A feedforward three-layer neural network is proposed to predict conductivity (k) of gas mixtures at a wide range of temperatures and mole fraction based on the pure component conductivities and mole fraction of first gas. The accuracy of the method is evaluated and tested by its application to experimental conductivities of various mixed gases which some of them are not used in the network training. Through the comparison of the calculated results between this method and one of the theoretical analysis in the literature, the validity of this method based on the ANN is confirmed. Conventional conductivity correlations are usually used for a limited range of temperature and mole fraction of components while the network method can cover a wide range of temperatures and substances.
