An evolutionary-based neural network approach to investigate heat and mass transportation by using non-Fourier double-diffusion theories for Prandtl nanofluid under Hall and ion slip effects

Achieving precise predictions and classifications with artificial neural networks (ANNs) while minimizing the consumption of computational resources and time continues to be a substantial objective in the realm of scientific inquiry. The primary objective of this research is to create ANNs that achieve a harmonious equilibrium between precision and computational speed, thereby enabling their implementation in a wide array of contexts. This study aims to examine the characteristics of Prandtl nanofluids using ANN-HCS-PNF, an innovative computational framework that integrates artificial neural networks and a hybridized cuckoo search method. The aforementioned framework is utilized in conjunction with the Cattaneo-Christov double heat flow model, with Hall and ion slip effects also taken into account. The central argument of this manuscript pertains to the examination of concentration, velocity, and temperature profiles in the flow of Prandtl nanofluid using the Cattaneo-Christov double heat flux model. This investigation focuses specifically on the consequences of Hall and ion slip effects. In order to explore the complexities of flow dynamics, scientists utilize an unsupervised ANN approach. The process entails converting the partial differential equations that regulate the flow of Prandtl nanofluids into a set of ordinary differential equations. Following this, the transformed equations are utilized in the construction of a reference dataset that supports the evolutionary methodology. The complex nanofluid flow velocities demonstrate a clear and direct relationship with non-dimensional parameters, such as the flexible number and Prandtl fluid parameter, in addition to the Hall and ion slip parameters. Temperature profiles demonstrate a strong correlation with the Brownian parameter and ion slip parameter as they increase, while the Prandtl number and thermal parameter exhibit an inverse correlation. In a similar fashion, the concentration profile exhibits an inverse correlation with the Hall parameter, Schmidt number, and concentration relaxation parameters, but a direct correlation with the Hartmann number. This study enhances the advancement of computationally efficient ANN models utilized in the analysis of complex fluid dynamics, thereby creating novel opportunities for their implementation across diverse domains.