Background: The utilisation of solar radiation for generating thermal energy has garnered a significant amount of attention, especially with the rising demand for sustainable power and heating sources. Nanofluids appeared as promising options for enhancing the efficiency of solar-thermal systems owing to their superior heat transmission properties. In this aspect, the present research deals with Ree-Eyring hybrid nanofluid ( SWCNT − MWCNT / H 2 O ) flow in the presence of thermal radiation effect and shape factor to examine the thermal behaviour in solar panels. Additionally, the innovation lies in the field of artificial neural networks and fluid flow analysis as an integrated approach. Methodology: The considered physical interpretation is stated in the form of nonlinear partial differential equations. To facilitate the computation process, the governing equations turned into nondimensional ordinary differential equations with the help of invariant transformations. The converted equations are treated scientifically using the NDSolve methodology, which automatically adjusts the step size. In further, an artificial neural network-based multi-layer perceptron deploying the Levenberg–Marquardt model is employed to forecast the measurements of physical quantities. Core Findings: The markable outcomes from this advanced work imply that the stretching surface has an important role in analysing the thickness of the boundary layer. The improving Weissenberg number enhances the boundary layer thickness, leading to a magnification in the momentum field of the non-Newtonian fluid. In addition, thermal radiation is captured as a second significant influence on temperature distribution. It follows that enhancing the Radiation parameter upsurges the thermal layer. The variation of the volume fraction of nanoparticles resulted in higher temperatures due to the Hamilton–Crosser model, which involves the cylindrical shape of the nanoparticles. Validation: The validation of the current work is explored by comparing it with the existing literature in certain instances. The incorporation of multi-layer perceptron for hybrid nanofluid delivers the performance capacity of high prediction with the error 4.83E−06, 1.11E−10 and 2.4E−08. Applications: The existing optimisation approach provides a new perspective on solar-powered offshore, solar vehicles, solar-powered charging stations, solar power greenhouse and solar water pump implementation.
Read full abstract