CO2 capture by benzene-based hypercrosslinked polymer adsorbent: Artificial neural network and response surface methodology

In this research, porous benzene-based hypercrosslinked polymeric adsorbents with different morphological properties were synthesized through Friedel–Crafts alkylation reaction. The resulting samples were applied for CO₂ capture at different operational conditions. Two modelling approaches, including artificial neural network (radial basis function [RBF] and multi layer perceptron [MLP]) and response surface methodology (RSM), were employed to investigate the effect of independent parameters on adsorption capacity. A semi-empirical quadratic model for adsorption capacity was presented based on RSM-central composite design technique. Additionally, the optimal structure of RBF was determined with 200 neurons, and the optimal structure of MLP was determined with three hidden layers and 10, 8, and 7 neurons. The modelling results demonstrate the better prediction of MLP and RBF approaches than the RSM method with correlation coefficient values of 0.999, 0.989, and 0.931, respectively. Finally, process optimization was carried out using RSM optimization module and the optimized values of synthesis time, crosslinker ratio (formaldehyde dimethyl acetal [FDA]/benzene), adsorption time, pressure, and temperature were obtained at 10.11 h, 1, 220 s, 9 bar, and 55°C, respectively. The optimum value of CO₂ uptake capacity was obtained around 167 (mg/g).     

Prediction of the gas solubility in polymers by a radial basis function neural network based on chaotic self‐adaptive particle swarm optimization and a clustering method

A novel model based on a radial basis function neural network (RBF NN), chaos theory, self‐adaptive particle swarm optimization (PSO), and a clustering method is proposed to predict the gas solubility in polymers; this model is hereafter called CSPSO‐C RBF NN. To develop the CSPSO‐C RBF NN, the conventional PSO was modified with chaos theory and a self‐adaptive inertia weight factor to overcome its premature convergence problem. The classical k‐means clustering method was used to tune the hidden centers and radial basis function spreads, and the modified PSO algorithm was used to optimize the RBF NN connection weights. Then, the CSPSO‐C RBF NN was used to investigate the solubility of N₂ in polystyrene (PS) and CO₂ in PS, polypropylene, poly(butylene succinate), and poly(butylene succinate‐co‐adipate). The results obtained in this study indicate that the CSPSO‐C RBF NN was an effective method for predicting the gas solubility in polymers. In addition, compared with conventional RBF NN and PSO neural network, the CSPSO‐C RBF NN showed better performance. The values of the average relative deviation, squared correlation coefficient, and standard deviation were 0.1282, 0.9970, and 0.0115, respectively. The statistical data demonstrated that the CSPSO‐C RBF NN had excellent prediction capabilities with a high accuracy and a good correlation between the predicted values and the experimental data. © 2013 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 130: 3825–3832, 2013     

Artificial neural network modelling of supercritical fluid CO2 extraction of polyunsaturated fatty acids from common carp (Cyprinus carpio L.) viscera

Common carp viscera, obtained from Tikveš Lake in Macedonia, was investigated as a possible source of polyunsaturated (PUFA) fatty acids. Supercritical fluid CO₂ extraction (SFE-CO₂) was employed for extraction of investigated bioactive components. The GC-FID analysis on the total extract obtained by supercritical fluid CO₂ extraction confirmed the assumption of presence of these bioactive components. A three layer artificial neural network was created for prediction and modelling of the extraction yield of polyunsaturated fatty acids from lyophilized viscera matrixes. Operating values of pressure, temperature, mass flow of CO₂ and extraction time were defined as input vectors to the ANN where PUFA extraction yield was considered as an output vector. Created ANN model provided adequate fitting of experimental data, with a correlation coefficient of 0.9968 for the entire data set. RSM-3D method was employed for mathematical modelling of the ANN output values as a function of operating variables and their interactions.     

基于模糊RBF神经网络的乙烯装置生产能力预测

针对传统的径向基函数(RBF)神经网络隐藏层节点的不确定和初始中心敏感性、收敛速度过慢等问题,提出一种基于模糊C均值的RBF神经网络(FCM-RBF)模型,通过模糊C均值聚类(FCM)得到各聚类中心,基于误差反传的梯度下降法训练隐藏层到输出层之间的权值,克服传统RBF模型对数据中心的敏感性,优化确定RBF神经网络隐藏层的节点数,提高网络训练速度和精度。最后将其用于乙烯装置生产能力预测中,分析预测不同技术、不同规模乙烯装置生产情况,指导乙烯生产,提高生产效率,结果验证了所提出算法的有效性和实用性。     

