Thermal performance enhancement of metal hydride reactor for hydrogen storage with graphene oxide nanofluid: Model prediction with machine learning

Abstract

Some metals and metal alloys can store gaseous hydrogen, making the storage of hydrogen in metal hydrides (MHs) possible. For the MH reactor to store hydrogen at a higher rate, improved heat transfer is required. The 2-D material graphene oxide (GO) attracted researchers’ attention due to its excellent thermal properties. The present work aims to improve heat transfer and hydrogen storage rate of the LaNi5 MH reactor. A 2-D axis-symmetric numerical model of the reactor is formed and simulated using COMSOL Multiphysics 5.6 software. Water and its based nanofluids (NFs), namely, GO, GO-SiO2 (50:50), GO-TiO2 (50:50), and Al2O3 are employed as heat transfer fluids (HTFs). The effect of inlet temperature and flow velocity of the HTF; and hydrogen supply pressure on the reactor performance is examined. The findings demonstrate that the storage rate is greatly improved by lowering the HTF inlet temperature; and increasing its inlet velocity and hydrogen supply pressure. In comparison to water and all other NFs, the GO NF with 1 vol% demonstrated comparatively better heat transfer. It reduces the duration by 61.7% of that of water to attain 90% hydrogen storage capacity at similar conditions. The data acquired in the numerical investigations were used to build a prediction metamodel using the evolutionary machine learning (ML) technique of gene expression programming (GEP).