A comprehensive review on the application of nanofluid in heat pipe based on the machine learning: Theory, application and prediction

This paper introduces three paramount factors i.e. viscosity, thermal conductivity and stability that affect the application of mono hybrid nanofluids in heat pipes. The applications various types pipes are reviewed mechanism transfer enhancement or inhibition is summarized. machine learning (thermal dynamic viscosity) charged with presented. main challenges include: (1) difference uncertainty on as well undesirability property nanofluid; (2) lack comprehension time-dependent pipes; (3) limitation predictive models based learning; (4) an appropriate standard for selecting algorithm. To tackle above imminent challenges, further opportunities revealed including: exploring at nanoscale establishing unified standards, effect surfactant smaller particle size; focusing nanoparticle deposition layer; large, exclusive databases expanding input variables; defining specific by horizontal comparison using more advanced algorithms. review-based study provides guidelines development establishes foundation technology nanofluids. • There some differences uncertainties viscosity a techniques. thermo-physical properties should be explored. large established variables expanded.