Modeling dynamic viscosity of n-alkanes using LSSVM technique

One of the important thermophysical properties is viscosity which expresses the resistance of fluid to flow. The least squares support vector machine (LSSVM) algorithm is proposed as a novel method for prediction of dynamic viscosity of different normal alkanes in a wide range of pressure and temperature. As this study is purely computational, 228 experimental data points were gathered from literature for training and validation of the model. The outcomes of the LSSVM algorithm were compared with the actual data with acceptable average absolute relative deviation and the coefficient of determination (R²) of 1.014 and 0.9968, respectively. The comparisons showed that the predicting model has the potential of prediction of n-alkane dynamic viscosity in terms of pressure, temperature, and carbon number of n-alkane, so this strategy can be used as a simple tool for predicting the behavior of reservoir fluids.     

CSA-LSSVM model for the estimation of solubility of n-alkane in supercritical CO2

Poor solubility of substantial hydrocarbons in CO₂ has constrained the use of CO₂-EOR (enhanced oil recovery) in the modern oil recovery industry to some extent. Subsequently, it is significant to research the solubility regularity of various hydrocarbons in supercritical carbon dioxide (scCO₂) in the first place. CO₂ injection as one of the popular methodologies in light of financially and environmentally friendly has wide applications in EOR. In this paper, our objective is to estimate the solubility of n-alkanes in scCO₂. This study highlights the application of a model based on least square support vector machine for estimation of solubility of n-alkanes in scCO₂. The tuning parameters of the developed model were determined by an optimization algorithm, namely coupled simulated annealing. A set of 184 data points of solubility was used to execute the new model. To assess the accuracy and effectiveness of the developed model for prediction of experimental data, statistical and graphical techniques were used. Moreover, the outcomes were compared with the results of literature correlations to predict the solubility of alkanes. Results demonstrate that the model is precise and viable for prediction of solubility data. The resulted values of R², root-mean-square error, SD, and % average absolute relative deviation for total data points are 0.99204, 0.12862, 0.6437, and 0.7753 for overall data, respectively.     

Comprehensive chemical kinetic modeling of the oxidation of 2-methylalkanes from C7 to C20

Conventional petroleum jet and diesel fuels, as well as alternative Fischer–Tropsch (FT) fuels and hydrotreated renewable jet (HRJ) fuels, contain high molecular weight lightly branched alkanes (i.e., methylalkanes) and straight chain alkanes (n-alkanes). Improving the combustion of these fuels in practical applications requires a fundamental understanding of large hydrocarbon combustion chemistry. This research project presents a detailed and reduced chemical kinetic mechanism for singly methylated iso-alkanes (i.e., 2-methylalkanes) ranging from C₇ to C₂₀. The mechanism also includes an updated version of our previously published C₈–C₁₆n-alkanes model. The complete detailed mechanism contains approximately 7200 species 31400 reactions. The proposed model is validated against new experimental data from a variety of fundamental combustion devices including premixed and non-premixed flames, perfectly stirred reactors and shock tubes. This new model is used to show how the presence of a methyl branch affects important combustion properties such as laminar flame propagation, ignition, and species formation.     

Analysis of equilibrium CO2 solubility in aqueous APDA and its potential blends with AMP/MDEA for postcombustion CO2 capture

The utility of polyamine-based solvent-activators for the possible application in postcombustion CO₂ capture technology has drawn considerable attention recently owing to its higher loading capacity as well as superior kinetics. The current work involves a comprehensive experimental cum theoretical investigation on the equilibrium solubility of CO₂ pertaining to aqueous N-(3-aminopropyl)-1,3-propanediamine and its blends with N-methyldiethanolamine and 2-amino-2-methyl-1-propanol. The analysis was conducted within the operating temperature and CO₂ partial pressure range of 303.2-323.2 K and 2-200 kPa, respectively. Two different mathematical models based on nonrigorous approaches such as equilibrium based modified Kent-Eisenberg (KE) model and a multilayer feedforward neural network model have been developed to correlate the CO₂ solubility data over a wide range of experimental conditions. Both the model predictions are well-validated with the experimental results. The reaction scheme as well as the prevalence of important reaction products was further confirmed with qualitative ¹³C NMR as well as ATR-FTIR analysis. Apart from these some of the important thermally induced transport properties viz, density, viscosity, and surface tension of the aqueous single and blended systems were measured and correlated with various consistent empirical models such as Redlich-Kister and Grunberg-Nissan model while surface tension data are modeled using temperature-based multiple linear regression technique.     

