Renewable Energy 181 (2022) 472e489

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: [Link]/locate/renene

Effects analysis on heat dissipation characteristics of lithium-ion

battery thermal management system under the synergism of phase
change material and liquid cooling method
Feng Yi a, Jiaqiang E a, Bin Zhang c, d, Hongyan Zuo b, *, Kexiang Wei b, **, Jingwei Chen a,
Hong Zhu d, Hao Zhu a, Yuanwang Deng a
a
College of Mechanical and Vehicle Engineering, Hunan University, Changsha, 410082, China
b
Hunan Provincial Key Laboratory of Vehicle Power and Transmission System, Hunan Institute of Engineering, Xiangtan, 411104, China
c
School of Mechanical Engineering, Hunan Institute of Engineering, Xiangtan 411104, China
d
Hunan Haibo Ruide Electric Intelligent Control Technology Co., Ltd, Xiangtan, 411100, China

a r t i c l e i n f o a b s t r a c t

Article history: In the work, a composite cooling system coupled with PCM (phase change material) and liquid cooling
Received 11 June 2021 was designed. The influence of parameters such as spacing, EG (expanded graphite) content, battery
Received in revised form direction, coolant flow rate and pipe diameter on the cooling performance was analyzed. The results
4 August 2021
revealed that: (1) The increase of spacing makes the temperature of the pack decrease, but the effect
Accepted 17 September 2021
degree of increasing the spacing on temperature performance decreases when the spacing exceeds
Available online 20 September 2021
14 mm. At the same time, CPCM formed by the added high thermal conductivity material expanded
graphite and paraffin can effectively improve the temperature control performance of the pack, which
Keywords:
Battery thermal management
can control the temperature at 313.15K under the condition of an ambient temperature of 308.15K (2) By
Phase change material comparing the maximum temperature and the temperature difference of the pack under different
Liquid cooling discharge methods, it is concluded that the poles facing inward can reduce the accumulation of the
Lithium-ion battery interior heat. (3) The convective heat transfer coefficient and Nu number of the pack are increasing as the
flow rate increases, but the slope of the temperature versus flow rate decreasing slowly. After increasing
the inlet size, the inner wall surface area of the pipe increases and the average convective heat transfer
coefficient of the pack reduces. However, the Nu number increases with the increase of the pipe
diameter, indicating that the growth of the pipe diameter can enhance the heat transfer rate. (4)
Orthogonal experiment and fuzzy correlation analysis were combined to analyze the parameters that
affect the thermal performance of the BTMS. It was determined that using a proper coolant flow rate can
maximize the thermal performance of the pack.
© 2021 Elsevier Ltd. All rights reserved.

1. Introduction between 20
C and 40
C, and the temperature difference cannot

exceed 5
C. Working outside this temperature range will have a
Lithium-ion batteries have the following advantages: high en- serious impact on the charging and discharging performance of
ergy, high specific power, long cycle life, and short charging time lithium battery [4e7]. Therefore, a suitable BTMS is extremely
[1,2]. Compared to many other types of power batteries, lithium-ion important to the performance and safety of the pack [8e10]. The
batteries have good overall performance, so most electric vehicles main function of BTMS is to absorb or transfer the heat generated
use lithium-ion batteries as the main energy carrier nowadays [3]. by the battery out of the module through the medium, and the
However, internal chemical reaction of lithium batteries is easily current thermal management method is realized by three cooling
affected by temperature. Studies have shown that the operating media mainly: air-based [11], liquid-based [12], and PCM-based
temperature of lithium-ion batteries needs to be maintained [13]. Each of the cooling methods has its advantages and disad-
vantages [14]. The latest research trend is to utilize multiple cooling
methods for coupled heat dissipation, which has better heat
dissipation performance than a single method [15]. Due to the large
* Corresponding author.
** Corresponding author.
size and capacity of the prismatic battery, the temperature differ-
E-mail addresses: zuohongyan18@[Link] (H. Zuo), wei-kx@[Link] (K. Wei). ence during charging and discharging may increase over time [16].

[Link]
0960-1481/© 2021 Elsevier Ltd. All rights reserved.
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Nomenclature TPCM the temperature of PCM, K

HPCM the enthalpy of the PCM, J
PCM phase change material L the latent heat of the PCM, J/kg
EG expanded graphite T0 the temperature of ambient, K
CPCM composite phase change material Ts the temperature of solid phase, K
BTMS battery thermal management system Tl the temperature of liquid phase, K
V0 the volume of battery, m3 rl the density of the coolant, kg/m3
E0 battery open circuit voltage, V cpl heat capacity of the coolant, J/(kg$K)
U the voltage of the battery, V kl the thermal conductivity of coolant, W/(m$K)
T battery temperature, K tl the temperature of coolant, K
Q± the heat generation of the pole, W !v the velocity vector of coolant, m/s
R± the resistance of the pole, U$m qave the average heat flux density, W/m2
Lelectrode the length of the pole, m Tw the average pipe wall temperature, K
s the cross-sectional area of the pole, m2 Tf average coolant temperature, K
Kx, Ky, Kz thermal conductivity of the cell in the x, y, and z D the diameter of the pipe, m
directions, W$m$K uwater the velocity of the water, m/s
Lx the thickness of the cell, kg vwater the dynamic viscosity of water, m2/s
Li the laminated thickness of i-material, m de the equivalent diameter of the pipe, m.
Ki the thermal conductivity of the i-material
respectively, W/(m$K) Greek letter
Mcell-interior the mass of the cell, kg relectrode the resistivity of the pole, U$m
Vcell-interio the volume of the cell, m3 rwick the density of the wick, kg$m3
Ccell-core the heat capacity of the cell, J/(kg$K) ε the porosity of the wick
mi the mass of the i-material of the cell, kg rv the density of the steam, kg$m3
Ci average heat capacity of the i-material, J/(kg$K) pv the saturation pressure of the steam, Pa
kwick the thermal conductivity of the wick, W$m-1$K-1 m the dynamic viscosity of the steam, N$s/m2
Cwick the specific heat capacity, J,kg-1,K-1 rPCM the density of PCM, kg$m3
dv the cross-sectional diameter of the steam core, m q the liquid fraction of the PCM
H the latent heat of water, kJ/kg lw the thermal conductivity of the water, W/(m$K)
Rv the thermal resistance of steam rwater the density of water, kg$m3
Tv the temperature of steam, K ▽ Gradient operator
kPCM the thermal conductivity of PCM, W$m-1$K-1

Characteristics of the PCM make the battery temperature change pack within a certain range, some auxiliary mechanisms are needed
very small during the phase change process, so using the PCM is an in order to improve the thermal performance of the pack in order to
effective solution [17,18]. In addition, the PCM-based cooling smoothly take the heat out. For instance, Xu et al. [28] founded that
method can also prevent battery aging, according to the research of adding EG and copper foam could enhance the conductivity of silica
Lv et al. [19] used PCM as a cooling medium which can increase the plate and the method based on natural convection and forced
cycle life of the battery by 65.3%. convection could not meet the temperature requirements during
Several standards to be taken into considered such as thermal the experiment. However, combined with the silica plate and the
conductivity, heat storage capacity, and operating temperature liquid cooling system can effectively transfer the heat out of the
range to select a suitable PCM [20]. However, the thermal con- pack through the water flowing in the pipe, which could bring a
ductivity of pure PCM is not sufficient to control the battery tem- good cooling effect. Bai et al. [29] studied the temperature control
perature, adding high thermal conductivity materials to enhance performance of the pack coupled with a cooling plate and PCM. The
the heat of transmission properties usually, such as metallic ma- illustrate show that the presence of the cooling plate could improve
terials, nanomaterials and porous materials [21e24]. For example, the temperature uniformity in the pack effectively. Chen et al. [30]
huang et al. [25] used porous structured EG as an additive for PCM founded that the combination of PCM and heat pipe could improve
to improve the structural stability and thermal conductivity of its the cooling performance of the BTMS, and the heat pipe was in
own. The results demonstrate that the increase of EG mass fraction contact with the battery and PCM at the same time. Ping et al. [31]
will increase the thermal conductivity of CPCM (Composite Phase studied a composite cooling system composed of PCM and liquid
Change Material) and improve thermal performance of BTMS. cooling tube. The results show that increasing the flow rate of the
Similarly, there are some other ways to alleviate problem. Choud- pipeline coolant could reduce the maximum and minimum tem-
hari et al. [26] added a fin structure to PCM as an optimized perature of the battery module. Similarly, Hekmat et al. [32]
structure. The effects of fin structure, PCM thickness, number of embedded the pipe into the PCM and founded that the use of PCM
fins, and external heat transfer coefficient on the temperature of can improve the temperature uniformity of the pack greatly. Zheng
lithium-ion batteries were studied. Luo et al. [27] wrapped a layer et al. [33] coupled the PCM and liquid cooling methods to studied
of graphite film on the outer surface of the battery to enhance heat the influence of factors including the flow direction of the coolant,
transfer before putting the battery into the PCM module. Under the the flow rate of the coolant, the presence or absence of a thermally
premise that no additional cooling system is required, the insulating intermediate layer, and the filling of the composite PCM
maximum temperature of the pack at 4C discharge rate is 33
C, and on the thermal performance of BTMS.
the temperature difference is 1.4
C. The above review shows that couples PCM and liquid cooling
Although the use of PCM can maintain the temperature of the can achieve better cooling performance. However, few researchers

473
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

pay attention to the degree of influence of each parameter on heat (1) Both liquid paraffin and coolant are incompressible Newto-
dissipation performance of BTMS, and the priority of each param- nian fluids.
eter is not considered when designing the thermal management (2) Thermal properties of all materials are constant.
system. Therefore, we establish a three-dimensional thermal model (3) The average specific heat capacity for cell unit and PCM in
which coupling PCM and liquid cooling plate to analyze the various applied the simulation.
parameters that affect BTMS, and obtain its effect on the heat (4) The physical properties of the batteries are assumed to be
dissipation performance of BTMS. Then, combined with orthogonal homogeneous during the simulation.
experiment and fuzzy gray relational analysis, the influence degree (5) Thermal contact resistance is not considered.
of the four parameters on each index of BTMS is analyzed. Distin-
guished from the existing work, the following major contributions
are made in the present study. 3.1. Model thermogenesis analysis

(1) Developed a hybrid thermal management system that com- The heat generate by battery can be divided into cell generation
bines PCM with liquid cooling panels, and the influence of and pole generation, cell heat generation in this article is using
the physical properties and thickness of different PCMs on BERNARDI model [40].
heat dissipation performance of BTMS was explored. It also


recommends to use reasonable types and dosages of PCM in I dE
BTMS. q¼ E0  U  T 0 (1)
V0 dT
(2) Under different PCM types and dosages, the impact of the
coolant flow rate, coolant type, and battery orientation on where V0 is the battery volume, m3; E0 is the voltage of the battery
the pack temperature and PCM utilization rate was explored in an open circuit condition, V; U is the voltage of the battery, V; T is
in detail. Revealing the coupling effects of the nature and the temperature of the battery, K; dE0/dT is the temperature in-
amount of PCM, battery orientation, coolant flow rate and fluence coefficient.
flow channel diameter on BTMS. Generally, the material of the electrode is a metal with good
(3) Fuzzy gray correlation analysis is used to analyze the key conductivity, such as copper, aluminum and other metals with low
parameters that affect the performance of the pack, and the resistance, but the current collected on the pole is relatively large,
priority of the key parameters' impact on the performance of so the heat generation of the pole cannot be ignored. The heat
the BTMS is obtained, which provides a reference for the generation rate of the pole can usually be described by the
subsequent optimization research of the battery thermal following formula:
management system.
8
> 2
> Q± ¼ I R±
>
>
>
>
< Q±
q± ¼ (2)
2. Thermal management model of battery pack V
>
> electrode
>
>
>
>
: R± ¼
relectrode Lelectrode
The battery cooling module which contains 16 commercial
s
LiFePO4 batteries designed in this paper is shown in Fig. 1. The
parameters of the battery are shown in Table 1.
where Q± is the heat generation of the pole, W; R± is the resistance
The surface area and battery capacity of the prismatic battery
of the pole, U$m; relectrode is the resistivity of the pole, U$m; Lelec-
are large, so the temperature unevenness of the prismatic battery
trode is the length of the pole, m; s is the cross-sectional area of the
will be larger than that of the general battery [34], PCM has the
pole, m2. The calculation results are presented in Table 3.
characteristic of small temperature change during the phase
change process and surrounds it around the square battery to
reduce the temperature difference between the batteries [35], the 3.2. Heat conduction
specific parameters of which are shown in Table 2. In order to avoid
the phenomenon of heat accumulation in the battery pack, a Analyzing the structure of the battery in the x direction (thick-
separator is utilized to separate the battery from the battery. The ness direction), it can be seen that the thermal conductivity of the
separator can improve the temperature uniformity of the battery battery in the thickness direction can be regarded as connecting the
pack and prevent the PCM from leaking during the phase change thermal conductivity of various materials in series. For the y and z
process [36]. In addition, in order to prevent the heat transfer directions, it can be regarded as connecting the thermal resistance
deterioration of the PCM after it has completely melted, it is an in parallel. Based on this assumption, the mathematical expression
effective way to reconcile PCM with heat pipe [37]. Heat pipes with of the thermal conductivity of the cell in the x, y, and z directions is:
high thermal conductivity are utilized to transfer the heat gener- 8
> Lx
ated by the pack to the liquid cooling plate. At the same time, the >
> Kx ¼ P L
>
>
part of the battery in contact with the liquid cooling plate directly >
>
i
>
> Ki
can transfer heat to the liquid cold plate. Using coolant to take heat >
> i
>
<
out of the battery module is an effective cooling method [38], the XKi Li
Ky ¼ (3)
liquid is delivered to the liquid cooling plate through the cooling >
> Lx
>
>
liquid system and the heat generated in the battery is removed of >
>
i
>
>
the battery module. >
> XKi Li
>
>
: Kz ¼ Lx
i
3. Mathematical module
where Kx, Ky, Kz are the thermal conductivity of the cell in the x, y,
In order to simplify the physical problem, some assumptions and z directions, W/(m$K); Lx is the laminated thickness of the cell,
need to be made as follows: m; Li is the laminated thickness of i-material, m; Ki represents the
474
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 1. Thermal management system model.

Table 1
The parameters of square battery.

Paremeter Value Paremeter Value

Positive electrode LiFeCoPO4 Size(mm) 260  110  25

Negative electrode Graphite Internal resistance(mU) z2

Rated capacity(A$h) 50 Rated voltage(V) 3.2
Working temperature(
C) 20~þ50 Changing/Discharging cut-off voltage(V) 3.6/2.75

Table 2
Physical parameters of phase change materials [39].

Paraffin Mass Fraction (%) Specific Heat (J$kg1K1) Thermal Conductivity (W$m1K1) Latent Heat (kJ$kg1) Melting Temperature(K) Density (kg$m3)

100 2300 0.282 196.62 310.15e316.15 800

95 2227.5 1.361 194.69 310.15e316.15 800
90 2155 6.395 187.88 310.15e316.15 800

Table 3
Discharge characteristics of cell.

Discharge rate I/A q/W qþ/W q-/W

0.5C 25 2153.84 92.67 1236.94

1C 50 8455.3 309.03 4947.91 Mcellinterior
2C 100 32062.12 1236.11 19791.67 rcellcore ¼ (4)
3C 150 70820.45 2781.25 44531.25
Vcellinterior

where Mcell-interior is the mass of the cell, kg; Vcell-interio is the vol-
ume of the cell, m3.
The heat capacity of the cell can be obtained by summing the
thermal conductivity of the i-material respectively, W/(m$K). product of the mass of the material and its own specific heat ca-
The calculation formula of average density of the battery is as pacity and the ratio of the mass to the heat capacity of the cell. The
follows: mathematical expression is:

475
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

P 8
mi Ci > vrl  
>
>
> þ V, rl ! v ¼0
Ccellcore ¼ i
(5) >
> vt
Mcellinterior >
< 
v   
>
rl !
v þ V, rl ! v! v ¼ Vp (13)
where Ccell-core is the heat capacity of the cell, J/(kg$K); mi is the >
> vt
> 
>   
>
> v
mass of the i-material of the cell, kg; Ci is the average heat capacity : rl cpl tl V, rl cpl !
v tl ¼ V,ðkl Vtl Þ
of the i-material, J/(kg$K). vt
The three areas of the heat pipe consist of a copper tube (shell),
wick, and a steam wick inside. The copper shell can be directly where rl is the density of the coolant, kg/m3; cpl is heat capacity of
regarded as heat conduction, and the characteristics of the wick can the coolant, J/(kg$K); kl is thermal conductivity of coolant, W/
be calculated according to the following formula [41]: (m$K); tl is the temperature of coolant, K; is the velocity vector of
coolant, m/s.
kl cu ½ðkl cu þ ks cu Þ  ð1  εÞðkl cu  ks cu Þ The average heat transfer coefficient between the fluid and the
kwick ¼ (6)
½ðkl cu þ ks cu Þ þ ð1  εÞðkl cu  ks cu Þ coolant flow channel needs to be based on the formula:
 
rwick ¼ εrl cu þ ð1  εÞrs cu (7) qave ¼ h Tw  Tf (14)

εrl cu Cl cu þ ð1  εÞrs cu Cs cu where qave is the average heat flux density, W/m2; h is the average
Cwick ¼ (8) heat transfer coefficient, W/(m2$K); Tw and Tf are the average pipe
rwick
wall temperature and the average coolant temperature respec-
where kwick is the thermal conductivity of the wick, W$m1$K1; tively, K.
rwick is the density of the wick, kg$m3; Cwick is the specific heat In the study of convective heat transfer, the Nu number (Nusselt
capacity, J,kg1,K1; ε is the porosity of the wick. The subscripts number) is also an important criterion number that characterizes
lcu and scu represent liquid and sintered copper powder, the convection strength of the cooling liquid and BTMS, and its
respectively. calculation formula is:
The physical parameters of the vapor can be calculated by the
h,D
following formula [42]: Nu ¼ (15)
lw
d2v H 2 rv pv
kvapor ¼ (9) where D is the diameter of the pipe, m; lw is the thermal con-
24mRv Tv2 ductivity of the water, W/(m$K).
The Reynolds number (Re) is defined by:
where dv is the cross-sectional diameter of the steam core, m; H is
the latent heat of water, kJ/kg; rv is the density of the steam, Kg$m3; de uwater rwater
pv is the saturation pressure of the steam, Pa; m is the dynamic Re ¼ (16)
vwater
viscosity of the steam, N$s/m2; Rv is the thermal resistance of
steam; Tv is the temperature of steam, K. The physical parameters of where uwater is the velocity of the water, m/s; rwater is the density of
the heat pipe are obtained from Ref. [43]. water, kg$m3; vwater is the dynamic viscosity of water, m2/s; de is
The heat transfer of PCM can be expressed by the following the equivalent diameter of the pipe, m.
equations [44]:
3.3. Boundary conditions and initial conditions
vHPCM
rPCM ¼ V$ðkPCM VTPCM Þ (10)
vt The initial temperature of the entire system is set to 308.15K.
The inlet of the coolant is set as the velocity inlet and the incoming
ð
TPCM
flow velocity is set to 0.04 m/s, and the temperature is set to
HPCM ¼ cPCM dT þ qL (11) 308.15K. The boundary condition of the outlet is set as the pressure
T0 outlet. In addition, the heat transfer coefficient of natural convec-
tion between the wall of the pack and the outside is set to 5W/
8 (m2$K) [46,47], and the environment is set to a temperature of
>
> 0
> 308.15K. According to calculations, the Reynolds numbers of cool-
< TPCM  TS TPCM < Ts ants with a flow velocity between 0 and 0.1 m/s are all less than
q¼ TL  TS
Ts  TPCM < Tl (12)
>
> 2300, so the laminar flow model can be used for calculation.
>
: TPCM  Tl
1
3.4. Grid independence
where rPCM, kPCM are the density and thermal conductivity of PCM
respectively. TPCM is the temperature of PCM, K; HPCM is the Five thermal management system models with different grid
enthalpy of the PCM, J; L is the latent heat of the PCM, J/kg; q is the numbers were simulated with Fluent 20.0, and the model with a
liquid fraction of the PCM; T0, Ts and Tl are the temperature of spacing of 14 mm was selected as the experimental object for grid
ambient, solid phase and liquid phase respectively, K. independence verification which is shown in Fig. 2(a), and the
The energy conservation equation, continuous equation, and number of grids is 356344, 670064, 780030, 936870, 1213799. The
momentum conservation equation are given as follows [45]: result of grid verification is shown in Fig. 2(b). The highest

476
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

temperature tends to be stable when the number of grids reach the same after discharge end. However, there is a deviation be-
670064, so the number of grids used for analysis and calculation in tween the experimental and the simulated in the process of dis-
this article will be about to 670064. charging the battery. The reason may be that the thermistor
installed on the surface of the battery exchanges heat with the air,
which causes a temperature error. In addition, changes in the
3.5. Model validation ambient temperature may also affect the experimental result
[48e50]. All in all, the battery model is accurate enough to research
The experimental condition of the pack is placed in a large space the cooling performance of BTMS.
for a discharge test at an ambient temperature of 300K and the
discharge rate of 1C. The environmental convective heat transfer
3.6. Uncertainty analysis
coefficient is 5W/(m2$K). In the experiment, the cell voltage, cur-
rent and other data were recorded by the BTS20.5V/4  200A, and
The uncertainty in the experiment mainly comes from tem-
the K-type thermocouple was used to collect the temperature
perature sensors, charging and discharging machines and ambient
change of the battery surface during the discharge process, the two
temperature. The charging and discharging machine used in this
installation positions of which are described in Fig. 3(a). The com-
paper is BTS250-5V/4  200A produced by Hubei Depu Electric
parison between the experimental measurement and the data
Company, the maximum current is 200A, and the tester could
obtained through the simulation is shown in Fig. 3(b). It can be
produce the current/voltage pattern with an accuracy and stability
observed that the numerical results are basically consistent with
of ±0.1% full scale. The thermocouple used to measure the surface
the experimental values, and the battery temperature is basically
temperature of the battery is a K-type thermocouple, and its tem-
perature measurement deviation is 0.1
C. Temperature control
accuracy of environmental chamber is within 0.1
C. Therefore, the
uncertainty in the temperature measurement was within 0.5
C
built on the root mean square method [51].

4. Analytical methods

EG content, spacing, coolant flow rate, and inlet size all affect the
temperature control performance of BTMS. In order to distinguish

Fig. 2. Simulation model grid and independence verification. Fig. 3. Experimental validation of cell.

477
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

the influence of the primary and secondary or interaction of these

different factors on the heat dissipation performance of the pack, 23 2 3
x1 x1 ð1Þ x1 ð2Þ/x1 ðnÞ
the structure of the BTMS was optimized and the pump energy loss 6 x2 7 6 x2 ð1Þ x2 ð2Þ/x2 ðnÞ 7
was reduced. Analytical methods will combine orthogonal experi- x¼6 7 6
4 « 5¼4 «
7
5 (17)
« « «
ments and fuzzy gray correlation analysis to optimize the param- xm xm ð1Þ xm ð2Þ/xm ðnÞ
eters that affect the thermal performance of the pack.
where x represents the comparison sequence, and x1, x2, …, xm are
the factors that affect the conclusion of the experiment. In this
article, they are the content of EG, spacing, coolant flow rate and
4.1. Orthogonal experimental design flow channel diameter. And x1(1), x1(2),$$$, x1(m) are the values of
the influencing factors x1 under different working conditions.
Orthogonal experimental design, also known as multi-factor Define the reference sequence. The sequence obtained by
optimization design, is an effective mathematical statistical orthogonal analysis as the reference sequence to improve the
method for rationally arranging and analyzing various experi- credibility of the reference sequence. The reference sequence ob-
mental factors scientifically. The research found that the four pa- tained by the orthogonal matrix in this article is:
rameters of EG content, spacing, coolant flow rate, and inlet size
have a greater impact on the performance of BTMS. They are in
y ¼ ½yð1Þ yð2Þ/ yðnÞ (18)
contact with each other in structure. The various factors are related
to each other and have a greater impact on the thermal perfor-
where y is the reference sequence, and the reference sequence in
mance of the BTMS designed in this paper. Therefore, four pa-
this article is the relevant temperature parameter of the battery
rameters are chosen as the factors of the orthogonal experiment.
thermal management system obtained by orthogonal experiment.
Table 4 shows the orthogonal table selected in this article.
(2) Standardization of the original data series

The original data obtained must be normalized first. The original

4.2. Fuzzy gray relational analysis
data may have large numbers or small data, it may cause by
different dimensions rather than its intrinsic nature [55]. The
Gray relational analysis is part of the most important aspects of
calculation formula as follows:
gray system theory. It is built on the measurement of similarity or
difference between data sequences. The purpose of gray correlation
analysis is to explore the qualitative and quantitative relationships xi ðkÞ  minxi ðkÞ
xi ðkÞ0 ¼ (19)
between the main system factors, to capture their dynamic char- maxxi ðkÞ  minxi ðkÞ
acteristics in the development process, and to measure the relative
influence of the sequence in order to compare with the reference where xi(k) is the original data; min xi(k) is the minimum value in
sequence [52]. In fact, it represents a measure of the absolute value the comparison sequence; max xi(k) is the maximum value in the
of the data difference between sequences, and can be used to assess comparison sequence; xi(k)0 is the result of standardization
the approximate correlation between sequences with a small processing.
amount of data. Therefore, the gray correlation coefficient can
represent the relationship between the expected and actual (3) Find the cosine value of fuzzy membership
experimental results, and the gray correlation level is calculated at
the same time corresponding to each performance feature [53,54]. P
n
On the basis of this method, this chapter combines orthogonal yðkÞxi ðkÞ
k¼1
experiment and fuzzy membership evaluation method, and ex- ri ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (20)
plores the four factors that affect pack temperature, temperature Pn
2 Pn
yðkÞ xi ðkÞ2
difference, pressure difference, and Nu number. The steps are as k¼1 k¼1
follows:

(1) Define the data series (4) Calculate the correlation coefficient

Table 4
Influencing factors and levels.

Case Factors

EG Content Spacing(mm) Coolant Flow Rate (m/s) Inlet Size (mm)

1 paraffin 12 0.02 15
2 paraffin 14 0.05 18
3 paraffin 16 0.08 21
4 5% EG CPCM 12 0.05 21
5 5% EG CPCM 14 0.08 15
6 5% EG CPCM 16 0.02 18
7 10% EG CPCM 12 0.08 18
8 10% EG CPCM 14 0.02 21
9 10% EG CPCM 16 0.05 15

478
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 4. Influence of spacing on maximum temperature and volume fraction.

Fig. 5. Temperature difference and temperature change curve of pack.

½c; 1:5c c < 1=3

Dmin þ hDmax h2 (26)
xi ðkÞ ¼ (21) ½1:5c; 2c c  1=3
Di ðkÞ þ hDmax
In the formula, if c < 1/3, then h ¼ 1.25k; if c  1/3, then
h ¼ 1.75c.
Dmin ¼ minfminjxi ðkÞ  yðkÞjg (22)
(5) Calculate the Euclidean gray correlation level
Dmax ¼ maxfmaxjxi ðkÞ  yðkÞjg (23) W ¼ ðw1 ; w2 ; /; wt Þt ¼ 1; 2; /; n (27)

where w1,w2, …, wt represent the weight coefficients under various

DðkÞ ¼ jxi ðkÞ  yðkÞj (24)
working conditions. The model in this paper does not have a
mutual influence relationship, 1/n is used as the weight coefficient,
where h is the resolution coefficient, its size is related to the data
and the European gray correlation degree is:
difference.
The average of the absolute difference is: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u n
uX   2
rij0 ¼ 1  2t w 1  zij ðkÞ (28)
1 X m Xn
k¼1
D¼ jxi ðkÞ  yðkÞj (25)
m,n i¼1
k¼1

Based on k ¼ D/Dmax, the calculation formula of the resolution

coefficient is: (6) Calculation of fuzzy association degree:
479
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 6. Effects of varying spacing on pack temperature.

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi of electric vehicles, and the change of the spacing will affect the
u  2
u 2 quality of the PCM at the same time. Therefore, the influence of the
tr ij þ r 0ij
Rij ¼ (29) change of the spacing and the PCM with different EG content on the
2 temperature of the battery module is simultaneously analyzed.
Fig. 4 shows the effect of changing two parameters of spacing
and EG content on the temperature performance of BTMS. From the
curve shown in Fig. 4(a), the maximum temperature drops fastest
5. Results and discussion
when the spacing is increased from 10 mm to 12 mm, the reason of
this phenomenon is that the mass of the PCM is too little at 10 mm,
5.1. Influence of spacing and EG content on thermal performance of
and the increased spacing of 2 mm can provide more space to
BTMS
accommodate more PCM, which can absorb more heat due to latent
heat storage enhanced. But the amount of PCM is not as much as
The appropriate spacing should be selected in the limited space
480
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

possible, which can be seen as the slope of the temperature versus constantly decreasing and tends to be constant during the process
spacing reduces. Although the improvement of EG mass can of increasing the spacing after the use of CPCM, and the effect of
improve the thermal conductivity of CPCM, it also reduces the increasing the spacing on the temperature difference gradually
latent heat of CPCM, thereby, comparing the difference of the weakens after the spacing exceeds the optimum distance, the
temperature of the three temperature curves, it is not difficult to reason of this phenomenon is that the quality of the PCM is also
see that the maximum temperature of the pack will decrease with increasing, which can absorb more heat. Fig. 5(b) is the transient
the increase of EG content, but the trend of which is slowing down temperature curve of the pack, it is not difficult to find that the heat
with the increase of EG content. In addition, judging from the generated by the battery at the beginning of the discharge causes
histogram of the change in the liquid phase fraction of the PCM in the temperature to continue to rise, the slope of temperature versus
Fig. 4(b). The liquid phase fraction of the PCM is close to 100% while time curve reduces because the PCM is closed its melting point and
the battery spacing is 10 mm, indicating that the amount of PCM is can absorb more heat due to latent heat storage. This is explained
insufficient at this time, which can also explain the maximum by the fact that the PCM starts to phase change (from solid to liquid)
temperature drop of the battery is greatest between 10 and 12 mm and release the latent heat during the phase change process, then
in the previous article. The liquid fraction of the three PCM grad- the temperature curve has a second turn when all the PCM are
ually drops with the increase of the distance, 10% EG CPCM drops transformed from solid to liquid completely. In addition, increase in
fastest, 5% EG CPCM is the second, and the decrease rate of paraffin EG mass fraction allows the heat generated by the battery to be
wax is the slowest. When the distance is 16 mme18 mm, the liquid better conducted to the outside world instead accumulate inside,
phase fraction decreases the fastest. At this time, the amount of therefore, the pack the time for the second turning point is delayed
PCM has exceeded the optimal thickness, and continuing to grow as the content of EG increases. The temperature difference curve
the spacing cannot effectively improve the heat dissipation ca- and the temperature rise curve has the same laws and principles.
pacity of BTMS. The temperature cloud map of the pack after the discharge is
Fig. 5(a) shows the effect of spacing change on battery tem- analyzed in order to discuss the temperature distribution of the
perature difference. The temperature difference of the pack is not pack in Fig. 6, which is a cloud diagram of the influence of changing
noticeable when the PCM is paraffin. Comparing the minimum the spacing on the pack temperature under the same PCM. The
temperature change curve of the pack, it can be seen that the highest temperature of the pack is concentrated at the position
minimum temperature change of the pack is larger when the PCM away from the cooling plate when spacing is 10 mm, the reason for
is paraffin, which is due to the low thermal conductivity of paraffin this appearance is that the quality of PCM is too small makes a high
making the minimum temperature of the pack higher than that of temperature of the battery core. It is not difficult to observe that the
pack using CPCM. The temperature difference of the pack is temperature on the left side of the pack is higher than that on the

Fig. 6. (continued).

481
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

right, the left side is close to the coolant inlet and the temperature can be seen that orientation of the battery will affect the heat
in the inlet direction is definitely lower than the temperature in the accumulation of the pack. Due to the non-heating insulating cover
outlet direction. In the process of increasing the spacing from in the direction of the pole, the heat generation in the direction of
10 mm to 18 mm, the maximum temperature of the pack and the the pole is lower than that of the end of the battery. When the poles
temperature of the middle part where heat accumulation is likely of the battery are in adjacent positions, the heat generation in the
to occur continue to decrease, the phase change latent heat direction of the poles will be lower than the heat generation at the
released by the PCM between the batteries also increase and which end of the battery, so there will be no obvious heat accumulation
can effectively absorb the heat released by cell during the phase effect and the temperature will be lower than the former.
change process, so the temperature in the middle of the battery
decreases as the distance increases. 5.3. Effect of flow rate on BTMS
In order to analyze the effect of changing the PCM under the
same spacing, simulation results are presented in Fig. 7. The heat It can be observed that the maximum temperature of the pack
dissipation performance of the pack increases gradually with the and the liquid fraction of the PCM decrease with the increase of the
increase of the EG content, and the specific manifestation is that the inlet flow rate in Fig. 10(a). The maximum temperature of the pack
heat accumulation effect in the middle of the pack weakens with is 315.64K and 314.40K when the flow rate is 0.01 m/s to 0.1 m/s
the increase of the EG content in Fig. 7(a). From the perspective of respectively. From the battery temperature distributions, it can be
temperature distribution, core temperature of the pack keeps seen that the core temperature of the pack is lower by enhance flow
declining with the increase of EG content, the reason of which is rate, which means lower plate temperature enhances heat transfer
that the temperature inside the pack can be better conducted from within the pack. Fig. 10(b) describe the temperature difference and
the pack with the increase of thermal conductivity of the PCM. pressure difference of the pack with the inlet flow rate. it can be
Fig. 7(b) is cloud chart of PCM liquid fraction over time, the liquid clear that the temperature difference of the pack decreases with the
fraction of the PCM is gradually increasing as the discharge process increase of the inlet flow rate. When the inlet flow rate increases
continues. In addition, the increase of EG content can delay the from 0.01 m/s to 0.1 m/s, the temperature difference of the pack
increase of the liquid fraction. drops from 4.79K to 4.13K. At the same time, the pressure differ-
The curve of the liquid phase fraction of different PCM with time ence increases with the increase of the flow velocity. When the
under the same spacing is illustrated in Fig. 8(a). The liquid fraction inlet flow velocity increases from 0.01 m/s to 0.1 m/s, the pressure
of the three PCM will suddenly rise after it reach PCM's melting difference of the flow channel increases from 0.83Pa to 9.73Pa. The
point, and the higher thermal conductivity of the PCM, the growth temperature difference and the cooling hydraulic pressure differ-
rate of the liquid fraction decreases faster, it is likely that the PCM ence of the pack present an opposite state when the flow rate
with high thermal conductivity can dissipating heat from the pack changes. From the temperature distribution of pipeline section, the
in a short period of time, and the PCM with high thermal conduc- reason is that increasing the flow rate can reduce the coolant and
tivity has a lower liquid phase fraction at the end of the discharge at pipe wall temperature.
the same time. The energy stored by the PCM after the discharge is In order to quantitatively analyze the above phenomenon, two
shown in Fig. 8(b). The heat absorbed by the three PCMs is criterion numbers, the convective heat transfer coefficient h and
increasing between 10 mm and 16mm, and the higher the thermal the Nusselt number Nu will be used for analysis and discussion.
conductivity of the PCM, the less heat absorbed by comparing three Fig. 11 indicates the heat exchange performance of the pack with
curves, the reason of this is that the PCM with high thermal con- the flow rate. It can be seen from the curve that when the flow
ductivity conducts the heat generated by the battery to the liquid velocity increases from 0.01 m/s to 0.1 m/s, the h of the pack in-
cooling plate faster, so the heat stored by the PCM is lower. In creases from 346 W m2 K1 to 679.17 W m2 K1. The growth rate
addition, when the spacing is increased from 16 mm to 18 mm, the curve of the convective heat transfer coefficient shows that the
stored heat of 5% EG CPCM and 10% EG CPCM does not increase but growth rate of the h is continuously decreasing, which means in-
decrease. This is attributed to the increase in spacing leads to an crease the inlet flow rate can increase the convection, but the
increase in the contact area between the PCM and the liquid cooling growth rate is already very small, and more additional pumping
plate, which make the PCM with strong thermal conductivity energy needs to be provided. When the flow velocity increases
conducts the heat to the liquid cooling plate more through thermal from 0.01 m/s to 0.1 m/s, the Nu number of the pack increases from
conduction. 8.35 to 16.43. The Nu number and h have the same changing law,
indicating that the heat exchange capacity of the pack and the
5.2. The influence of the battery direction on BTMS coolant is also increasing. From the analysis of the growth rate
curve of Nu number, which has the highest growth rate when the
Considering the heat generation of the battery poles and the flow velocity is 0.02 m/s.
insulating plastic inside the pack, the placement direction of the
battery is also a factor that may affect the battery temperature. The 5.4. Influence of different inlet sizes on the temperature of BTMS
battery modules with a spacing of 14 mm are selected as an
experimental object, the results obtained as shown in Fig. 9. Except for the inlet flow rate has a greater impact on the per-
According to Fig. 9(a), placing the pole inward can reduce the formance of the BTMS, the inlet radius of the coolant has an impact
maximum temperature of the pack. But the influence of the cell on the temperature control capability and temperature uniformity
orientation on the temperature of the pack is gradually reduced as of the BTMS. With the increase of the diameter of the inlet size, the
the thermal conductivity of the PCM increases, which are attributed flow rate flowing through the liquid cooling plate and the contact
to the increase in thermal conductivity of PCM can alleviate the area of the flow channel and the fluid also increases, the temper-
phenomenon of heat accumulation. Fig. 9(b) shows the influence of ature distribution of pack reveals that the increase in the diameter
the pole placement direction on the battery temperature differ- of the pipe also enhances the ability of the battery module to
ence. It can be seen from the curve that the battery poles facing conduct heat outward. It can be seen that the maximum temper-
inward can reduce the temperature difference of the pack. In order ature of the pack is 315.54K while the inlet size is 6 mm, and the
to explore the reason, Fig. 9(c) and (d) are temperature cloud dia- maximum temperature of the pack decreases to 314.51K when it
grams corresponding to different battery placement directions. It increases to 21 mm in Fig. 12(a), and the liquid fraction of the PCM
482
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 7. Pack temperature and PCM liquid fraction over time.

483
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 8. Liquid fraction change and energy storage curve of PCM.

Fig. 9. Effect of battery orientation on the heat dissipation performance of BTMS.

484
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

Fig. 10. Effect of flow rate on BTMS performance. Fig. 12. Effect of inlet size on BTMS performance.

