforward propagation with the Levenberg–Marquardt algorithm was employed to train the network. The use of ANN and ANFIS models for experiments was used to optimize, construct, and develop prediction models for Ni (II) adsorption using blend hydrogels. The adsorption isotherm and kinetic models were also used to describe the process. The results show that ANN and ANFIS models are promising prediction approaches that can be applied to successfully predict metal ions adsorption. According to this finding, the root mean square errors (RMSE), absolute average relative errors (AARE), average relative errors (ARE), mean squared deviation (MSE), and R² for Ni (II) in the training dataset were 0.061, 0.078, 0.017, 0.019, and 0.986, respectively, for ANN. In the ANFIS model, the RMSE, AARE, ARE, MSE, and R² were 0.0129, 0.0119, 0.028, 0.030, and 0.995, respectively. The adsorption process was spontaneous and well explained by the Langmuir model, and chemisorption was the primary control. The morphology, functional groups, thermal characteristics, and crystallinity of blend hydrogels were all assessed.     

Heat transfer and fluid flow modeling in serpentine microtubes using adaptive neuro-fuzzy approach

An adaptive neuro-fuzzy inference system (ANFIS) is applied to predict thermal and flow characteristics in serpentine microtubes. Heat transfer rate and pressure drop were experimentally measured for six serpentine microtubes with different geometrical parameters. Thermal and flow characteristics were obtained in various flow conditions. The ANFIS models were trained using the experimental data to predict Nusselt number (Nu) and friction factor (f) in the studied serpentine microtubes as a function of geometric parameters and flow conditions. The model was validated through testing data set, which were not previously introduced to the developed ANFIS. For Nu prediction, the root mean square error (RMSE), mean relative error (MRE), and absolute fraction of variance (R²) between the predicted results and experimental data were found 0.2058, 1.74%, and 0.9987, respectively. The corresponding calculated values for f were 0.0056, 2.98%, and 0.9981, respectively. The prediction accuracy of the ANFIS models was compared with that of corresponding classical power-law correlations and its advantages are illustrated.     

Development of an artificial neural network correlation for prediction of overall gas holdup in bubble column reactors

In the literature, several correlations have been proposed for gas holdup prediction in bubble columns. However, these correlations fail to predict gas holdup over a wide range of conditions. Based on a databank of around 3500 measurements collected from the open literature, a correlation for gas holdup was derived using a combination of Dimensional Analysis and artificial neural network (ANN) modeling. The overall gas holdup was found to be a function of four dimensionless groups: Re_g, Fr_g, Eo/Mo, and ρ_g/ρ_L. Statistical analysis showed that the proposed correlation has an average absolute relative error (AARE) of 15% and a standard deviation of 14%. A comparison with selected correlations in the literature showed that the developed ANN correlation noticeably improved prediction of overall gas holdup. The developed correlation also shows better prediction over a wide range of operating conditions, physical properties, and column diameters, and it predicts properly the trend of the effect of the operating and design parameters on overall gas holdup.     

Determination of surface area of carbon-black by simple cyclic-voltammetry measurements in aqueous H2SO4

Determination of the surface area of commercial carbon-black (CB) by cyclic-voltammetry (CV) measurements of the electrochemical double-layer charge (Q) in aqueous sulfuric acid was investigated. Various factors that affect the Q value associated with CB, including: the presence of redox-reversible function-groups, the binders used for the formation of thin-film CB electrodes, the scan rates of the CV measurement, H₂SO₄ electrolyte concentration and the volume of air contained in the pores of the CB samples were examined. Conditions for measuring Q without interference from these factors were investigated and the data derived, without interference, was found to correlate well with the surface area of the CB. The results show that the total Brunauer–Emmett–Teller (BET) surface area shows good correlation (R² = 0.993) with the Q value corresponding to a full charge/discharge of the CB (Q₀), obtained by extrapolation at a zero scan rate; additionally, the CB micropore surface area (diameter < 2 nm) shows good linear correlation with the Q deficiency (Q₀–Q), measured at 5 mV s^?1 (R² = 0.998).     

