A Wavelet-Based Stochastic Interpretable Physics-Guided Neural Network to Study the Periodic Heat Transport in a Radiative Convex Fin

The current study investigates the heat transport in the convex profiled fin with radiation and internal heat generation impact. The one-dimensional energy equation of the convex profiled fin is nondimensionalized utilizing appropriate nondimensional variables. The obtained nondimensional energy equation is solved by employing a computational analysis of periodic heat transfer in a radiative convex fin with internal heat generation using a wavelet-inspired physics-based neural network. Further, the influence of several dimensionless constraints on the temperature profile of the convex profiled fin is discussed with the aid of graphs. The convex fins' temperature profile declines as the value of the convection–conduction parameter and the radiation–conduction parameters escalates. The increase in the values of the internal heat generation parameter increases the temperature of the convex profiled fin, and the thermal variation in the convex fin oscillates periodically with varying time. The results of the proposed method are equated with numerical results to validate the efficacy of the approach, signifying accuracy.