Predicting MHD mixed convection in a semicircular cavity with hybrid nanofluids using AI

Area of Science:

• Fluid Dynamics and Heat Transfer
• Nanomaterials Science
• Magnetohydrodynamics (MHD)

Background:

• Investigates magnetohydrodynamic (MHD) mixed convection within a semicircular enclosure.
• Focuses on heat transfer enhancement using various nanofluids and hybrid nanofluids, including Al2O3-TiO2-SWCNT-water.
• Examines the influence of a rotating inner cylinder on fluid flow and thermal performance.

Purpose of the Study:

• To numerically analyze MHD mixed convection heat transfer in a semicircular enclosure with a rotating cylinder.
• To evaluate the performance of different nanofluids and hybrid nanofluids, including Al2O3-water, TiO2-water, SWCNT-water, and Al2O3-TiO2-SWCNT-water.
• To develop and validate an artificial neural network (ANN) model for predicting heat transfer outcomes.

Main Methods:

• Numerical simulation of magnetohydrodynamic mixed convection.
• Utilized Al2O3-water, TiO2-water, SWCNT-water, and Al2O3-TiO2-SWCNT-water hybrid nanofluids.
• Developed an Artificial Neural Network (ANN) model for prediction, achieving high accuracy (97.34% training, 97.41% testing for average Nusselt number).

Main Results:

• Heat transfer increased significantly with higher cylinder rotation speeds (21.12% for Ω=10) and larger cylinder sizes (66.14% for SWCNT-water).
• Higher concentrations of SWCNT and Al2O3 in hybrid nanofluids improved heat transfer performance.
• Increased Hartmann number reduced heat transfer, while higher Richardson numbers enhanced it for SWCNT-water nanofluid.

Conclusions:

• Hybrid nanofluids, particularly those with higher SWCNT and Al2O3 content, offer superior heat transfer enhancement.
• Cylinder rotation speed and size are critical parameters for optimizing convective heat transfer in this configuration.
• The developed ANN model accurately predicts heat transfer performance, offering a valuable tool for design and optimization.