Fig. 11. Effect of flow rate on h and Nu. Fig. 13. Effect of inlet size on h and Nu.

is gradually decreasing too. Fig. 12(b) shows the curve of the tem- highest convective heat transfer coefficient when the inlet diam-
perature difference and the pressure difference of the pipeline after eter is 6 mm, the reason is that the thermal boundary layer of the
changing the inlet diameter. At this time, as the inlet diameter in- coolant at 6 mm is thicker than at 22 mm according Fig. 12. The
creases, the pressure difference and temperature difference are value of the h gradually decreases with the increase of the inlet
decline slightly, but a larger pipe radius means that a larger pump diameter, but the decrease rate of h gradually descends with the
which requires a larger installation space. increase of the inlet diameter according to the decrease rate curve
The flow channel convective heat transfer coefficient h and the of the h. The reason is that as the inlet diameter increases, the
decline rate of h, the Nu number, and the growth rate of Nu number contact surface area between the coolant and the liquid-cooled
change with the increase of the inlet diameter in Fig. 13. There is the plate is also increasing. Although the average heat transfer

485
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

coefficient is constantly decreasing, the total heat transfer is The calculation result is displayed in Fig. 14. The European gray
constantly increasing. Judging from the growth curve of Nu num- correlation between the four factors and the maximum tempera-
ber, the heat transfer intensity of the system is continuously ture of the pack is basically the same, while the fuzzy membership
increasing, which also verifies that the aforementioned total between the coolant flow rate and the maximum temperature of
average heat transfer coefficient is continuously increasing. the packs is the largest. According to formula (20), the greater the
degree of fuzzy membership, the better the consistency of the
change trend between the studied factor and the maximum tem-
5.5. Results of fuzzy gray relational analysis
perature, which means that the coolant flow rate has a greater
impact on the maximum temperature of the battery pack. Ac-
EG content, spacing, coolant flow rate, and flow channel diam-
cording to formula (28), the correlation difference between the four
eter have a significant impact on the heat dissipation of the pack. In
factors and the maximum temperature is close. In addition, fuzzy
this part, the fuzzy gray relational analysis method is used to
correlations of the four factors (EG content, spacing, coolant flow
analyze the influence degree of each factor on the performance of
rate, inlet size) and the maximum temperature of the pack are
BTMS. First, the results obtained by orthogonal experiments
0.575, 0.649, 0.699, and 0.688. Therefore, the degree of influence on
(maximum temperature, temperature difference, pressure differ-
the maximum temperature of the pack in the BTMS is arranged in
ence and Nu number) are used as the reference matrix Y1(k), Y2(k),
order of the coolant flow rate, the inlet size, the spacing and the EG
Y3(k), Y4(k), the time point where the liquid phase fraction is 77% is
content in descending order. In addition, the fuzzy gray correlation
selected as the reference sequence, and its matrix is Y5(k). EG
between the spacing and the EG content also exceeds 0.5, indi-
content, spacing, coolant flow rate and inlet size will be treated as
cating that the maximum temperature of the pack is also significant
x1(k), x2(k), x3(k), x4(k). The reference sequence and comparison
influences.
sequence obtained by numerical simulation are:
2 3 (b) Effect analysis on temperature difference based on fuzzy gray
0 0 0 5% 5% 5% 10% 10% 10%
6 12 relational analysis
14 16 12 14 16 12 14 16 7
x¼6
4 0:02
7
0:05 0:08 0:05 0:08 0:02 0:08 0:02 0:05 5
15 18 21 21 15 18 18 21 15 The influence of the four factors on the temperature difference
of the pack in Fig. 15. The largest fuzzy membership with the
temperature difference is the inlet size, and the European gray
y1 ¼ ½324:08 322:51 318:43 319:48 318:67 319:07 correlation grade between four factors and pack temperature dif-
316:11 312:22 312:28 ference is closed. According to formula (20), the greater the degree
of fuzzy membership, the better the consistency of the change
y2 ¼ ½5:13 5:68 3:38 4:51 3:59 3:79 1:25 2:13 2:21 trend between the studied factor and the temperature difference,
which means that the coolant flow rate and the temperature dif-
ference have the highest degree of similarity. In addition, according
y3 ¼ ½1:14 2:92 4:97 2:44 7:17 0:9 5:18 0:72 3:83
to formula (28), the correlation difference between the four factors
and the maximum temperature is close. In addition, the fuzzy gray
y4 ¼ ½10:09 14:50 18:75 16:67 15:81 10:68 16:56 11:72 12:50 correlations between EG content, spacing, coolant flow rate and
inlet size and pack temperature difference are 0.586, 0.694, 0.670,
y5 ¼ ½294; 354; 384; 840; 846; 726; 1140; 1158; 1120 and 0.702. Therefore, in order to reduce the temperature difference
of the pack and avoid local deterioration, a proper flow channel
diameter must be adopted first.
(a) Effect analysis on maximum temperature based on fuzzy
gray relational analysis (c) Effect analysis on pressure difference based on fuzzy gray
relational analysis

Fig. 14. Relational grades of maximum temperature. Fig. 15. Relational grades of difference temperature.

486
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

seen that the coolant flow rate and the temperature difference have
the highest degree of similarity and smaller correlation differences.
In addition, the fuzzy gray correlation degree of EG content,
spacing, coolant flow rate, inlet size and pressure difference is
0.6766, 0.6811, 0.9267, 0.6183. Sorting according to the results, the
four factors have a descending effect on the cooling hydraulic
pressure difference: coolant flow rate, spacing, EG content, and
inlet size. In order to reduce the energy consumed by the cooling
pump in the cooling process, the appropriate flow rate and spacing
should be selected first.

(d) Effect analysis on Nu number based on fuzzy gray relational

analysis

The degree of influence of different factors on the Nu number of

the pack is presented in Fig. 17. The highest degree of fuzzy mem-
bership and European gray correlation with the Nu number of the
pack is the coolant flow rate. According to formulas (20) and (28), it
Fig. 16. Relational grades of difference pressure.
can be seen that the coolant flow rate and the Nusselt number have
the highest degree of similarity and the smaller degree of correla-
tion difference. In addition, the fuzzy gray correlation degrees of EG
content, spacing, coolant flow rate, and inlet size are 0.627, 0.649,
0.886, and 0.756, respectively. According to the calculation results,
the factors that affect the Nu number of the pack are sorted, coolant
flow rate has the largest impact, followed by the diameter of the
flow channel and the spacing. EG content has the lowest impact on
the Nu number of the pack. Therefore, when choosing to increase
the Nu number of the pack, choosing an appropriate coolant flow
rate is a great method.

(e) Effect analysis on melting time based on fuzzy gray relational

analysis

The greater the degree of fuzzy membership, the better the

consistency of the changing trends of the studied factors. It dem-
onstrates in Fig. 18 that the type of PCM has the best consistency
with the melting rate of the PCM. And the result of European gray
correlation provides that the correlation differences between PCM
Fig. 17. Relational grades of Nu number. types and melting time is small. At the same time, the type of PCM
and the melting time of PCM have the highest European gray cor-
relation and fuzzy gray correlation. Therefore, the type of PCM is a
The influence of four factors on the pressure difference is
key factor in determining the melting time.
illustrated in Fig. 16. The fuzzy membership and European gray
correlation between the coolant flow rate and the pressure differ-
6. Conclusion
ence are the highest. According to formulas (20) and (28), it can be
Nowadays, due to the shortage of resources and environmental
pollution, new energy vehicles [56e58] have being paid more and
more attention for researchers today. In this work, a BTMS coupled
CPCM and liquid cooling is developed in combination with related
theories for the ATL-R5B0S4 battery produced by ATL. Analyzed
related parameters: the influence of spacing, EG content, direction,
separator, coolant flow rate and inlet size on the cooling perfor-
mance of BTMS. The fuzzy gray correlation analysis is used to
analyze the influence of several important parameters (EG content,
spacing, coolant flow rate and inlet size) that affect the perfor-
mance of the pack on the maximum temperature, temperature
difference, pressure difference and Nu number of BTMS. The main
conclusions of this article are as follows:

