Impacts of double stratification on thermally radiative third-grade nanofluid flow on elongating cylinder with homogeneous/ heterogeneous reactions by implementing machine learning approach

This work investigates the influences of double stratification on a thermally radiative third-grade nanofluid flow. The fluid flowed on a stretching cylinder. It is considered that the homogeneous reaction occurs at ambient flow, while the heterogeneous reaction occurs at the wall of the cylinder. The modeled equations were solved using an artificial neural network (ANN) approach. The outcomes of this work revealed that the maximum performance of the modeled problem was obtained at epochs 295, 1,000, 687, and 117 by using an ANN design. The velocity characteristics of the fluid decreased with an increase in the magnetic and curvature factors and increased with an increase in the third-grade dimensionless factor. The temperature distribution diminished with an increase in the curvature factor of the cylinder and thermal stratification factor, and increased with an increase in the radiation factor. The heat transfer field weakened with an increase in the Schmidt number and quotient of the diffusion coefficients and augmented with an increase in the homogeneous/heterogeneous reaction strength factors. The absolute errors are evaluated for all the four scenarios that fall within the range of 10⁻³–10⁻⁸ and are associated with the corresponding ANN configuration that demonstrates a fine degree of accuracy.