In this study, an Adaptive Neuro-Fuzzy Inference System (ANFIS) is used for modeling the effect of geometrical parameters on heat transfer in helically coiled tubes. An experimental study was carried out to investigate the heat transfer rate in nine helically coiled tubes with different geometrical parameters including: curvature ratio and coil pitch. In the experiments, hot water was passed in the coiled tubes, which were placed in a cold bath. The output data of the ANFIS is Nusselt number and input data are Reynolds number, Prandtl number, coil pitch, and curvature ratio. The validity of the method was evaluated through a test data set, which were not employed in the training stage of the ANFIS. The predicted results of the developed ANFIS were compared with the experimental data for both training and testing data sets. The results show that the adaptive neuro-fuzzy model is a capable method in order to predict performance of thermal systems in engineering applications.
