Heat Transfer Prediction of Supercritical R134a in Horizontal Pipe by Artificial Neural Network

The present study proposes a method to predict the heat transfer of supercritical R134a in both smooth and internally ribbed tubes using an artificial neural network (ANN). A detailed analysis of the relevant factors affecting the prediction accuracy was conducted, and the ANN model was optimized accordingly. Four correlations were selected for comparison with the ANN model. The results demonstrate that the prediction accuracy of the ANN is higher than that of the traditional correlation models. Furthermore, the ANN model can simultaneously predict the heat exchange in different types of tubes (smooth pipe and internally ribbed pipe), indicating its superior generalization capabilities. This suggests that the ANN model can address the limited application range of traditional correlation methods.