Impact of different numerical approaches on the magnetocaloric effect modeling

As an ecological alternative to the conventional refrigeration technology, magnetocaloric refrigeration is still facing scientific and technological challenges hindering their application. Magnetocaloric devices rely on the magnetocaloric effect, where temperature variations result from magnetic field changes. The correct implementation of the magnetocaloric effect in numerical models is crucial before prototyping the related solutions. Here, we present a comparison between the three most used numerical methods to simulate the magnetocaloric effect: continuous temperature change, discrete temperature change step and heat source obtained from adiabatic temperature. By varying the time and space steps, it was observed that the continuous temperature change method is the most appropriate for small time steps, but has the largest computational cost. The discrete method can only be applied to small time steps, but is the fastest method. Finally, the adiabatic temperature change power source method can be applied in the entire range and is the one that presents the best results for larger time steps.