Optimization of Process Parameters for the Preparation of CaO/CaSO4/Coal Fly Ash Sorbent for Sulfur Dioxide (SO2) Removal: Part II

A two-step optimization strategy was employed to optimize the surface area of sorbent prepared from coal fly ash, calcium oxide (CaO) and calcium sulfate (CaSO₄) for flue gas desulfurization. In the first step, a 3 level full factorial design of experiment was used to develop a regression model equation to correlate the significant experimental sorbent preparation variables to the surface area of the resulting sorbent. The three experimental sorbent preparation variables studied are hydration period (x ₁), ratio of CaO to fly ash (x ₂) and amount of CaSO₄ (x ₃). In the subsequent step, response surface methodology was used to identify the experimental sorbent preparation variables that maximize the surface area of the sorbent. Through this two-step optimization strategy, it was found that at a hydration period of 10 hrs and drying temperature of 100°C, optimum surface area of 67.0 m²/g could be attained by using 5 grams of CaO, 13.7 grams of fly ash, and 7.4 grams of CaSO₄ in the preparation mixture. The prediction was verified with experimental runs.     

Machine learning approach to predict the thermal performance of closed-loop thermosyphon

This paper presents the machine learning (ML) algorithm to predict the thermal performance of closed-loop thermosyphon (CLT). The experimentation is carried out on the acetone-charged CLT at different test conditions such as heat inputs, filling ratios, and adiabatic lengths. The test data is used to calculate the performance parameters such as thermal resistance, heat transfer coefficient, and effectiveness of the system. Based on the experimental dataset, the ML algorithms are developed to predict the performance parameters of the CLT system. The ML algorithms such as linear regression, decision tree (DT), random forest (RF), and lasso regression are used for the development of the prediction model. The hyperparameters are well-tuned and optimized. The prediction measuring parameters (mean absolute error, R²) are analyzed carefully. It is noticed that the DT model outperformed the prediction of the other used models. The R² score of the DT model was 98.504; whereas, the R² scores of the RF model and linear regression model were about 94.76 and 92.17, respectively. This study will become a roadmap to the ML approach in the thermosyphon system.     

Computational research on rotating channel by a modified anisotropic turbulence model

Based on the simplified format of the Reynolds stress equations,a fire-new rotational-modification method for the anisotropic turbulence model has been presented.A three-dimensional Navier-Stokes code with this new rotational modified k-ω turbulence models(β=0.1 and β=1) and the standard k-ω turbulence model have been used for the prediction of flow and heat transfer characteristics in a rotating smooth square channel.The Reynolds number Re based on the inlet velocity of the cooling air and hydraulic diameter is 6000.The rotating speed are 300,600,900,1200rpm respectively.The calculations results of using three turbulence models have been compared with the experimental data.The research results show that(1) the rotational modification coefficient Rf13 used in the new anisotropic k-ω model would increased/decreased the predictions of heat transfer on the trailing surface/leading surface compared to the standard k-ω model.And this tendency would be increased with the increased β.(2) The simulation performance of the standard k-ω model was well on the leading surface.However,on the trailing surface it under-predicted the heat transfer at high rotating speed.(3) The calculation results of the new anisotropic k-ω model with β=0.1 proposed by the present paper agreed well with experimental data,both on the leading and trailing surfaces.Besides,compared to 1,0.1 is a more appropriate magnitude of β at conditions in the present paper.     

Effect of microgravity on flow boiling heat transfer of liquid hydrogen in transportation pipes

The flow boiling phenomenon of liquid hydrogen (LH2) during transportation in microgravity is very different from that under terrestrial condition. In this study, a saturated flow boiling of LH2 in a horizontal tube has been simulated under microgravity condition using coupled level-set and volume of fluid method. The validation of the developed model shows good agreement with the experimental data from the literature. The changes of heat fluxes and pressure drops under different gravitational accelerations were analyzed. And, the variation of heat fluxes with different wall superheat and contact angle were compared between microgravity (10⁻⁴g) and normal gravity (1g) condition. Also, the influence of surface tension were studied under microgravity. The numerical results indicate that the heat flux decrease with the decrement of gravitational acceleration. And the heat transfer ratio decrease with the increment of wall superheat in the nucleate boiling regime. The heat transfer slightly reduce when considering surface tension. In addition, the changes of contact angle have a more significant impact on heat transfer under microgravity condition.     