SOFT COMPUTING METHOD FOR PREDICTION OF CO2 CORROSION IN FLOW LINES BASED ON NEURAL NETWORK APPROACH

An important aspect of corrosion prediction for oil/gas wells and pipelines is to obtain a realistic estimate of the corrosion rate. Corrosion rate prediction involves developing a predictive model that utilizes commonly available operational parameters, existing lab/field data, and theoretical models to obtain realistic assessments of corrosion rates. This study presents a new model to predict corrosion rates by using artificial neural network (ANN) systems. The values of pH, velocity, temperature, and partial pressure of the CO₂ are input variables of the network and the rate of corrosion has been set as the network output. Among the 718 data sets, 503 of the data were implemented to find the best ANN structure, and 108 of the data that were not used in the development of the model were used to examine the reliability of this method. Statistical error analysis was used to evaluate the performance and the accuracy of the ANN system for predicting the rate of corrosion. It is shown that the predictions of this method are in acceptable agreement with experimental data, indicating the capability of the ANN for prediction of CO₂ corrosion rate in production flow lines.     

Development of artificial neural network models for supercritical fluid solvency in presence of co-solvents

This paper presents the application of artificial neural networks (ANN) to develop new models of liquid solvent dissolution of supercritical fluids with solutes in the presence of cosolvents. The neural network model of the liquid solvent dissolution of CO₂ was built as a function of pressure, temperature, and concentrations of the solutes and cosolvents. Different experimental measurements of liquid solvent dissolution of supercritical fluids (CO₂) with solutes in the presence of cosolvents were collected. The collected data are divided into two parts. The first part was used in building the models, and the second part was used to test and validate the developed models against the Peng-Robinson equation of state. The developed ANN models showed high accuracy, within the studied variables range, in predicting the solubility of the 2-naphthol, anthracene, and aspirin in the supercritical fluid in the presence and absence of co-solvents compared to (EoS). Therefore, the developed ANN models could be considered as a good tool in predicting the solubility of tested solutes in supercritical fluid.     

Thermal performance prediction of wickless heat pipe with Al2O3/water nanofluid using artificial neural network

A wickless heat pipe (WHP) comprises of an evacuated-close tube filled with an appropriate amount of working fluid. In this study, the effect of Al₂O₃/water nanofluid as the working media on thermal performance of WHP investigated and compared with pure water by designing an optimized Artificial Neural Network (ANN). ANN trained with the collected test data obtained from experimental setup and validated. Multilayer Perceptron configuration (MLP) adopted for the ANN. The MLP architecture consists of four input nodes representing the parameters; input power, volume concentration of nanofluid, filling ratio and mass rate in condenser section, and a single output node representing the thermal efficiency of WHP. According to sensitivity analysis results, volume concentration is the most significant parameter which affects the WHP performance. Also, since the ANN test output data are sufficiently close to experimental one, it can be inferred that the ANN model can be applied to accurately model WHP thermal performance.     

Predicting the vapor-liquid equilibrium of carbon dioxide+alkanol systems by using an artificial neural network

A multi-layer feed-forward artificial neural network has been presented for accurate prediction of the vapor liquid equilibrium (VLE) of CO₂+alkanol mixtures. Different types of alkanols namely, 1-propaol, 2-propanol, 1-butanol, 1-pentanol, 2-pentanol, 1-hexanol and 1-heptanol, are used in this study. The proposed network is trained using the Levenberg-Marquardt back propagation algorithm, and the tan-sigmoid activation function is applied to calculate the output values of the neurons of the hidden layers. According to the network’s training, validation and testing results, a six layer neural network is selected as the best architecture. The presented model is very accurate over wide ranges of experimental pressure and temperatures. Comparison of the suggested neural network model with the most important thermodynamic correlations shows that the proposed neuromorphic model outperforms the other available alternatives. The predicted equilibrium pressure and vapor phase CO₂ mole fraction are in good agreement with experimental data suggesting the accuracy of the proposed neural network model for process design.     

Experimental and modeling of CO2 capture by dry sodium hydroxide carbonation

In this work, CO₂ capture from the air using dry NaOH sorbents has been studied. The influences of the main operating parameters such as temperature, air humidity, and NaOH loading on the CO₂ removal rate have been experimentally investigated using Taguchi method. The results revealed that the appropriate value of the temperature to maximize the rate was in the range of 35–45 °C. A multilayer artificial neural network (ANN) was also used to model the process in order to find the optimal conditions. A procedure reported in the literature was modified and applied to design the ANN model. The model predictions were validated by conducting some more experiments. The experimental results proved the accuracy of the model to predict the optimal conditions. The effects of NaOH particle size and multiple carbonation cycles have also been investigated.     

