Browsing by Author "Borjini M.N."

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Control of Magnetohydrodynamic Mixed Convection and Entropy Generation in a Porous Cavity by Using Double Rotating Cylinders and Curved Partition
    (American Chemical Society, 2021) Hassen W.; Selimefendigil F.; Ben Khedher N.; Kolsi L.; Borjini M.N.; Alresheedi F.
    In this work, mixed convection and entropy generation analyses in a partitioned porous cavity with double inner rotating cylinders are explored under magnetic field effects. A curved partition shape is considered with identical rotating cylinders and an inclined magnetic field, while the right vertical wall moves with a constant speed in the y-direction. Numerical simulations are performed by considering various values of Rayleigh number, Hartman number, Darcy number, inclination of the magnetic field, size of the curved partitions, and rotational speeds of the inner cylinders and their vertical locations with the cavity. Complicated flow field with multicellular structures are observed due to the complex interaction between the natural convection, moving wall, and rotational effects of inner cylinders. Improved heat-transfer performance is obtained with higher values of magnetic field inclination, higher values of permeability/porosity of the medium, and higher rotational speeds of the cylinders. Almost doubling of the average Nu number is obtained by decreasing the value of the Hartmann number from 25 to 0 or varying the magnetic field inclination from 90 to 0. When rotational effects of the cylinders are considered, average heat-transfer improvements by a factor of 5 and 5.9 are obtained for nondimensional rotational speeds of 5 and −5 in comparison with the case of motionless cylinders. An optimum length of the porous layer is achieved for which the best heat-transfer performance is achieved. As the curvature size of the partition is increased, better heat transfer of the hot wall is obtained and up to 138% enhancement is achieved. Significant increments of entropy generation are observed for left and right domains including the rotating cylinders. The magnetic field parameter also affects the entropy generation and contributions of different domains including the curved porous partition. © 2021 The Authors. Published by American Chemical Society.
  • No Thumbnail Available
    Item
    Jet impingement cooling using shear thinning nanofluid under the combined effects of inclined separated partition at the inlet and magnetic field
    (Springer Science and Business Media Deutschland GmbH, 2022) Selimefendigil F.; Kolsi L.; Ayadi B.; Aich W.; Alresheedi F.; Borjini M.N.
    Combined effects of using inclined partition and magnetic field on the cooling performance of double slot jet impingement are analyzed with finite element method. Two different shear thinning nanofluids are used while experimental data is available for the rheological properties. Different values of of Reynolds number (Re between 100 and 1000), velocity ratio (VR, between 0.2 and 1), opening ratio (OR, between 0.05 and 0.95), magnetic field strength (Ha, between 0 and 30) and inclination of partition (Ω , between 0 and 40) are used. It is observed that varying VR of the jets, size/inclination of the partition, magnetic field strength and nanfluid type, can be used to control the local and average convective heat transfer and cooling performance features effectively. The average Nusselt number (Nu) rises with higher VR while at the highest VR the amount of increments are 23.5% and 28.5% with first (NF1) and second (NF2) nanofluid (NF). When magnetic field is imposed, effects of OR becomes important with NF1 at the lowest strength of magnetic field. Average Nu reduces with higher magnetic field strength for NF1 while 14.4 % reduction for the highest strength at OR = 0.95 is achieved. However, for NF2 the trend is opposite and 18.8 % increment is obtained. Variations in the average Nu becomes 7.6 % and 1.8 % for NF1 and NF2 when inclination of the partition is changed. The cooling performance is estimated by using a feed-forward network modeling approach in terms of average Nu for NF1 and NF2 by using 25 neuron in the hidden layer. © 2022, The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature.