One of the equipment that plays a significant role in reducing energy consumption is heat exchangers. Therefore, their optimal design is very important. One type of heat exchanger is the spiral heat exchanger, which is widely used today because of its advantages over other types. In this research, spiral tubes with a square cross-section and conical angles of 10, 30, and 50 degrees and step lengths of 15, 30, and 45 mm were modeled by computational fluid dynamics (CFD) to investigate the thermal performance of the computational fluid. The CFD data was validated and used to develop an artificial neural network (ANN). The input variables are Reynoldth number (Re), step length (b), and cone angle (Ө). 70% of the data were used for training the network and 30% were used to check the validity of the model. The results show high accuracy of the artificial neural network in predicting Nusselt number (Nu).