Hybrid neural network for prediction of CO<Subscript>2</Subscript> solubility in monoethanolamine and diethanolamine solutions

The solubility of CO₂ in single monoethanolamine (MEA) and diethanolamine (DEA) solutions was predicted by a model developed based on the Kent-Eisenberg model in combination with a neural network. The combination forms a hybrid neural network (HNN) model. Activation functions used in this work were purelin, logsig and tansig. After training, testing and validation utilizing different numbers of hidden nodes, it was found that a neural network with a 3-15-1 configuration provided the best model to predict the deviation value of the loading input. The accuracy of data predicted by the HNN model was determined over a wide range of temperatures (0 to 120 °C), equilibrium CO₂ partial pressures (0.01 to 6,895 kPa) and solution concentrations (0.5 to 5.0M). The HNN model could be used to accurately predict CO₂ solubility in alkanolamine solutions since the predicted CO₂ loading values from the model were in good agreement with experimental data.     

Comparison between the artificial neural network,SAFT and PRSV approach in obtaining the solubility of solid aromatic compounds in supercritical carbon dioxide

The aim of this study is to model the solubilities of solid aromatic compounds in supercritical carbon dioxide (SCCO₂) using feed-forward artificial neural network (ANN). Temperature, pressure, critical properties and acentric factor of each solute have been used as independent variables of ANN model. The parameters of multi-layer perceptron (MLP) network have been adjusted by back propagation learning algorithm using experimental data which have been collected from various literatures. In order to find the optimal topology of the MLP, different networks were trained and examined and the network with minimum absolute average relative deviation percent (AARD%), mean square error (MSE) and suitable regression coefficient (R²) has been selected as an optimal configuration. By this procedure a single hidden layer network composed of nineteen hidden neurons has been found as an optimal topology. Sensitivity error analyses confirmed that the optimal ANN can predict experimental data with an excellent agreement (AARD% = 4.99, MSE = 7.08 × 10⁻⁷ and R² = 0.99699). Capability of the proposed ANN model has compared with those published results which have obtained by SAFT combined with eight different mixing rules (one, two and three parameters mixing rules) and PRSV equation of state (EOS). The best presented overall AARD% for SAFT approach with one, two and three parameters mixing rules are 16.15, 12.32% and 7.65%, respectively while PRSV EOS showed AARD% of 21.10%. The results emphasize that the proposed ANN model can predict the solubilities of solid aromatic compounds in SCCO₂ more accurate than SAFT and PRSV EOS.     

Artificial neural network for the second virial coefficient of organic and inorganic compounds: An ANN for B of organic and inorganic compounds

An artificial neural network (ANN) to estimate the second virial coefficient, valid for organic and inorganic compounds, is presented. First, we statistically analyzed 6,531 experimental points, belonging to 234 fluids, collected from literature. The data were investigated with a factor analysis approach to identify the most significant parameters that influence the second virial coefficient. The factor analysis, combined with physical considerations, allowed to find four (T_r, T_c, P_c, ω) or five (μ_r) parameters as input variables for the ANN, according to the specific chemical family. The architecture of the proposed multi-layers perceptron (MLP) neural network consists of one input layer with five input variables (T_r, T_c, P_c, ω, μ_r), one output layer with one neuron (B) and two-hidden-layers with 19 neurons each. We trained, validated and tested several configurations of the neural network to obtain this network topology that minimizes the deviations between experimental and calculated points. Results show that the ANN is able to calculate the second virial coefficient with greater accuracy (RMSE?=?29.38?cm³/mol) than that of correlations available in literature. To identify the outliers and applicability domain of the proposed MLP neural network, an outlier diagnosis based on the Leverage approach was performed. This analysis shows that the model is statistically valid.     

Hybrid modeling of ethylene to ethylene oxide heterogeneous reactor

In this research a dynamic grey box model (GBM) of ethylene oxide (EO) fixed bed reactor has been presented. In the first step of the study, kinetic model of the existing reactions was obtained using artificial neural network (ANN) approach. In order to build the ANN model industrial data of a typical EO reactor were employed. Time, C₂H₄, C₂H₄O, CO₂, H₂O and O₂ mole fractions were network inputs and the multiplication of reaction rate and catalyst deactivation (r * a)was ANN output. From 164 data, 109 data were employed to train ANN. After employing different training algorithms, it was found that, the radial basis function network (RBFN) training algorithm provides the best estimations of the data. This best obtained network was tested against fifty five unseen data. The network estimations were close to unseen data which confirmed generalization capability of the obtained network.In the next step of study, (r * a) was estimated with ANN and then the hybrid model of the reactor was solved. Simulation results were compared with EO mechanistic model and also with plant industrial data. It was found that GBM is 8.437 times more accurate than the mechanistic model.     

Design of CO2 hydrogenation catalyst by an artificial neural network

An artificial neural network (ANN) is firstly applied to CO₂ hydrogenation catalyst design. CO₂ catalytic hydrogenation activity is represented as a function of the catalyst composition and reaction condition. After training, the network can predict CO₂ catalytic hydrogenation activity and design CO₂ hydrogenation catalyst successfully. The estimated results are in good agreement with the experimentally observed ones.     

