Optimization of a novel magnetic refrigerator based on the demagnetizing effect using a particle swarm-like algorithm

Abstract

The current state of the art of magnetic refrigerator prototyping relies on changes in magnetic field magnitude for promoting heat transfer processes from cold to hot regions. Although this is the most widespread method, some disadvantages arise, such as requiring massive magnets and complex operation mechanisms to alternate the field intensity. Recently, a new operation approach has been proposed using the demagnetizing effect of polycrystalline and anisotropically shaped magnetocaloric materials, which explores the rotation of the magnetic field rather than changes in its magnitude. In this work, we numerically study the performance of a magnetic refrigerator based on this novel approach using a finite element method, considering the half-plate and half-fluid channel model. After performing 92 simulations with random parameters, the no-load temperature span of the system was optimized using a particle swarm-based algorithm to determine the global best. The initial global best from the simulated results was 5.89 K. A new global best of 7.27 K was obtained after applying the algorithm, which represents an increase of over 20%. For the same volume of magnetocaloric material, the temperature span decreases with the inverse aspect ratio. The maximum temperature span of the remaining inverse aspect ratios was successfully determined. In particular, it was possible to increase the temperature span by 22.91% and 18.28% for the inverse aspect ratios 0.26 and 0.4, respectively. This study highlights the feasibility of a novel kind of magnetic refrigerators based on the demagnetizing field.