Artificial neural networks application on friction factor and heat transfer coefficients prediction in tubes with inner helical-finning

A study presents new model of the artificial neural network to predict friction factor and heat transfer coefficients for the turbulent flow in tubes with inner helical-finning. A fin geometry differs in its form, shape and fabrication method. The generalized equations, correlating thermal performance in such tube, available in earlier works by other authors primarily apply to a single type of inner helical-finning. In present work, we compile experimental results of other authors to an extended database that has been used further for artificial neural network training procedure. The presented model of artificial neural network applies to all types of inner helical tube finning. The mean average percent error values of 11.8% for friction factor and 16.3% for Nusselt number values for the ANN model over the whole database have been achieved. The performance validation of the obtained model was based on a comparison of predicted data with the independent experimental results obtained by authors, yielding excellent accuracy.

Automated summary

A new model of artificial neural network is proposed in this study to predict friction factor and heat transfer coefficients for turbulent flow in tubes with inner helical-finning. The model applies to all types of inner helical tube finning and achieves good accuracy over the entire database.