This paper presents a hybrid approach for investigation of heat transfer enhancement performance using computational fluid dynamics and artificial neural network. More than 5,000 CFD simulations are carried out for turbulent flow in pipes provided with artificial roughness of transverse rectangular ribs to analyze heat transfer, pressure drop, and thermal hydraulic performance. The rib height and pitch are widely varied along with the flow Reynolds number, working fluid, and material of roughness elements. To accurately predict major parameters (Nusselt number, friction factor, and thermal hydraulic performance) a deep neural network is developed, trained, and tested by current CFD data. The ANN allowed finding optimal rib roughness parameters for the current problem and opened perspectives of industrial application due to low computational cost and prediction error of less than 1.5%.
This paper presents a comparison of three different approaches for modeling enhanced heat transfer characteristics of turbulent airflow in a circular tube with artificial roughness of transverse ribs. A number of CFD simulations are carried out forming the first dataset as well as the second dataset extracted from a number of classical works. A deep feed-forward neural network is developed to predict Nusselt number and friction factor for a variety of rib roughness and flow parameters. The ANN is trained by the first dataset (the CFD and ANN approach) and the second dataset (the experiment and ANN approach) independently and by a combination of datasets (the hybrid approach) showing good quality predictions in all the cases. All results are compared with experimental data and CFD modelled values showing the best results of the experiment and ANN approach.
This paper presents a numerical analysis of convective heat transfer enhancement of transverse ribs in circular tubes. Several CFD simulations are carried out for turbulent airflow to analyze heat transfer and pressure drop provided with transverse rectangular ribs. The rib height and pitch are widely varied along with the flow Reynolds number. The effect of each parameter is examined and discussed. To accurately predict major parameters (Nusselt number, friction factor, and thermal hydraulic performance parameter) a deep neural network is developed, trained, and tested by current CFD data. The result demonstrates that artificial neural network shows better performance compared to other methods of prediction (e.g. power-law approximation) and can offer an economical and powerful approach for modeling optimal heat enhancement parameters.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.