Page 119 - Defrosting for Air Source Heat Pump

Modeling study on uneven defrosting 111

4.4 Concluding remarks

Following on the experimental work on defrosting performance of an experimental
ASHP unit having a three-circuit outdoor coil draining away locally the melted
frost by using water-collecting trays, a modeling study on the dynamic defrosting
process, at the two experimental settings of with and without the use of water-
collecting trays between circuits, was carried out. Two empirical models
corresponding to the two settings were therefore developed. It was the first time
for a set of defrosting models focusing on the uneven defrosting performance for
a multicircuit outdoor coil, but not for a whole heat pump system. It is also the first
time for a defrosting model to fully consider the melted frost locally drained away
and the energy stored in the metal of the coil. Additionally, the two models were
validated with experimental data, with the totally same experimental conditions
considered. Therefore, the two models were expected to be used to predict some
physical parameters difficult or hardly possible to be measured in a dynamic
defrosting process. Finally, the two models could help increase awareness and spur
efforts in exploring and maximizing the potential uses of ASHP units to realize more
energy savings.
Moreover, by using the validated defrosting Model 1 corresponding to a typical
traditional ASHP unit, without trays installed between circuits, the model extrapola-
tion was carried out. A series of modeling results was quantitatively predicted, includ-
ing the temperature of melted water on each circuit’s surface during defrosting, the
thermal resistance of the refrigerant during defrosting, the mass of melted frost during
defrosting, and the energy used from the refrigerant at each 5 s. Additionally, based on
the validated Model 1, three control strategies were proposed and tested in the model.
Results showed that the best operating defrosting performances in terms of defrosting
durations and energy use were achieved by fully closing the modulating valve on the
top circuit when its defrosting was terminated.
However, as discussed, the two validated models have their limitations, includ-
ing the listed assumptions, the validation errors, empirical formulas, and the exper-
imental data from the fixed configuration and operating conditions for an ASHP
unit, etc. For example, the frost accumulationoneachcircuit wasassumedevenly
distributed at the start of a defrosting process. But in practice, it is impossible that
the frost was evenly formed and accumulated on the surface of each circuit in a
vertically installed multicircuit outdoor coil. This was because of the uneven dis-
tribution for the refrigerant inside the tube and as the entrance air outside the tube.
Moreover, the refrigerant was also assumed evenly distributed during defrosting
for each circuit in the preliminary model extrapolation work. Clearly, this is impos-
sible because the heating loads imposed on the outdoor coil kept changing. The
uneven distributed frost on the surfaces of each circuit would further promote
uneven refrigerant distribution, and thus result in uneven defrosting. Therefore,
the frost accumulation condition at the start of a defrosting process and the refrig-
erant distribution during defrosting should be further studied, and are presented in
the next chapter.