Design Improvement of a Vertically Oriented Thermal …
251
step size of 0.5 s. Different grid sizes with 16,000, 32,000 and 64,000 nodes are
considered. There is an increase in the total melting time of PCM by 1.59% as the
number of nodes is increased from 16,000 to 32,000, whereas there is no change in
the total melting time of PCM when the number of nodes is increased from 32,000 to
64,000. Therefore, the grid size with 32,000 nodes is considered further for the time
step independence study. No effect of time step size is found on the total melting
time of PCM. Consequently, 32,000 nodes with 0.5 s time step size were considered
suitable for the present study.
5 Results and Discussion
For the geometric configurations and flow behavior of the TES system considered
in the present study, the melting front propagates downward as observed from the
previously reported work [1]. This causes the late melting of PCM situated in the
bottom region of the TES system and affects the total melting time of PCM. At
the initial stage of melting, PCM near the HTF wall will be melted completely and
create a zone connecting the top and bottom regions. The instability results in the
establishment of a natural convection current which assists the melting of PCM. Since
natural convection is one the driving mechanism during melting, to take advantage
of it, it is proposed to provide a partition in the PCM domain which can create two
zones of circulation for more uniform melting of the PCM throughout the domain.
The effect of introducing the partition and its configuration on melting behavior and
thermal performance of the TES system is studied in the present work. Further, the
study is performed to determine the optimum PCM mass proportion in each zone
created by dividing the overall PCM domain. This is done because the total melting
time of PCM is not only affected by the type of partition but also by the PCM mass
in each zone. The results are presented and discussed in the following subsections.
5.1 Effect of Type of Partition
This section discusses the study on the effect of introducing the two types of partitions
(annular—case 1 and radial—case 2) in the simple TES system (base case). Effect
of introducing partitions is also analyzed on the performance of the TES system
by considering cases with partition (case 1 and case 2) and without partition (base
case). The temporal variation of a liquid fraction in the PCM domain is presented
in Fig. 2 for all the three cases studied. The total melting time of PCM for the base
case is found to be 256 min. It is noticed that as partition is introduced in the PCM
domain of the TES system, the total melting time of PCM is reduced. For case 1, the
total melting time of the PCM in inner and outer zones is found to be 108 min and
208 min, respectively. The total melting time of PCM for case 1 is 18.75% lower
than the total melting time for the base case. For case 2, the total melting time of the
251
step size of 0.5 s. Different grid sizes with 16,000, 32,000 and 64,000 nodes are
considered. There is an increase in the total melting time of PCM by 1.59% as the
number of nodes is increased from 16,000 to 32,000, whereas there is no change in
the total melting time of PCM when the number of nodes is increased from 32,000 to
64,000. Therefore, the grid size with 32,000 nodes is considered further for the time
step independence study. No effect of time step size is found on the total melting
time of PCM. Consequently, 32,000 nodes with 0.5 s time step size were considered
suitable for the present study.
5 Results and Discussion
For the geometric configurations and flow behavior of the TES system considered
in the present study, the melting front propagates downward as observed from the
previously reported work [1]. This causes the late melting of PCM situated in the
bottom region of the TES system and affects the total melting time of PCM. At
the initial stage of melting, PCM near the HTF wall will be melted completely and
create a zone connecting the top and bottom regions. The instability results in the
establishment of a natural convection current which assists the melting of PCM. Since
natural convection is one the driving mechanism during melting, to take advantage
of it, it is proposed to provide a partition in the PCM domain which can create two
zones of circulation for more uniform melting of the PCM throughout the domain.
The effect of introducing the partition and its configuration on melting behavior and
thermal performance of the TES system is studied in the present work. Further, the
study is performed to determine the optimum PCM mass proportion in each zone
created by dividing the overall PCM domain. This is done because the total melting
time of PCM is not only affected by the type of partition but also by the PCM mass
in each zone. The results are presented and discussed in the following subsections.
5.1 Effect of Type of Partition
This section discusses the study on the effect of introducing the two types of partitions
(annular—case 1 and radial—case 2) in the simple TES system (base case). Effect
of introducing partitions is also analyzed on the performance of the TES system
by considering cases with partition (case 1 and case 2) and without partition (base
case). The temporal variation of a liquid fraction in the PCM domain is presented
in Fig. 2 for all the three cases studied. The total melting time of PCM for the base
case is found to be 256 min. It is noticed that as partition is introduced in the PCM
domain of the TES system, the total melting time of PCM is reduced. For case 1, the
total melting time of the PCM in inner and outer zones is found to be 108 min and
208 min, respectively. The total melting time of PCM for case 1 is 18.75% lower
than the total melting time for the base case. For case 2, the total melting time of the