A Novel Learning Approach for Different Profile Shapes of Convecting-Radiating Fins Based on Shifted Gegenbauer LSSVM

The purpose of this paper is to introduce a novel learning approach to solve the heat transfer problem from convecting-radiating fin model. This model is a nonlinear differential equation in which different boundary conditions cause different profile shapes including rectangular, triangular, trapezoidal and concave parabolic. We consider one-dimensional, steady conduction in the fin and neglect radiative exchange between adjacent fins and between the fin and its primary surface. Our method is based on using the quasilinearization method to linearize the nonlinear models and applying shifted Gegenbauer polynomials as new kernel in least squares support vector machines method. The results of fin efficiency and heat transfer rate of the problems which compared with available previous results indicate better efficiency and accuracy of the proposed approach.