Correlating of Thermal Conductivity of monatomic Gases Using Artificial Neural Networks

A set of 119 monatomic Gas was used to train validation and test the performance of the ANN, good correlations were found ($R=0.993$ for NN). The root mean squared errors for the total data set were l%, and mean square errors (MSE) 0.01 for NN. Moreover, it was revealed by the comparison between the forecasted out comes and other models that the neural network models provided greater results.