A Model for the Prediction of the Collection Efficiency Characteristics of a Small,Cylindrical Aerosol Sampling Cyclone

Velocity data from a previous study were nondimensionalized and used in conjunction with a computer program which solves the equations for particle trajectory to predict the collection efficiency for the cyclone. Results for the prediction of cutpoint at the same Reynolds number as that for which the velocities were measured, both for a large cyclone of 88.9 mm diameter and another geometrically similar at one half the scale, are excellent. The model predicts cutpoints of 10 μm and 5.1 μm for the large and small cyclone, respectively, while the actual cutpoints determined from aerosol tests were 9.9 μm and 5.2 μ m. The efficiency curve generated by the model was steeper (geometric standard deviation of 1.1) than the efficiency curve determined through the aerosol testing (geometric standard deviation of 1.4). A simplification of the Dirgo and Leith equation fitting Barth's design curve is suggested which provides a significantly better fit of the aerosol data (geometric standard deviation of 1.3). At 1.5N _{R Q}, where N _{R Q} = (4pg)/(πμD _c), the error in prediction of the cutpoint in the large cyclone is less than 8% while at 04N _{R Q} the error is less than 2%. Although results are good over a limited range of Reynolds numbers, the model is strictly applicable only for flows which are dynamically similar to those studied here.     

Crystallization behavior and melt structure of typical CaO–SiO2 and CaO–Al2O3-Based mold fluxes

Crystallization behavior and melt structure of two typical mold fluxes A (CaO–SiO₂-based) and B (CaO–Al₂O₃-based) for casting high-aluminum steel were investigated using double hot thermocouple technology (DHTT), X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy. The results suggest that the crystallization temperature of Flux B is higher, and its crystallization incubation time is shorter compared with Flux A. The precipitated phase in Flux A is CaSiO₃, whereas BaAl₂O₄ and Ca₂Al₂SiO₇ form in Flux B. The structure analyses suggest that the degree of polymerization of Flux A is larger than that of Flux B. In addition, the major structural units of Flux A are Si–O–Si, Q₀^Si, Q₁^Si, Q₂^Si and Q₃^Si, but those of Flux B are mainly aluminate (Al–O–Al, Al–O^-), aluminosilicate (Al–O–Si) and silicate units (Q₀^Si, Q₁^Si, Q₂^Si and Q₃^Si). These different melt structures are the main reasons why the precipitated phases in these two mold fluxes are different, and the crystallization ability of Flux A is weaker than Flux B.     

Determination of EMC isotherms and appropriate mathematical models for canola

Canola seed needs to be dried to minimize damage during subsequent unit operations. In order to optimize the drying and storage processes of seeds it is necessary to know the equilibrium moisture content at different air equilibrium relative humidities and temperatures. In the present work adsorption equilibrium moisture content isotherms were determined for a specific local canola variety at 25, 40 and 55 °C and seven air relative humidifies within the range of 11-90%. Experimental data were used to model the adsorption isotherm process using non-linear regression analysis. Thirteen available mathematical and semi empirical models were employed to find the best fit isotherm curve model. The Halsey model showed the best results at 25 and 40 °C and the values of coefficient of determination (R²), chi-square (χ²) and root of mean square error (RMSE) were 0.993, 0.122 and 0.295 at 25 °C and 0.994, 0.042 and 0.174 at 40 °C. Another best fit mode was the GAB model at 55 °C which resulted in the values of 0.997, 0.023 and 0.120 for R², χ² and RMSE respectively. The adsorption monolayer moisture content (m₀) was also evaluated using BET equation. The m₀ values at 25 °C, 40 °C and 55 °C were 0.017, 0.016 and 0.015 g H₂O/g solid and the corresponding constant values of the BET equation were found to be −4.733, −5.129 and −10.299 respectively.     

