Design optimization and thermal analysis of C₂H₆O₂ hybrid nanofluid flow over solar plate under solar radiation using artificial neural network

This study, we investigate how solar radiation affects C₂H₆O₂ hybrid nanofluid over solar plate under LTNE (local thermal non-equilibrium effects). This study will be useful for applications pertaining to thermal management, renewable energy technologies, and industrial heating systems. The study of Stefan blowing and solar radiation effects on ethylene glycol-based hybrid nanofluids under LTNE conditions provides heat transport enhancement to advance the performance of solar thermal collectors, heat exchangers, electronic cooling devices, thermal storage units, and other manufacturing and chemical engineering processes. The HAM (Homotopy analysis method) is used to first obtain the analytical solution. Our model is unusual in that it uses artificial intelligence neural networks (AI-NN) in combination with the back-propagated Levenberg-Marquardt algorithm (BLM), resulting in a more sophisticated computational solution than traditional numerical methods. The resultant synthetic dataset is divided into three sections: testing (15%), validation (15%) and training (70%). The resultant PDE was converted into an ODE. The artificial neural networks (ANNs) are evaluated for accuracy, validity, and efficiency using regression analysis, MSE (mean-squared error), EH (error histograms), and an examination of the relationship between numerical repetitions using the stochastic (Levenberg-Marquardt) approach. As the Stefan blowing factor values rise, the velocity field improves while the temperature and concentration characteristics decrease.