(1) Increasing the spacing can improve the thermal performance

of the battery thermal management system. When the
spacing increases from 12 mm to 14 mm, the maximum
temperature drops most obviously, but after increasing the
spacing, the temperature optimization performance of the
Fig. 18. Relational grades of melting time. battery thermal management system gradually weakens. In
487
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

addition, the liquid fraction of the three phase change ma- References
terials gradually decreases with the increase of the spacing,
of which 10% EG CPCM decreases the most, 5% EG CPCM is [1] J. Wen, D. Zhao, C. Zhang, An overview of electricity powered vehicles:
lithium-ion battery energy storage density and energy conversion efficiency,
the second, and the paraffin wax decreases the slowest when Renew. Energy 162 (2020) 1629e1648.
the spacing is 16 mme18 mm, and the amount of phase [2] J.Q. E, F. Yi, W.J. Li, et al., Effect analysis on heat dissipation performance
change material has exceeded the optimal thickness at this enhancement of a lithium-ion-battery pack with heat pipe for central and
southern regions in China, Energy 226 (2021) 120336.
time. [3] Y. Li, K. Li, Y. Xie, J. Liu, C. Fu, B. Liu, Optimized charging of lithium-ion battery
(2) The temperature cloud chart of the battery module explains for electric vehicles: adaptive multistage constant currenteconstant voltage
the reasons for the influence of changing the battery direc- charging strategy, Renew. Energy 146 (2020) 2688e2699.
[4] X. Luo, Q. Guo, X. Li, Z. Tao, S. Lei, J. Liu, L. Kang, D. Zheng, Z. Liu, Experimental
tion on the temperature of the battery pack. The pole facing investigation on a novel phase change material composites coupled with
inward can reduce the heat accumulation inside. graphite film used for thermal management of lithium-ion batteries, Renew.
(3) The increase of the flow rate can improve the thermal per- Energy 145 (2020) 2046e2055.
[5] W. Nam, J.Y. Kim, K.Y. Oh, The characterization of dynamic behavior of Li-ion
formance of the cooling system. When the flow rate is set to
battery packs for enhanced design and states identification, Energy Convers.
0.02 m/s, the growth rate of the heat exchange capacity be- Manag. 162 (2018) 264e275.
tween the liquid-cooled plate and the coolant is the largest. [6] S. Panchal, M.H. Akhoundzadeh, K. Raahemifar, M. Fowler, R. Fraser, Heat and
With the increase of the flow rate, the growth rate of the heat mass transfer modeling and investigation of multiple LiFePO4/graphite bat-
teries in a pack at low C-rates with water-cooling, Int. J. Heat Mass Tran. 135
exchanges intensity decrease. Increasing the inlet diameter (2019) 368e377.
will lead to a decrease in the convective heat transfer coef- [7] H. Liu, Z. Wei, W. He, et al., Thermal issues about Li-ion batteries and recent
ficient, but the Nu number is increasing, indicating that progress in battery thermal management systems: a review, Energy Convers.
Manag. 150 (2017) 304e330.
increasing the inlet diameter can improve the heat transfer [8] J. Kim, J. Oh, H. Lee, Review on battery thermal management system for
intensity. electric vehicles, Appl. Therm. Eng. 149 (2019) 192e212.
(4) The biggest influence on the maximum temperature of the [9] P. Ping, R. Peng, D. Kong, G. Chen, J. Wen, Investigation on thermal manage-
ment performance of PCM-fin structure for Li-ion battery module in high-
pack is the flow rate of the coolant, followed by the inlet size, temperature environment, Energy Convers. Manag. 176 (2018) 131e146.
the spacing and the EG content. The factors that affect the [10] Z. Rao, Z. Qian, Y. Kuang, et al., Thermal performance of liquid cooling based
battery temperature difference is arranged in descending thermal management system for cylindrical lithium-ion battery module with
variable contact surface, Appl. Therm. Eng. 123 (2017) 1514e1522.
order: inlet size, spacing, coolant flow rate and EG content. [11] J.Q. E, M. Yue, J. Chen, H. Zhu, Y. Deng, Y. Zhu, F. Zhang, M. Wen, B. Zhang,
The influence of factors on the cooling hydraulic pressure S. Kang, Effects of a different air cooling strategies on cooling performance of a
difference from large to small is: coolant flow rate, spacing, lithium-ion battery module with baffle, Appl. Therm. Eng. 144 (2018)
231e241.
EG content and runner diameter. In order to reduce the en-
[12] J.Q. E, S. Xu, Y. Deng, W. Zuo, H. Wang, J. Chen, Q. Peng, Z. Zhang, Investigation
ergy consumed by the cooling pump in the cooling process, on thermal performance and pressure loss of the fluid cold-plate used in
an appropriate flow rate should be selected first. The factors thermal management system of the battery pack, Appl. Therm. Eng. 145
that affect the Nu number of the pack are sorted from largest (2018) 552e568.
[13] K. Jiang, G.L. Liao, J.Q. E, et al., Thermal management technology of power
to smallest, coolant flow rate has the largest impact, followed lithium-ion batteries based on the phase transition of materials: a review,
by the diameter of the flow channel, and then the spacing, EG Journal of Energy Storage 32 (2020) 2352, 152X.
mass fraction has the smallest impact on the Nu number. [14] J. Jaguemont, N. Omar, P. Bossche, et al., Phase-change materials (PCM) for
automotive applications: a review, Appl. Therm. Eng. 132 (2018) 308e320.
[15] T.H. Tran, S. Harmand, B. Sahut, Experimental investigation on heat pipe
cooling for Hybrid Electric Vehicle and Electric Vehicle lithium-ion battery,
J. Power Sources 265 (2014) 262e272.
[16] M. Fleckenstein, O. Bolhen, M. Roscher, et al., Current density and state of
CRediT authorship contribution statement charge inhomogeneities in Li-ion battery cells with LiFePO4 as cathode ma-
terial due to temperature gradients, J. Power Sources 196 (2011) 4769e4778.
Feng Yi: Resources, Methodology, Software, Data curation, [17] J. He, X. Yang, G. Zhang, A phase change material with enhanced thermal
conductivity and secondary heat dissipation capability by introducing a bi-
Writing e original draft, Writing e review & editing. Jiaqiang E: nary thermal conductive skeleton for battery thermal management, Appl.
Methodology, Writing e review & editing. Bin Zhang: Software, Therm. Eng. 148 (2019) 984e991.
Writing e review & editing. Hongyan Zuo: Funding acquisition, [18] W. Yuan, X. Yang, G. Zhang, et al., A thermal conductive composite phase
change material with enhanced volume resistivity by introducing silicon
Software, Data curation, Writing e review & editing. Kexiang Wei:
carbide for battery thermal management, Appl. Therm. Eng. 144 (2018)
Methodology, Data curation, Writing e review & editing. Jingwei 551e557.
Chen: Data curation, Writing e review & editing. Hong Zhu: Data [19] Y. Lv, X.Q. Yang, G.Q. Zhang, Durability of phase-change-material module and
curation, Writing e review & editing. Hao Zhu: Data curation, its relieving effect on battery deterioration during long-term cycles, Appl.
Therm. Eng. 179 (2020) 1359e4311.
Writing e review & editing. Yuanwang Deng: Conceptualization, [20] P. Kumar, D. Chaudhary, P. Varshney, et al., Critical review on battery thermal
Writing e review & editing. management and role of nanomaterial in heat transfer enhancement for
electrical vehicle application, The Journal of Energy Storage 32 (2020) 102003.
[21] W. Situ, G. Zhang, X. Li, et al., A thermal management system for rectangular
LiFePO4 battery module using novel double copper mesh-enhanced phase
Declaration of competing interest change material plates, Energy 141 (2017) 613e623.
[22] M. Mehrabi-Kermani, E. Houshfar, M. Ashjaee, A novel hybrid thermal man-
agement for Li-ion batteries using phase change materials embedded in
The authors declare that they have no known competing copper foams combined with forced-air convection, Int. J. Therm. Sci. 141
financial interests or personal relationships that could have (2019) 47e61.
appeared to influence the work reported in this paper. [23] Z.T. Yu, X. Fang, L.W. Fan, et al., Increased thermal conductivity of liquid
paraffin-based suspensions in the presence of carbon nano-additives of
various sizes and shapes, Carbon 53 (2013) 277e285.
[24] P. Gilart, A. Martínez, M.G. Barriuso, et al., Development of PCM/carbon-based
Acknowledgements composite materials, Sol. Energy Mater. Sol. Cells 107 (2012) 205e211.
[25] Q. Huang, X. Li, G. Zhang, et al., Thermal management of Lithium-ion battery
pack through the application of flexible form-stable composite phase change
This work is supported by the Natural Science Foundation of materials, Appl. Therm. Eng. 183 (2021) 116151.
Hunan Province (Grant No. 2021JJ50019) and the Aid program for [26] V.G. Choudhair, A.S. Dhoble, S. Panchal, Numerical analysis of different fin
structures in phase change material module for battery thermal management
Science and Technology Innovative Research Team in Higher
system and its optimization, Int. J. Heat Mass Tran. 163 (2020) 120434.
Educational Institutions of Hunan Province [No. (2014) 207].
488
F. Yi, J. E, B. Zhang et al. Renewable Energy 181 (2022) 472e489