Water recovery from humidified waste gas streams: Quality control using membrane condenser technology

In this work “membrane condenser” is utilized for the selective recovery of evaporated waste water from industrial gases and for the control of the composition of the recovered liquid water. A simulation study of the process has been developed for predicting the membrane-based process performance. The achieved results indicate that feed flow rate (Q^Feed), interfacial membrane area (A^Membrane), the ratio Q^Feed/A^Membrane, the temperature difference (ΔT) between the fed flue gas T_feed and the membrane module are the parameters controlling the process. In particular, the amount of recovered water rises at increasing ΔT and Q^Feed/A^Membrane, whereas its quality is made worse at increasing ΔT and Q^Feed. The data obtained have been also supported by an experimental study of the process, confirming the validity of the simulation study and its suitability for a screening of the operative conditions to be chosen in a membrane condenser.     

Scaling laws and internal structure for characterizing electrospun poly[(R)-3-hydroxybutyrate] fibers

Using chloroform/dimethylformamide (CF/DMF) co-solvent, electrospinning of poly[(R)-3-hydroxybutyrate] (PHB) solutions was carried out at ambient temperature. The effects of the applied voltage (V), flow-rate (Q), and solution viscoelastic properties on the Taylor cone, electrified jet, and fiber morphology were investigated. In addition, the electric field developed by the needle-plate electrode configuration was calculated using a finite element analysis to reveal the tip-to-collector (H) effect. Among the processing parameters (V, Q and H), it was found that Q played a key role in determining the jet diameter (d_j) and electrospun fiber diameter (d_f), and scaling laws existed between them, i.e., d_j-Q^0.61 and d_f-Q^0.33. The diameter reduction ratios of D_o/d_j (D_o is the needle diameter) and d_j/d_f were measured as 50-120 and 5-10, respectively; it suggested that major jet stretching took place in the straight electrified jet region, and further chain orientation could be gained by the subsequent process of jet whipping. By changing PHB concentrations from 5 to 15 wt%, the solution viscosity (η_o) was increased from 100 to 4900 cP, whereas the surface tension and solution conductivity remained unchanged; it provided a good model solution to exclusively reveal the η_o effect on the electrospinning process. Our results showed that the η_o-dependence of d_j and d_f also followed simple scaling laws: d_j-η_o^0.06, and d_f-η_o^0.39, with a prefactor depending on the processing variables, mainly the flow-rate. Regardless of the PHB concentrations used, the obtained PHB fibers showed a similar crystallinity fraction of ca. 0.63 and possession of major α-crystals together with a small amount of β-crystals with zigzag chain conformation.     

Development of support vector regression (SVR)-based correlation for prediction of overall gas hold-up in bubble column reactors for various gas-liquid systems

The objective of this study was to develop a unified data-driven correlation for the overall gas hold-up for various gas-liquid systems using support vector regression (SVR)-based modeling technique. Over the years, researchers have amply quantified the hydrodynamics of bubble column reactors in terms of the overall gas hold-up. In this work, about 1810 experimental points were collected from 40 open sources spanning the years 1965-2007. The model for overall gas hold-up was established as a function of several parameters which include superficial gas velocity, superficial liquid velocity, gas density, molecular weight of gas, sparger type, sparger hole diameter, number of sparger holes, liquid viscosity, liquid density, liquid surface tension, operating temperature, operating pressure and column diameter of the gas-liquid system. For understanding the hold-up behavior, the data used for training the model was grouped into various gas-liquid systems viz., air-water, gas-aqueous viscous liquids, gas-organic liquids, gas-aqueous electrolyte solutions and gas-liquid systems operated over a wide range of pressure. A generalized model established using SVR was evaluated for its performance for various gas-liquid systems. Statistical analysis showed that the proposed generalized SVR-based correlation for overall gas hold-up has prediction accuracy of 97% with average absolute relative error (% AARE) of 12.11%. A comparison of this correlation with the selected system specific correlations in the literature showed that the developed SVR-based correlation significantly gives enhanced prediction of overall gas hold-up.     

