Prediction of Enthalpy of Solvation for organic solutes and gases Dissolved in Solvent (N,N-dimethylformamide and tert-butanol) With Combining Genetic Algorithm and Artificial neural Network

In This paper we utilized the Genetic Algorithms (GAs) and Artificial Neural Networks (ANNs), for prediction of enthalpy of solvation for organic solutes and gases dissolved in tow solvent. Tow solvent of interest are N,Ndimethylformamide and tert-butanol. This prediction is based on five characteristics of solute and experimentally enthalpy of solvation values for tow solvent of interest. The experimental value for enthalpy of solvation was measured using, direct calorimetric data and gas-liquid chromatography data. The performance of ANN was evaluated by a regression analysis between the predicted and the experimental values. The regression Analysis such as R2 and standard deviation and consequently their error percentage are determined and reported. This method by using the GA can optimize the weights and biases of the ANN, so raise the rate of the prediction and shorten the time of the design. At the same time, this method can simultaneously searched in many directions, thusgreatly increasing the probability of finding a global optimum. Comparisons between Genetic Neural Network (GNN) and famous correlation model like Abraham and Goss model, proofed that GNN is the best model for prediction of enthalpy of solvation and is more accurate.

Keywords:

enthalpy of solvation , artificial neural network , genetic algorithm , N , N dimethylformamide , tert-butanol