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. |