NEURAL NETWORK MODEL FOR FLUIDISED BED DRYERS

The application of an artificial neural network (ANN) to model a continuous fluidised bed dryer is explored. The ANN predicts the moisture and temperature of the output solid. A three-layer network with sigmoid transfer function is used. The ANN learning is made by using a set of data that were obtained by simulating the operation by a classical model of dryer. The number of hidden nodes, learning coefficient, size of learning data set and number of iterations in the learning of the ANN were optimised. The optimal ANN has five input nodes and six hidden nodes. It is able to predict, with an error less than 10%, the moisture and temperature of the output dried solid in a small pilot plant that can treat up to 5 kg/h of wet alpeorujo. This is a wet solid waste that is generated in the two-phase decanters used to obtain olive oil.     

Measurement and correlation of supercritical CO2 and ionic liquid systems for design of advanced unit operations

Ionic liquids combined with supercritical fluid technology hold great promise as working solvents for developing compact processes. Ionic liquids, which are organic molten salts, typically have extremely low volatility and high functionality, but possess high viscosities, surface tensions and low diffusion coefficients, which can limit their applicability. CO₂, on the other hand, especially in its supercritical state, is a green solvent that can be used advantageously when combined with the ionic liquid to provide viscosity and surface tension reduction and to promote mass transfer. The solubility of CO₂ in the ionic liquid is key to estimating the important physical properties that include partition coefficients, viscosities, densities, interfacial tensions, thermal conductivities and heat capacities needed in contactor design. In this work, we examine a subset of available high pressure pure component ionic liquid PVT data and high pressure CO₂-ionic liquid solubility data and report new correlations for CO₂-ionic liquid systems with equations of state that have some industrial applications including: (1) general, (2) fuel desulfurization, (3) CO₂ capture, and (4) chiral separation. New measurements of solubility data for the CO₂ and 1-butyl-3-methylimidazolium octyl sulfate, [bmim][OcSO₄] system are reported and correlated. In the correlation of the CO₂ ionic liquid phase behavior, the Peng-Robinson and the Sanchez-Lacombe equations of state were considered and are compared. It is shown that excellent correlation of CO₂ solubility can be obtained with either equation and they share some common characteristics regarding interaction parameters. In the Sanchez-Lacombe equation, parameters that are derived from the supercritical region were found to be important for obtaining good correlation of the CO₂-ionic liquid solubility data.     

The use of artificial neural network (ANN) in dry flue gas desulphurization modelling: Levenberg–Marquardt (LM) and Bayesian regularization (BR) algorithm comparison

This research project aims to investigate the efficacy of artificial neural networks (ANN) in mapping dry flue gas desulphurization (DFGD). Bayesian regularization (BR) and Levenberg–Marquardt (LM) training algorithms were used for DFGD modelling. The input layer feed data contained diatomite to Ca(OH)₂ ratio, hydration time, hydration temperature, sulphation temperature, and inlet gas concentration, while the output layer metadata were sorbent conversion and sulphation responses. The hyperbolic tangent (tansig), sigmoid (logsig), and linear (purelin) activation functions were compared to ascertain the best network learning model. The number of hidden layer cells also varied between 7 and 10, given the existence of multiple output feed data. BR and LM performance evaluation was based on coefficient of determination (R²), root mean square error (RMSE), and mean square error (MSE) mathematical analysis. BR was a superlative training tool compared to LM, with lower RMSE and MSE values. The goodness of fit data for both techniques was close to unity, clarifying that ANN using BR and LM tools can be used to predict DGFD outcome. The shrinking core model was used to analyze the desulphurization reaction and concluding the chemical reaction was the reaction controlling step.     

Modeling of CO2 Solubility in Selected Imidazolium-Based Ionic Liquids

This study explores the use of COSMO-RS model and Peng-Robinson (PR) equation of state (EoS) to predict the solubility of carbon dioxide (CO₂) in specific ionic liquids (ILs). COSMO-RS was employed to estimate of CO₂ solubility at atmospheric pressure in eight imidazolium-based ILs resulting from the combination of ethyl, butyl, hexyl, and octyl-imidazolium cations with two anions: bis(trifluoromethylsulfonyl)imide ([Tf₂N]) and Trifluoromethanesulfonate ([TFO]). The results indicated relatively acceptable qualitative consistency between the experimental and predicted values. The PR EoS was employed at high pressure by tuning the interaction parameters to fit the experimental data reported in the literature. The model demonstrated excellent accuracy in predicting the solubility of CO₂ at pressure values less than the critical pressure of CO₂; however, at higher pressures, the calculated solubility diverged from the experimental values. Furthermore, the type of anion and cation used in the IL affected the performance of the PR EoS.