380 J. Therm. Sci., Vol.28, No.2, 2019
layer. Chen et al. [17] established the evaporator surface
water vapor mass transfer mathematical model, and the
frost layer growth process for certain kinds of finned tube
evaporator outside surface frosting process change and
frost thickness were simulated, and analyzed the
parameter change on the influence of the thickness of the
frost. Cui et al. [18] used a new CFD model to predict the
formation and growth of frost formation on the flat
surface, which could better reflect the effect of surface
structure on the growth of frost layer in the initial stage
and the full development stage. However, the macro scale
modeling fails to capture the frost layer local ice nucleus
formation and growth of ice branch of dynamic change,
and is also unable to obtain the influence of topological
characteristics of frost layer on frost layer physical
properties (average thickness, average density, average
solid phase volume fraction, etc.).
Therefore, this paper attempts to establish a lattice
Boltzmann (LB) model for the frost growth process by
first coupling the nucleation probability model with the
enthalpy-LB model from the mesoscopic scale. The
process of thickening and densification of the growth of
the overall frost layer can be simulated at the
macroscopic scale, and the dynamic change process of
the frost layer structure caused by the ice branch growth
in the frost layer can also be described on the
microscopic scale, and the frost layer can be obtained
during the growth of the frost layer. The average
thickness, average density, frost quality and other
non-steady changes in the internal physical quantity of
the frost layer could be studied by examining the
influence of factors such as cold surface temperature,
relative humidity, and surface infiltration characteristics
on the frosting characteristics.
2. Mathematic Model
The frosting process is a complicated physical process
of multiphase flow heat transfer, which involves the
process of the formation and migration of phase interface
and the process of transfer and diffusion of interphase
mass (heat). Lattice Boltzmann method, as an emerging
meso-scale simulation method, has natural advantages in
capturing and tracking the phase interface.
The initial stage of frost formation is a complicated
process of random nucleation. The existing lattice
Boltzmann method (LBM) multiphase flow model cannot
accurately describe this process. This article introduces
the probability function and the multiphase flow phase
change model to accurately describe this complex
process.
The mathematical model is mainly divided into three
sub-models: (1) The multiphase flow model is used to
solve the physical quantities of the flow field and
temperature field during the frosting process, and
describe the formation and migration process of the
phase interface caused by frost formation (2) The
probabilistic model is used to simulate the formation of
ice cores during the initial stage of frosting. (3) The phase
transition model is used to simulate the growth of ice
nucleus and the growth process of frost formation.
2.1 Multiphase model
Lattice Boltzmann method is a method of mesoscopic,
which describes the motion of a fluid particle distribution
function with discrete velocities on a fixed lattice. The
evolution equation of the particle distribution function is
expressed as [19]:
,,
1
,,
xe x
xx
iitti
eq
ii
tft
ftft
(1)
The time step is
t
/
x
c
. If a uniform grid is used, the
lattice spacing is calculated by
/
xx
LN
. The
evolution process is divided into two processes:
Collision process:
1
,,,,xxxx
eq
iti i i
tftftft
(2)
Migration process:
,,xe x
iitti
tft
(3)
D3Q19 model is used in this article. Fig. 1(a) gives
discrete velocities in D3Q19 model. This discrete
velocity is shown as [20]
01 100001 11 11 1 1100 0 0
0001 1001 1 1100 0 01 11 1
000001 100 0 01 11 11 1 11
e c
The equilibrium particle distribution function is given as
2
2
242
1
22
eu
eu
u
eq
i
i
ii
ss
f
ccc
(4)
where, the lattice speed of sound is defined as c
s
= 3c .
The weight coefficients are:
2
22
22
0
1/3
1/18
1/36
2
e
e
e
i
ii
i
c
c
(5)
The kinematic viscosity of the fluid is
2
0.5
t
vc
(6)