Optimization of Active Magnetic Regenerative Refrigeration Systems Using Design of Experiments

This paper attends to demonstrate the usefulness of Design of Experiments (DOE) method in magnetic refrigeration (MR) understanding and optimization. A numerical DOE is applied to a simple 1D finite difference model describing an Active Magnetic Regenerative Refrigeration (AMRR) system. The heat transfer fluid is water, the regenerator consists of stacked gadolinium plates and the model is based on the assumption of an equivalent single plate. A two-level 2 fractional DOE based on Box methodology is used to evaluate the effect of seven parameters on the temperature span, namely material and fluid thicknesses, length, equivalent width, mass flow rate, cycle frequency and magnetocaloric effect (MCE).