HEAT TRANSFER COEFFICIENT AND FRICTION FACTOR PREDICTION OF CORRUGATED TUBES COMBINED WITH TWISTED TAPE INSERTS USING ARTIFICIAL NEURAL NETWORK

In the research described here, artificial neural network ( ANN) approach has been utilized to characterize the thermohydraulic behavior of corrugated tubes combined with twisted tape inserts in a turbulent flow regime. The experimental data sets were extracted from 57 tubes, 9 and 3 spirally corrugated tubes with varying geometries combined with 5 and 4 twisted tapes with different pitches. The tests were carried out with Reynolds numbers ranging from 3000 to 60,000. The experimental data sets have been utilized in training and validation of the ANN in order to predict the heat transfer coefficients and friction factors inside the corrugated tubes combined with twisted tape inserts, and the results were compared to the experimental data. The mean relative errors between the predicted results and experimental data were less than 2.9% for the heat transfer coefficients and less than 0.36% for the friction factor. The performance of the neural networks was found to be superior in comparison with the models correlated in the form of mathematical functions with their own assumptions. The results of this study suggested that ANN can be considered as a powerful tool and can be easily utilized to predict the performance of thermal systems in engineering applications.