Comparison of the different artificial neural networks in prediction of biomass gasification products

In this study, artificial neural networks (ANNs) and a nonlinear autoregressive exogenous (NARX) neural network model were employed in order to model a fixed bed downdraft gasification. The relation between the feature group and the regression performance was investigated. First, feature group consists of the equivalence ratio (ER), air flow rate (AF), and temperature distribution (T0‐T5) obtained from the fixed bed downdraft gasifiers, while the second group includes ultimate and proximate values of biomasses, ER, AF, and the reduction temperature (T0). Models constructed to predict the syngas composition (H₂, CO₂, CO, CH₄) and calorific value. Experimental gasification data that involve 3831 data samples that belong to pinecone and wood pellet were used for training the ANNs. Different ANN architecture and NARX time series model have been constructed to examine the prediction accuracy of the models. The results of the ANN models were consistent with the experimental data (R² > 0.99). The overall score of NARX time series networks is found to be higher than other architecture types. A successful method is proposed to reduce the number of features, and the effect of the features on the prediction capability was examined by calculating the relative importance index using the Garson's equation.     

A modified model for estimation of daylight factor for skylight integrated with dome roof structure of mud-house in New Delhi (India)

The daylight factor model given by Charted Institute of Building Services Engineers (CIBSE) was modified in this paper to incorporate time variations with respect to zenith angle (θ_z) and vertical height (h) of working surface above ground surface which was normalized with central height (H) of skylight dome. The modified model contains constant exponents which are determined using linear regression analysis based on hourly experimental data of inside and outside illuminance for each month of the year 2007–2008. The prediction of modified model is found in good agreement with experimental observed inside illuminance data on the basis of values of root mean square percentage error (e) and correlation coefficient (r). The annual average daylight factor values for big and small dome skylight rooms are determined as 2.3% and 4.4% respectively. The energy saving potential of skylight rooms for selected climatic locations in India is also presented in this paper.     

Prediction of Thermal Conductivity and Viscosity of Ionic Liquid-Based Nanofluids Using Adaptive Neuro Fuzzy Inference System

Nowadays, ionic liquid-based nanofluids are introduced as a new class of heat transfer fluids, which exhibit superior thermal properties compared to their base ionic liquids. Potential applications of these nanofluids make it necessary to know their thermophysical properties such as thermal conductivity and viscosity. Therefore, adaptive neuro fuzzy inference system (ANFIS) has been successfully developed to predict thermal conductivity and viscosity of ionic liquid-based nanofluids. The developed models have investigated the influence of temperature, nanoparticle concentration, and ionic liquid molecular weight on the thermophysical properties of nanofluids. After developing ANFIS structure, the capability and accuracy of the developed neuro fuzzy models have been evaluated by comparison of model predictions with experimental data extracted from the literature and calculation of statistical parameters such as coefficient of determination (R²) and average relative deviation (ARD). The ARD of ANFIS model in prediction of thermal conductivity of nanofluids is 0.72%, with a high R² of 0.9959. The values of ARD and R² for estimation of nanofluids viscosity are 5.1% and 0.9934, respectively, which indicates a satisfactory degree of accuracy for the proposed models.     

A New Inundation Correlation for the Prediction of Heat Transfer in Steam Condensers

A new correlation used to account for the inundation effect on the prediction of heat transfer between steam vapor and cooling water in tube-and-shell condensers is proposed in this work. The proposed correlation is validated by comparing the predicted results with the experimental data of a steam surface condenser. A modified k–ε turbulence model for gas–liquid two-phase flows with distributed flow resistance is used in the numerical simulation. The predicted results using the proposed correlation agree better with the experimental data than those obtained using the existing correlations for inundation.     