Effects of Ni2+ substitution on the crystal structure,bond valence,and microwave dielectric properties of BaAl2–2xNi2xSi2O8–x ceramics

BaAl_2?2xNi_2xSi₂O_8?x (x = 0, 0.005, 0.01, 0.02, 0.03) ceramics were prepared using traditional solid phase reaction method. The microwave dielectric properties, including permittivity (ε_r), quality factor (Q × f), and temperature coefficient of resonant frequency (τ_f), were discussed based on the bond valence theory. The first-principle calculation was adopted to determine the site (Ba, Al, and Si) where doping element (Ni²⁺) would be inclined to occupy. The substitution of Ni²⁺ for Al³⁺ contributed to the breaking of Al-O and Si-O bonds and then facilitated the BaAl₂Si₂O₈ (BAS) hexacelsian-celsian transformation. Moreover, this substitution could change the bond strength between cation and oxygen anion due to the variation of the bond valence, which reasonably explained the variation of ε_r, Q × f, and τ_f values. Well-sintered and completely transformed celsian ceramics can be obtained after doping with Ni²⁺. When x = 0.01, compact BaAl_1.98Ni_0.02Si₂O_7.99 ceramic exhibited highly promising microwave dielectric properties: ε_r = 6.89, Q × f = 53, 287 GHz and τ_f = -25.31 × 10^?6 /°C.     

Rapid Determination of Free Fatty Acid in Extra Virgin Olive Oil by Raman Spectroscopy and Multivariate Analysis

We introduce a visible Raman spectroscopic method for determining the free fatty acid (FFA) content of extra virgin olive oil with the aid of multivariate analysis. Oleic acid was used to increase the FFA content in extra virgin olive oil up to 0.80% in order to extend the calibration span. For calibration purposes, titration was carried out to determine the concentration of FFA for the investigated oil samples. As calibration model for the FFA content (FFA%), a partial least squares (PLS) regression was applied. The accuracy of the Raman calibration model was estimated using the root mean square error (RMSE) of calibration and validation and the correlation coefficient (R ²) between actual and predicted values. The calibration curve of actual FFA% obtained by titration versus predicted values based on Raman spectra was established for different spectral regions. The spectral window (945–1600 cm⁻¹), which includes carotenoid bands, was found to be a useful fingerprint region being statistically significant for the prediction of the FFA%. High R ² and small RMSE values for calibration and validation could be obtained, respectively.     

Electrophoresis in suspensions of charged porous spheres in salt-free media

Electrophoresis in either dilute or concentrated suspensions of charged porous spheres in salt-free media is investigated theoretically in this study. The Brinkman model and the Kuwabara’s unit cell model are adopted to simulate the porous structure and the suspensions, respectively.We found, among other things, that the polarization effect due to the convection flow within the porous sphere is a crucial factor in determining its electrophoretic behavior. An induced electric field opposite to the applied electric field is generated, which deters the particle motion significantly when the particle is highly permeable. Approximate analytical prediction for dilute suspensions neglecting convection flow can overestimate the mobility severely in this situation. The approximate analytical prediction is satisfactory when the permeability of particle is low, though. Counterion condensation happens at high fixed charge density which decreases the mobility drastically and the mobility approaches a constant value asymptotically. The mobility profile of the particles with increasing volume fraction can exhibit local minimum if the corresponding dimensionless parameter Q_fix/(λa)² is high, where Q_fix and λa are, respectively, the fixed charge density and the friction coefficient of the porous particles in dimensionless form. This is due to the overlapping of counterion clouds surrounding particles, which offsets the polarization effect, becomes significant as the suspension gets concentrated. No such phenomenon for low Q_fix/(λa)², where the mobility profile decreases monotonously with increasing volume fraction. Comparison with experimental data available in the literature for polyelectrolyte suspensions is excellent, indicating the reliability of this analysis, as well as the success of using charged porous sphere to model a polyelectrolyte system.     