[27] X. Luo, Q. Guo, X. Li, et al., Experimental investigation on a novel phase change [43] Y.H. Gan, L.F. He, J.L. Liang, et al., A numerical study on the performance of a
material composites coupled with graphite film used for thermal manage- thermal management system for a battery pack with cylindrical cells based on
ment of lithium-ion batteries, Renew. Energy 145 (2020) 2046e2055. heat pipes, Appl. Therm. Eng. 179 (2020) 115740.
[28] Y. Xu, X. Li, X. Liu, et al., Experiment investigation on a novel composite silica [44] D. Kong, R. Peng, P. Ping, et al., A novel battery thermal management system
gel plate coupled with liquid-cooling system for square battery thermal coupling with PCM and optimized controllable liquid cooling for different
management, Appl. Therm. Eng. 184 (2020) 1359e4311. ambient temperatures, Energy Convers. Manag. 204 (2020) 112280.
[29] F. Bai, M. Chen, W. Song, et al., Thermal management performances of PCM/ [45] Y. Huo, Z. Rao, X. Liu, et al., Investigation of power battery thermal manage-
water cooling-plate using for lithium-ion battery module based on non- ment by using mini-channel cold plate, Energy Convers. Manag. 89 (2015)
uniform internal heat source, Appl. Therm. Eng. 126 (2017) 17e27. 387e395.
[30] K. Chen, J.S. Hou, M.X. Song, et al., Design of battery thermal management [46] W.Y. Liu, Y.Q. Jia, Y.Q. Luo, et al., Experimental investigation on thermal
system based on phase change material and heat pipe, Appl. Therm. Eng. 188 management of cylindrical Li-ion battery pack based on vapor chamber
(2021) 116665. combined with fin structure, Appl. Therm. Eng. 162 (2019) 1359e4311.
[31] P. Ping, Y. Zhang, D. Kong, et al., Investigation on battery thermal management [47] K. Chen, J.S. Hou, M.X. Song, et al., Design of battery thermal management
system combining phase changed material and liquid cooling considering system based on phase change material and heat pipe, Appl. Therm. Eng. 188
non-uniform heat generation of battery, Journal of Energy Storage 36 (2021) (2021) 1359e4311.
102448. [48] T. Yang, N. Yang, X. Zhang, et al., Investigation of the thermal performance of
[32] S. Hekmat, G. Molaeimanesh, Hybrid thermal management of a Li-ion battery axial-flow air cooling for the lithium-ion battery pack, Int. J. Therm. Sci. 108
module with phase change material and cooling water pipes: an experimental (2016) 132e144.
investigation, Appl. Therm. Eng. 166 (2020) 114759. [49] F. Chen, R. Huang, C. Wang, et al., Air and PCM cooling for battery thermal
[33] Y. Zheng, Y. Shi, Y.H. Huang, Optimisation with adiabatic interlayers for liquid- management considering battery cycle life, Appl. Therm. Eng. 173 (2020)
dominated cooling system on fast charging battery packs, Appl. Therm. Eng. 115154.
147 (2019) 636e646. [50] N. Javani, I. Dincer, G. Naterer, et al., Heat transfer and thermal management
[34] M. Fleckenstein, O. Bohlen, M.A. Roscher, et al., Current density and state of with PCMs in a Li-ion battery cell for electric vehicles, Int. J. Heat Mass Tran.
charge inhomogeneities in Li-ion battery cells with LiFePO4 as cathode ma- 72 (2014) 690e703.
terial due to temperature gradients[J], J. Power Sources 196 (10) (2011) [51] R. Moffat, Describing the uncertainties in experimental results[J], Exp. Therm.
4769e4778. Fluid Sci. 1 (1) (1988) 3e17.
[35] N. Javani, I. Dincer, G. Naterer, et al., Heat transfer and thermal management [52] Z.W. Wang, T.Z. Lei, X. Chang, et al., Optimization of a biomass briquette fuel
with PCMs in a Li-ion battery cell for electric vehicles[J], Int. J. Heat Mass Tran. system based on grey relational analysis and analytic hierarchy process: a
72 (2014) 690e703. study using cornstalks in China, Appl. Energy 157 (2015) 523e532.
[36] Z. Sun, R. Fan, F. Yan, et al., Thermal management of the lithium-ion battery by [53] K.T. Chiang, F.P. Chang, Optimization of the WEDM process of particle-
the composite PCM-Fin structures[J], Int. J. Heat Mass Tran. 145 (2019) reinforced material with multiple performance characteristics using grey
118739. relational analysis, J. Mater. Process. Technol. 180 (2006) 96e101.
[37] Z. Jiang, Z. Qu, Lithiumeion battery thermal management using heat pipe and [54] A. Kadier, P. Abdeshahian, Y. Simayi, et al., Grey relational analysis for
phase change material during dischargeecharge cycle: a comprehensive nu- comparative assessment of different cathode materials in microbial electrol-
merical study[J], Appl. Energy 242 (2019) 378e392. ysis cells, Energy 90 (2015) 1556e1562.
[38] H. Liu, S. Ahmad, Y. Shi, et al., A parametric study of a hybrid battery thermal [55] S. Balasubramanian, D. Ganapathy, Grey relational analysis to determine op-
management system that couples PCM/copper foam composite with helical timum process parameters for wire electro discharge machining (WEDM), Int.
liquid channel cooling[J], Energy 231 (2021) 120869. J. Eng. Sci. Technol. 3 (1) (2011) 95e101.
[39] W. Wang, X. Zhang, C. Xin, et al., An experimental study on thermal man- [56] H. Zuo, B. Zhang, Z. Huang, K. Wei, J. Tan, Effect analysis on SOC values of the
agement of lithium ion battery packs using an improved passive method, power lithium manganate battery during discharging process and its intelli-
Appl. Therm. Eng. (2018) 163e170. gent estimation, Energy 238 (2022), 121854, [Link]
[40] D. Bernardi, E. Pawlikowski, J. Newman, A general energy balance for battery [Link].2021.121854.
systems, J Electrochemical Scioty 132 (1985) 1. [57] J. E, B. Zhang, Y. Zeng, M. Wen, Z. Huang, K. Wei, J. Chen, H. Zhu, Y. Deng,
[41] Y. Ye, L. saw, Y. Shi, et al., Numerical analyses on optimizing a heat pipe Effects analysis on active equalization control of lithium-ion batteries based
thermal management system for lithium-ion batteries during fast charging[J], on intelligent estimation of the state-of-charge, Energy 238 (2022), 121822,
Appl. Therm. Eng. 86 (2015) 281e291. [Link]
[42] R. Prasher, A simplified conduction based modeling scheme for design [58] L. Hu, X. Hu, Y. Che, F. Feng, X. Lin, Z. Zhang, Reliable state of charge estimation
sensitivity study of thermal solution utilizing heat pipe and vapor chamber of battery packs using fuzzy adaptive federated filtering, Appl. Energy 262
technology[J], J. Electron. Packag. 125 (3) (2003) 378e385. (2020), 114569.

489