Implication of radiation on the thermal behavior of a partially wetted dovetail fin using an artificial neural network

The simultaneous convection-radiation heat transfer of a partially wetted dovetail extended surface is investigated in this study. Also, the temperature variance behavior of the dovetail extended surface (DES) is estimated through thermal models for partially wet and dry conditions using the neural network with the Levenberg-Marquardt scheme (NNLMS). The corresponding governing energy equations of a dovetail fin are presented as a set of ordinary differential equations (ODE), which are reduced to a non-dimensional form using dimensionless terms. Further, the resulting coupled conductive, convective, and radiative dimensionless ODEs are numerically solved utilizing the Runge-Kutta-Fehlberg fourth-fifth order (RKF-45) scheme. Using graphical illustrations, the resultant solutions are physically determined by considering the effects of various nondimensional variables on thermal behavior. From the outcomes, it is established that the thermal conductivity parameter enhances the thermal distribution in a partially wetted dovetail fin, and an upsurge in convection-conduction variable, temperature ratio parameter, radiation-conduction, and wet parameter diminishes the temperature profile of the considered extended surface. The modelled problem's NNLMS efficacy is demonstrated by achieving the best convergence and unique numerically assessed quantified results. The outcomes indicate that the strategy successfully resolves the partially wetted fin problem.

Automated summary

This study investigates the simultaneous convection-radiation heat transfer of a partially wetted dovetail extended surface and estimates the temperature variance behavior using a neural network model. The resulting numerical solutions provide insights into the thermal behavior of the partially wetted fin.