Search

In this communication, modeling of carbon dioxide absorption by various amino acid solutions is presented as a function of operational parameters using the Least‐Squares Support Vector Machine (LSSVM) algorithm integrated with three different evolutionary algorithms, namely Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Hybrid GA and PSO (HGAPSO). A databank containing 255 data of carbon dioxide absorption by amino acids of potassium taurate, potassium glycinate, potassium prolinate, and potassium lysinate at different temperatures, partial pressures, and concentrations was prepared from different sources to train and test the proposed algorithms. The models were applied to estimate carbon dioxide absorption by amino acid solutions in terms of temperature, molar concentration, equilibrium partial pressure, molecular weight, number of hydrogen bond donor, number of hydrogen bond acceptor, and number of rotatable bond. The R2 values of LSSVM optimized by HGAPSO, PSO, and GA are 0.9944, 0.9915, and 0.9891, respectively, and the various errors were determined close to zero. On the other hand, the visual comparison of models outputs and actual carbon dioxide adsorption data was used to clarify performances of the models. By comparison analysis, it was found that the LSSVM‐HGAPSO is the most accurate model for estimation of carbon dioxide loading. Also, comparison of our proposed models results with previously reported artificial neural network results indicates the impressive estimation capability of LSSVM algorithm. According to sensitivity analysis, it becomes obvious that pressure is the most effective parameter on carbon dioxide absorption. [ABSTRACT FROM AUTHOR]

In the present work, a novel machine learning computational investigation is carried out to accurately predict the solubility of different acids in supercritical carbon dioxide. Four different machine learning algorithms of radial basis function, multi-layer perceptron (MLP), artificial neural networks (ANN), least squares support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS) are used to model the solubility of different acids in carbon dioxide based on the temperature, pressure, hydrogen number, carbon number, molecular weight, and the dissociation constant of acid. To evaluate the proposed models, different graphical and statistical analyses, along with novel sensitivity analysis, are carried out. The present study proposes an efficient tool for acid solubility estimation in supercritical carbon dioxide, which can be highly beneficial for engineers and chemists to predict operational conditions in industries. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

In the recent years due to increasing demand for energy and declination of reservoir production, an impressive notice on enhancement of oil recovery has been found. The gas injection especially carbon dioxide injection due to low cost and friendly environmentally of this approach the special attention to CO2injection increased. The miscibility is known as key factor which effects on enhancement of recovery. The miscibility is controlled by interfacial tension of hydrocarbons and carbon dioxide so the importance of investigation of the interfacial tension becomes highlighted.in this investigation by using radial basis function (RBF) artificial neural network (ANN) as a novel approach the interfacial tension of hydrocarbons and carbon dioxide in terms of pressure, temperature, liquid and gas densities and molecular weight of alkane. The graphical and statistical results illustrated the fact that RBF-ANN algorithm is applicable for estimation of interfacial tension between hydrocarbons and carbon dioxide with great accuracy. [ABSTRACT FROM AUTHOR]

One of problematic topics in petroleum engineering is Asphaltene precipitation issue which causes problems such as tubing plugging and formation damage due to temperature, pressure and composition changes so the notability of this issue increases. In the present investigation a novel Fuzzy c-means (FCM) algorithm was developed to predict precipitated asphaltene as function of dilution ratio, carbon number of precipitants and temperature for solving the problem. The results showed that this novel approach has great ability to predict precipitated asphaltene in terms of aforementioned parameters. The coefficients of determination (R2) for training and testing steps are calculated as 0.9828 and 0.9387 respectively. This great degree of accuracy expresses that the predicting algorithm has potential to be utilized as software for prediction of asphaltene behavior. [ABSTRACT FROM AUTHOR]

Asphaltene precipitation is known as one of the challenging problems in petroleum industries which have significant effects on production such as formation damage and wellbore plugging. To solve this problem, calculation of precipitated asphaltene becomes highlighted so in the present study a novel approach is proposed based on ANFIS algorithm to estimate precipitated asphaltene in terms of dilution ration, carbon number of precipitants and temperature. The particle swarm optimization (PSO) method is applied to optimize ANFIS algorithm parameters. The proposed model was evaluated based on statistical parameters and the calculated R2, AARD and RMSE for the total data are 0.90309, 9.4908 and 7.9468. They showed the predicting algorithm performed in acceptable manner so a high accurate and simple tool was proposed to predict the precipitated asphaltene as function of Dilution ration, temperature and carbon number of precipitants. [ABSTRACT FROM PUBLISHER]