Reduction of transient refrigeration time by modulation of fluid displacement ratio and operating frequency of a magnetocaloric device

摘要

In the literature, there are a lot of one-dimensional magnetic refrigeration model developed using lessormore well adapted exchange coefficients. In order to reduce assumptions and approximations, a globaltwodimensional(2-D) multiphysics model has been developed for an AMRR magnetocaloric device. Itintegrates successively:- a magnetostatic model based on a semi-analytical modeling of the magnetostatic phenomena, whichtakesinto account the nonlinear behaviour of the ferromagnetic external circuit as well as the activemagnetocaloric material (MCM) properties. The analytical model calculates the values of the internalmagnetic field and the internal magnetic flux density at each point of the regenerator volume;- a magnetocaloric model for calculating the magnetic power density produced in the magnetocaloricmaterial as a result of the magnetic field variation (provided by the magnetostatic model), combinedwiththe interpolation of the local magnetization of the magnetocaloric material (from experimental data) asafunction of local magnetic field and temperature values;- a thermo-fluidic model, which solves the energy and momentum equations using an implicit finitedifference method. It also calculates the heat capacity and the thermal conductivity of themagnetocaloricmaterial as a function of temperature and internal magnetic field, allowing to update the newtemperaturesof both the fluid and the material.The capacities and performances of the 2-D multiphysics model (Plait et al. (2021)) are studied by theinfluence of input parameters on the temperature difference and the time necessary to obtain thesteady state in adiabatic mode. The first study consists in calculating the influence of the two mainoperating parameters – the fluid displacement ratio A0 and the AMR frequency f – in order to obtain themaximal temperature difference between the two ends of the regenerator. The fluid displacement ratioranges from 5 % to 100 % of the channel volume for every specific operating frequency between 0.1 Hzand 1 Hz. The first figure shows the maximal temperature mapping obtained at steady state and permitsto observe a maximal span temperature of 16 K, with a combination {A0, f} = {25 %, 0.3 Hz}. However,some other combinations permit to obtain a similar temperature difference (in equipotential zonescombining whether lower A0 with higher f or combination of higher A0 with lower f). This kind of studywas realized in the literature (Bahl et al. (2008), Almanza et al. (2015)), which permits to validate thisstudy. Thus, it is interesting to identify which combination permits to reduce the time necessary toobtain these results.For that, the second study consists in calculating the transient temperature between the start and thestationary regime, allowing to determine the best combination {A0, f}, which ensures the fastest cooling rate. The second figure shows the mapping of the time necessary to achieve the steady state. A highfrequency combined to a high fluid displacement ratio permits to reduce the time necessary to obtainthesteady state. In our configuration the best combination to obtain the maximal temperature difference inaminimal time is {20 %, 0.5 Hz}.