Computational intelligence paradigm with Levenberg-Marquardt networks for dynamics of Reynolds nanofluid model for Casson fluid flow

Abstract

The work in hand examines the dynamics of Reynolds nanofluid model for Casson fluid flow (RNFM-CFF) through a slandering sheet by using computational intelligence of Levenberg-Marquardt networks (CILMNs). For two-dimensional and two-directional flows, mathematical formulations of Reynolds nanofluid model are developed. The partial differential equations are reduced to ordinary differential equations by using appropriate transformations. The evaluation of the Prandtl number, Casson fluid parameter, velocity power index parameter, and Surface thickness parameter over velocity profiles and temperature profiles is accomplished on synthetic dataset for various variations of RNFM-CFF by exploiting CILMNs. For the investigation and examination of the estimated solution of the validation, training and testing process are utilized, and the performance is verified through error histogram, regression index, mean square error-based fitness in the range of 10−09 to 10−13.