Heat transfer enhancement of serpentine channels with twisted tape insert by computational fluid dynamics and artificial intelligence

Improving the thermal–hydraulic performance of heat exchangers is a critical challenge. In this study, we investigated the impact of twisted tape insert geometrical variables, namely pitch length (P) and diameter (W), on the Nusselt number (Nu) and friction factor (f) within serpentine channels. To accomplish this, we employed computational fluid dynamics (CFD) as our simulation technique, with water as the working fluid. The validity of the CFD data was confirmed, leading to the development of artificial intelligence (AI) subsets: artificial neural network (ANN) and genetic algorithm (GA). In the prediction models, the following input variables were considered: Reynolds number (Re), (P/D_h), and (W/D_h). The ANN and GA models achieved a mean relative error (MRE) of 0.125% and 1.326% for Nu, and 0.139% and 4.104% for f, respectively. These results demonstrate the high accuracy of the correlations, with the ANN model showcasing superior performance.     

Maize [<Emphasis Type="Italic">Zea Mays (</Emphasis>L.)] crop-nutrient response functions extrapolation for Sub-Saharan Africa

There is a need for methods to scale up site specific field trial results for maize to larger areas. The objectives of this research for SSA were to establish relationships between maize crop-nutrient response functions and biophysical variables; determine prediction equations for maize crop-nutrient response functions; and extrapolate (predict) maize nutrient response functions using the predictor equations to areas of interest and evaluate the goodness-of-fit of the data. Geo-referenced maize-nutrient response functions of 736, 488, and 152 for N, P and K, respectively, determined from past and recent research results by the project OFRA were used. New geo-referenced 4646 points were identified across SSA to predict crop-nutrient response functions using the prediction equations. The independent variables considered were elevation, location, climate and soil properties and their square and two-way interactions. Data were subjected to GLM at P ≤ 0.05. The coefficients of maize nutrient-response functions model’s goodness-of-fits were evaluated using coefficient of determination (R²), cross-validated R² (q²), R _o ² and R _o ^′2 and RMSE with mean values of 0.67, 0.65, 0.67, 0.68 and 0.16 Mg ha^?1, respectively. These values indicate the robustness and predictive ability of the predictive models for the study area conditions. In conclusion, these predictive equations can be used to estimate maize nutrient response functions for important maize growing areas throughout SSA.     

AN EXPERIMENTAL PHASE DOPPLER ANEMOMETER STUDY OF ELECTROHYDRODYNAMIC ATOMIZATION

In this work, measurements of the size of the primary droplets generated by electrospraying 10 different liquids with different densities and surface tensions (measured by a tensiometer, Lauda LD1), viscosities, electrical conductivities (WTW LF530) and permittivities (capacitance bridge, Philips) were carried out. To measure the size of the droplets, a Phase Doppler Anemometer (Dantec/INVENT) was used. Gañán-Calvo et al. (1997) obtained on the basis of a theoretical model the so-called scaling laws, which permit the prediction of the spray current and droplet size. The empirical constants in these formulas were obtained from the experimental measurements. Two different behaviors were found, depending on the dimensionless parameter δ_μδ^1/3=(ε²₀γ²/κ²μ³Q)^1/3. From our measurements, an intermediate regime with a much smaller slope is also found. This might also be due to the small flow rate, as the measurements which lie early below Gañán-Calvo’s fitting curve all satisfy the relation (Q/(ε_r−1)^1/2Q₀)^1/3<3.     

Gas-liquid-liquid extraction in a novel rotating microchannel extractor

In this work, a novel rotating microchannel extractor (RME) is designed and further used for the extraction of chromium (III) from water. Unexpectedly, the micro-extraction had the same effect as carrying out 2.9-stage cross-flow extractions. Various factors, including the gas intake methods, gas intake quantity (Q_g), distance between inner rotor and outer wall (D), rotational inner rotor speed (R) and volumetric flow rate (Q_a, Q_o), were selected to investigate their effect on the extraction efficiency (η) thoroughly. The relation map of η with We_a and We_o−g for RME provides a comprehension for the gas–liquid–liquid extraction process in this RME system.