Integrating artificial intelligence with numerical simulations of Cattaneo-Christov heat flux on thermosolutal convection of nano-enhanced phase change materials in Bézier-annulus

Abstract

The numerical analysis based on incompressible smoothed particle hydrodynamics (ISPH) is introduced to examine the impacts of Cattaneo-Christov (Ca-Ch) heat flux and exothermic chemical reaction on thermosolutal convection of nano-enhanced phase change materials (NEPCM) in Bézier-annulus. The used annulus is formed between inner connected Bézier curves and outer connected spline-Bézier curves. The inner shape of connected Bézier curves is maintained at Th&Ch, left/right walls of spline-Bézier curves are kept at Tc&Cc and other walls are adiabatic. The governing equations, after being converted into non-dimensional form, have been solved by ISPH which is an accurate meshless algorithm the treatment of internal flows inside complex geometries. The simulations are executed for Frank-Kamenetskii number FK, Ca-Ch heat flux δC, buoyancy ratio parameter N, Soret-Dufour numbers SrDu, Rayleigh number Ra, and nanoparticle parameter φ on thermosolutal convection of a suspension fluid. From the numerical simulation values for the average Nusselt and Sherwood numbers (Nu¯, Sh¯) were obtained for some fluid flow scenarios. Then, based on those values an artificial neural network (ANN) model was developed to predict the values of Nu¯, and Sh¯ without the need to perform the ordinary simulations again for the new cases which is a high cost compared to ANN. From the obtained simulations, it was concluded that the ANN model is an accurate tool to be used to predict the needed values. Also, the Frank-Kamenetskii number significantly influences the enhancement process of the temperature distributions and velocity field as well as phase change material in Bézier-annulus.