Iranian Journal of Chemical Engineering (IJChE)

Abstract Time-consuming and costly experiments to measure the cetane number (CN) of biodiesel make computations even more valuable. In the current study, two artificial intelligence (AI) models have been used to predict the biodiesel CN by using comprehensive datasets (440 datasets). They were the gradient-based artificial neural network (GB-ANN) and the multi-layer-perceptron ANN optimized by the genetic algorithm (GA-ANN) for the first time. The three input variablesof the model for predicting the target variable of the biodiesel CN are the average number of carbon atoms, average number of double bonds, and average molecular weight of the fatty acid methyl esters. The learning function, transfer function, number of hidden layers, and number of neurons in the hidden layers are some of the optimized parameters in the current AI-models. The developed models were compared using statistical criteria such as the coefficient of determination (R²), mean square error (MSE), average absolute relative deviation (AARD), standard deviation (STD) and mean absolute percentage error (MAPE). The resulting outcomes revealed that the highest R² and the lowest MSE were related to the GB-ANN model with two hidden layers, trainbfg learning method and logsig-tansig-purelin transfer function. The R² and MSE for the optimized model are equal to 0.9296 and 0.0005 respectively. Although the GA-ANN achieved acceptable outcomes, its statistical analyses produced weaker outcomes than the AI-model based on GB-ANN.

Abstract The artificial neural network (ANN) approach was applied to develop simple correlations for predicting the thermal conductivity of nitrogen-methane and carbon dioxide-methane mixtures. The genetic algorithm method was used to obtain global optimum parameters (weights and biases) of the ANNs. The methane mole fraction, temperature, pressure, and density as effective parameters on thermal conductivity were network input variables. 171 and 180 data points related to the nitrogen-methane and carbon dioxide-methane gas mixtures, respectively, divided to test and train datasets. Simple correlations were obtained due to the small number of optimal neurons in the ANN structures. The mean relative errors of 0.206% and 0.199% for the testing dataset indicate the high accuracy and validation of the correlations. The work indicates that artificial intelligence approaches are very useful for thermal conductivity modeling in natural gases. A sensitivity analysis was performed on all input variables that indicates that the gas mixture density has the greatest impact on the thermal conductivity.

Abstract In the study, the thermal-hydraulic performance of the zigzag channels with circular cross-section was analyzed by Computational Fluid Dynamics (CFD). The standard K-Ꜫ turbulent scalable wall functions were used for modeling. The wall temperature was assumed constant 353 K and water was used as the working fluid. The zigzag serpentine channels with bend angles of 5 - 45° were studied for turbulent flow from 4000 to 40,000 Reynolds number (Re). The thermal performance of the zigzag 45° channel was better than the other channels and also it had the highest friction factor (f). The bends caused secondary flow, and as the bend angle increased, the secondary flow increased. This Phenomenon had a positive effect on thermal performance and a negative effect on hydraulic performance by increasing the friction factor. The obtained CFD data used to develop correlations for predicting the Nu and f as the functions of Re and bend angles. The correlation constants were optimized by the genetic algorithm method which leads to the mean relative errors of 3.32% and 6.94% for Nu and f estimation, respectively.

Abstract Utilization of membrane humidifiers is one of the methods commonly used to humidify reactant gases in polymer electrolyte membrane fuel cells (PEMFC). In this study, polymeric porous membranes with different compositions were prepared to be used in a membrane humidifier module and were employed in a humidification test. Three different neural network models were developed to investigate several parameters, such as casting solution composition, membrane thickness, operating pressure, and flow rate of input dry air which have an impact on relative humidity of the exhausted air after humidification process. The three mentioned models included Feed- Forward Back- Propagation (FBP), Radial Basis Function (RBF), and Feed- Forward Genetic Algorithm (FFGA). The developed models were verified by experimental data. The results showed that the feed- forward neural network models, especially FFGA, were suitable for prediction of the effect of membrane composition and operating conditions on the performance of this type of membrane humidifiers

Abstract The present work is concerned with the development of a new algorithm for determination of the equilibrium composition of gaseous detonations. The elements balance equations, and the second law of thermodynamics (i.e., the minimization of the Gibbs free energy of products), are used to determine the equilibrium composition of the detonation products. To minimize the Gibbs free energy with traditional deterministic methods one needs to solve a set of highly nonlinear equations. The numerical methods in the existing equilibrium codes suffer from several drawbacks such as the divergence possibility in some equivalent ratios, and the possibility of converging to a local relative minimum in the minimization process. To overcome these drawbacks, a genetic algorithm is presented in the present study. Converging to the global minimum of Gibbs function in all equivalent ratios, and having a reasonable CPU time are the notable aspects of the proposed algorithm.
 
 

Abstract An industrial ethane thermal cracking reactor was modeled assuming a molecular mechanism for the reaction kinetics coupled with material, energy, and momentum balances of the reactant-product flow along the reactor. To carry out the multi-objective optimization for two objectives such as conversion and ethylene selectivity, the elitist non-dominated sorting genetic algorithm was used. The Pareto optimum set was obtained successfully and finally the effect of the decision variable was discussed.