Application of ANFIS-PSO algorithm as a novel method for estimation of higher heating value of biomass

One of the important parameters in economic study of energy sources and bioenergy is higher heating value (HHV). In this investigation, adaptive neuro fuzzy inference system (ANFIS) was applied as a novel method to predict HHV of biomass in terms of fixed carbon (FC), ash content (ASH), and volatile matters (VMs). Due to the fact that experimental investigations are time- and cost-consuming, this investigation was selected purely computational and a total number of 350 experimental data were extracted from literature for different steps of modeling. The proposed algorithm was evaluated by statistical indexes such as coefficient of determination (R²), root mean squared error (RMSE), and average absolute relative deviation (AARD), which are 0.90757, 1.1792, and 5.266, respectively. The reported indexes showed that ANFIS-particle swarm optimization can be used as a novel computational approach for prediction of HHV as function of proximate analysis.     

Theoretical and experimental study on the transient adsorption characteristics of a vertical packed porous bed

This paper presents a theoretical and experimental study of the transient adsorption characteristics of vertical packed porous bed. The theoretical model describes the effect of independent parameters (time and vertical distance through the bed) on the vertical gradient of adsorbable fluid in the bed. A simplified analytical solution, for specific operating conditions, is also presented. In the experimental study, porous granules of burned clay are applied as a desiccant carrier in the fixed bed. The granules of the packed bed are impregnated with liquid calcium chloride solution to form the porous adsorbing surface. The isothermal adsorption of water vapour from atmospheric air using the prepared bed is experimentally studied. Transient values of the mass of adsorbed water vapour, solution concentration and vapour pressure through the bed layers are evaluated from the experimental measurements. The model output, which show the effect of dimensionless relative time (Tr) on the potential ratio (C−C^*)/(C−C^*₀), is compared with the experimental results and good agreement is found.     

Influence of thermodynamic models on the prediction of pool boiling heat transfer coefficient of dilute binary mixtures

In this paper, large number of experiments has been performed on saturated pool boiling heat transfer to three different dilute binary mixtures at various heat fluxes (up to 100 kW/m²) and five different concentrations (1–5 vol.% of heavier component). The test mixtures include water/glycerol, water/mono‐ethylene glycol (MEG), and water/di‐ethylene glycol (DEG). The effects of the main operating parameters such as heat flux, concentration, and bubble dynamics on the pool boiling heat transfer coefficient are also investigated. Furthermore, it is shown that physical properties of the mixtures have a considerable effect on the prediction of pool boiling heat transfer coefficients using the available correlations. In almost all of the existing correlations, some physical properties are strongly involved which can be estimated using different thermodynamic models. These models for the calculation of specific heat, density, heat of vaporization, and surface tension do not give exactly similar results and consequently, the heat transfer coefficient obtained from a specified predictive correlation can be tolerated according to the method used for the calculation of the physical properties. This point is usually ignored by investigators and they compare their experimental data with the correlations without reporting which thermodynamic models have to be used for the calculation of the physical properties. In this study, the prediction of Schlünder correlation has been compared with the present experimental data. Results show that the prediction ability of the Schlünder correlation is strongly dependent on the method used for the estimation of the required physical properties.     

Creep crack growth of seam-welded P22 and P91 pipes with artificial defects. Part II. Data analysis

This paper examines several methods for assessing experimental creep and fatigue crack growth data obtained on P22 (2.25Cr1Mo) and P91 (9Cr1MoVNb) axially notched, seam-welded pipes tested at 565 and 625 °C, respectively [Creep crack growth of seam-welded P22 and P91 pipes with artificial defects—part I: experimental study and post-test metallography. Second International HIDA Conference, Advances in Defects Assessment in High Temperature Plant, MPA, Stuttgart, Germany, 4–6 October, 2000]. The overall objective of this work is to identify the nature of any correlation between component and conventional testpiece creep crack growth rates and thereby provide a supplementary tool for structural integrity analysis. Creep crack growth rate of the notch located in the heat-affect-zone of the weld was assessed in terms of both stress intensity factors, K_I, and the C^*-integral. To estimate the C^*-integral, reference stresses were developed by deriving limit load solutions which reconcile the different collapse loads of the axially notched pipes. Both minimum and average creep rate laws were utilised in the analysis to accommodate the strain rate in the C^* relation. Each test was examined independently, but the general conclusion from each analysis was the same, in that C^*-integral, rather than the stress intensity factor, gave better correlation with respect to conventional data generated using compact tension (CT) specimens. The assessment of creep crack growth demonstrates one particular aspect of interest. In terms of the C^* based correlation of creep crack growth rates, the analysis was found to be relatively independent of the stress state and correlates well with CT specimen data when appropriate reference stresses are used. In addition, cracking in the tested pipes was observed to occur between plane stress and plane strain conditions, inferring that both creep ductility and ligament straining contribute towards the failure mechanism.     

