Artificial intelligence – Numerical study of melting and solidification heat transfer in a bundle of petal tubes embedded in metal foam

Maximizing energy efficiency is a topic of great interest to scientists and engineers. Energy recovery through thermal energy storage (LHTES) units is a promising technique that contributes to improving energy efficiency. The current article examines the thermal performance in petal-shaped tubes implanted in a metal foam-phase change material (PCM) domain using the local thermal non-equilibrium model. The enthalpy–porosity method was employed for phase transition. The impact of petal number, amplitude, and tube position on melting and solidification of PCM was analyzed by numerical simulations, resulting in a dataset of 161,000 records. These records were used to train the artificial neural network (ANN) in order to optimize the LHTES unit. The numerical findings indicated that the petal shape significantly improved the thermal performance of the system compared to the normal tube shape. It was also that petal number, amplitude, and tube position remarkably affect the thermal activities within the LHTES unit. Increasing the amplitude from 0.3 to 0.6 improved the melting and the solidification time by 13.6 and 16.2 %, respectively. The ANN models successfully captured complex thermal interactions, offering a powerful predictive tool for optimizing LHTES systems.