Economic Aspects of the Magnetocaloric Energy Conversion - Magnetocaloric Energy Conversion: From Theory to Applications - page 377

purpose of the comparison in Fig. 9.12 , the best results for the magnetic chiller were
taken (60 % volume fraction, thickness of the magnetocaloric material
100 microns). The results reveal that the most important issue to be addressed for
future research is to target substantial improvements in the frequency of operation,
which are very strongly related to the AMR geometry and heat transfer.
The authors have further emphasized [ 29 , 35 ] that devices based on permanent
magnets and the AMR principle have a certain restriction (upper limit) for cooling
power up to 200 kW, which is related to the geometry of the permanent magnet.
Namely, for a large cooling power the dimensions of the permanent-magnet
assemblies can be such that this can lead to substantially higher manufacturing
costs. The authors added an important note [ 29 , 35 ] that the problems associated
with cost, power density, compactness and ef ciency may be solved in the future;

however, with new approaches, which are for instance based on the application of
advanced working

uids, new heat transfer principles with the application of
thermal diodes, advanced design approaches, etc.
In addition, Egolf et al. [ 29 ] did a comprehensive analysis on the feasibility of a
superconducting rotary magnetic chiller. The study revealed that when magnetic
chillers are compared to conventional ones, this shows an advantage for both
ef

fl

ciency and cost. Namely, the cost ratio between the manufacturing costs of the
superconducting chiller versus compressor chillers were estimated to be from 0.55
to 0.75. However, the minimum cooling power for such devices should be not less
than 1 MW.
In 2011, a more accurate model for characterization of AMR s operation was

applied for an investigation of magnetic superconducting chillers by Kitanovski
et al. [ 36 ]. Since this was reported only internally we present here some of the
results of the calculations that were performed at the University of Ljubljana. For
the purpose of the study the supply and return temperatures for the chilled

'

fl uid

were taken to be 7/12 C, respectively. Such a

temperature difference at the heat source or heat sink is relatively low, i.e. about
5

° C and for the cooling fl uid 27/32 °

6 K can also be obtained with many different magnet-
ocaloric materials where the process of adiabatic magnetization (demagnetization)
is performed for a magnetic

6 K. A difference of 5

- -

eld change of less than 3 T, without the need for very

high magnetic elds. This is also why the analyses in the study considered magnetic

eld changes of superconducting magnets being 1.5, 2, and 3 T.
In the study it was assumed that the rotary superconducting chiller (i.e. the
rotation of the magnetic

eld source or the rotation of the AMRs) performs the

AMR Brayton-like cycle.
The numerical analysis of the characteristics of the operation at given parameters
was based on a dynamic simulation model, developed and described by Tu š ek et al.

in 2011 [ 37 ]. The following parameters were considered in the analysis:

￿ Heat-transfer fl uid: water;

Return temperature of the chilled fl uid 12 ° C;

￿

Supply temperature of the cooling fluid 27 ° C;

￿

Frequency of the operation <5 Hz;

￿