Due to the lack of energy all over the world, the importance of sustainable energy resources and new energy technologies is a consideration for both practitioners and also academicians. Nowadays, the application of thermal engineering is at the top of the sustainable approaches that focus on energy efficiency. Also, as a necessity for improving the efficiency of all of the engineering research fields, the Artificial Neural Network (ANN) model is used to forecast different types of energy efficiency problems in thermodynamic literature. In this paper, the ANN model for thermodynamic evaluation of nitrogen that is obtained by cryogenic liquefaction method in refrigeration systems is applied by two different software programmes. The Engineering Equation Solver (EES) program is used to obtain 441 experimental measurements of the liquefaction. Then, both toolboxes were evaluated with artificial neural networks. This study aims to forecast the exergy efficiency of the cryogenic liquefaction system by applying different neural network architecture types (back propagation, general regression, etc.) in two different developed model. This ANN has been trained using temperature, pressure, enthalpy, and entropy data as inputs for both models. The results obtained from the two models were also compared with the input-output model estimates with reasonable accuracy. This study presents a new procedure for identifying the structure and parameters of hidden layers in ANN models with a higher accuracy ratio and can make predictions in terms of thermodynamic results. Finally, the results of user-friendly and interoperable programs in the package programs based on ANN are compared and meaningful comparative results found.