The heat transfer characteristics of the refrigerant HFC-134a are investigated, such as convective heat transfer coefficient and pressure drop during evaporation inside a vertical smooth and five pieces of corrugated tube, using experimental data with the aim of numerically determining the best artificial intelligence method. The double tube test sections are 0.5 m long with refrigerant flowing in the inner tube and heating water flowing in the annulus. Input of the ANNs are the 14 numbers of dimensional and dimensionless values of test section, such as mass flux, heat flux, temperature difference between the tube wall and saturation temperature, average vapour quality, evaporating temperature, two-phase friction factor, two-phase multiplier, liquid and vapour Reynolds numbers, Bond number, Froude number, Weber number, depth of corrugation and helix angle for the tested corrugated tubes, whereas the outputs of the ANNs are the experimental condensation heat transfer coefficient and measured pressure drop from the analysis. The evaporation heat transfer characteristics of R134a are modelled to decide the best approach, using several ANN methods such as multi layer perceptron (MLP) and radial basis networks (RBFN). The performance of the method of MLP with 10-5-1 architecture and RBFNs with the spread coefficient of 100,000 and a hidden layer neuron number of 200 were found to be in good agreement, predicting the evaporation heat transfer coefficient and pressure drop. Dependency of outputs of the ANNs from input values is investigated and new ANN-based heat transfer coefficient correlations are developed as a result of the analyses.