Ex situ and modeling study of two-phase flow in a single channel of polymer electrolyte membrane fuel cells

In this study, we investigate the air-water two-phase flow in a single flow channel of polymer electrolyte membrane (PEM) fuel cells. In the ex situ study, both straight and serpentine channels with various gas diffusion layer (GDL) surfaces are studied. Focus is placed on the two-phase flow patterns, which are optically characterized using a microscope with a high-resolution camera, and the two-phase pressure amplifiers. We find that the GDL surface properties slightly affect the flow pattern and two-phase pressure amplifier in the flow field configuration. Flow pattern transition occurs at the superficial gas velocity of around 1 m s⁻¹, and the pressure amplifier can reach as high as 10. A two-fluid model is also presented together with one dimensional (1-D) analytical solution, and acceptable agreement is achieved between the model prediction and experimental data at high gas flow rates.     

Mathematical modeling of batch bioethanol generation from carob extract in the suspended-cell stirred-tank bioreactor

In this study, various functions were evaluated and utilized to forecast observed values and kinetic parameters of the batch ethanol fabrication from carob extract in the suspended-cell stirred tank reactor (SCSTR). The best model was detected with the model comparison parameters (root-mean-square-error [RMSE], mean-absolute-error [MAE], and R²). The results indicated that the model Stannard (ST) successfully predicted biomass production data (RMSE = 0.26 g L⁻¹, MAE = 0.18 g L⁻¹, and R² = 0.9910), ethanol fabrication data (RMSE = 2.44 g L⁻¹, MAE = 1.88 g L⁻¹, and R² = 0.9809), and sugar depletion data (RMSE = 2.82 g L⁻¹, MAE = 2.17 g L⁻¹ and R² = 0.9938). Nevertheless, the lowest value of the objective function (Φ-factor) was also yielded as 0.041 using the model ST. Additionally, in the estimation of the kinetic data, the model ST also gave well-directed results. Besides, when an independent set of the observed values was utilized to confirm the mathematical functions, the satisfactory consequences were achieved in terms of both the experimental and kinetic values. Consequently, the model ST can work as a universal function in predicting observed values and kinetics of batch ethanol generation from carob extract in an SCSTR.     

Estimation of monthly average daily global solar irradiation using artificial neural networks

This study explores the possibility of developing a prediction model using artificial neural networks (ANN), which could be used to estimate monthly average daily global solar irradiation on a horizontal surface for locations in Uganda based on weather station data: sunshine duration, maximum temperature, cloud cover and location parameters: latitude, longitude, altitude. Results have shown good agreement between the estimated and measured values of global solar irradiation. A correlation coefficient of 0.974 was obtained with mean bias error of 0.059 MJ/m² and root mean square error of 0.385 MJ/m². The comparison between the ANN and empirical method emphasized the superiority of the proposed ANN prediction model.     

Experimental and computational analysis of stall cells on rectangular wings

The present paper is the second part of a combined (experimental and computational) study on stall cells (SCs) on a rectangular wing. In the first part, tuft data were used in order to geometrically characterize a stabilized SC resulting from a localized spanwise disturbance introduced by a zigzag tape. Here, pressure measurements on the model and in the wake and aerodynamic polars at midspan are reported. The wing model had an aspect ratio value of 2, the Reynolds number was 10⁶ and the range of angles of attack (α) was from ?6^° to 16^°. Experimental results confirm previous findings. Furthermore, two‐dimensional and three‐dimensional Reynolds Averaged Navier‐Stokes RANS simulations are used in order to better understand the structure of SCs. 3D simulations reproduce the experimental data with a 3° delay in α and permit a qualitative analysis. It is found that the SC vortices start normal to the wing surface and extend downstream in the wake; the evolution of the SC vortices in the wake is in strong interaction with the separation line vortex and the trailing edge line vortex; as the SC vortex develops downstream in the wake, its centreline is contracted towards the SC centre; the wing wake is pushed upstream at the centre of the SC and downstream at the sides by the SC vortices; spanwise lift and drag distributions always attain their minimum at the SC centre. Copyright © 2013 John Wiley & Sons, Ltd.