Applied Thermal Engineering 237 (2024) 121791

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: [Link]/locate/apthermeng

Research Paper

Numerical investigation of a novel cold plate design with uniform circular

hollow fins for battery thermal management systems
Muhammad Mubashir a, b, Jun Xu a, b, *, Zhechen Guo a, b, Xingzao Wang a, b, Haitao Wang a, b,
Fengpu Qiao c, Enhu Li d, Xuesong Mei a, b
a
State Key Laboratory for Manufacturing Systems Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi 710049 China
b
Shaanxi Key Laboratory of Intelligent Robots, School of Mechanical Engineering, Xi’an Jiaotong University, Xi’an, Shaanxi 710049 China
c
China North Vehicle Research Institute, Beijing 100072 China
d
Gresgying Digital Technology Ltd., Xi’an, Shaanxi 710086 China

A R T I C L E I N F O A B S T R A C T

Keywords: Cooling plate design plays an important role in an efficient battery thermal management system (BTMS). An
Cooling plate effective cooling plate should have a low cost, lightweight, and dissipate heat with minimal pressure drop.
Liquid cooling Therefore, in this study a new lightweight cold plate design with circular hollow fins is proposed and heat
Thermal management
generation is validated by experiment. A 3D numerical model is developed and the performance of the cold plate
Lightweight
Hollow fins
is evaluated by maximum temperature, temperature standard deviation, mass and power consumption. Firstly,
the effects of structural parameters like fins diameter, the distance between hollow fins, cold plate thickness, and
flow rate on the battery maximum temperature, temperature standard deviation, and pressure drop are studied.
The results indicate that increasing the size of the hollow fins reduces the cold plate mass and leads to lower
battery temperature, whereas the pressure drop increases. When the distance between hollow fins decreases
more heat is taken away by the coolant. Furthermore, the performance of the optimized cooling structure is
compared in terms of maximum temperature, temperature difference, mass of cooling plate, and power con­
sumption with conventional liquid cooling structures under the same operating conditions. The results indicate
that the optimized cooling structure can reduce the maximum temperature, temperature difference, mass of the
cooling plate and power consumption by 6%, 29%, 55.2%, and 85.9%, respectively.

change materials (PCM) [18,19]. However, air cooling results in lower

1. Introduction heat transfer coefficient and temperature non-uniformity [20,21].
Although phase change material (PCM) may absorb battery heat, its
Due to environmental pollution and combustion instabilities in the limitations in being used for long periods due to volume changes and
energy conversion systems [1,2], electric vehicles (EVs) are the most packaging issues restrict its application [22]. The coolant’s high thermal
promising option for a clean and green environment as the globe de­ conductivity and heat capacity make liquid cooling systems widely used.
pends more on renewable energy sources [3,4]. Lithium-ion batteries These systems have proven effective, offering the benefits of a compact
(LIBs) are regarded as the most viable power source for EVs due to their structure and convenient layout. Liquid cooling systems are classified
high energy density and long cycle life [5,6]. The restricted battery into the direct contact and the indirect contact cooling systems. In in­
service life, temperature-caused degradation, or even thermal runaway direct contact liquid cooling systems, cold plates are used to cool down
when operating at a high temperature are only a few of the practical the battery temperature. In contrast, in the direct methods, the battery is
application’s ongoing challenges [7–9]. Although it is well known that immersed in cooling fluid such as water or mineral oil to lower the
LIBs perform best when operated at a temperature range from 15 ◦ C to battery temperature. However, the safety of the battery pack is the major
35 ◦ C, and the maximum difference in temperature should be lower than concern in direct cooling method. Short circuit may occur due to
5 ◦ C [10–12]. improper insulation and the coolant should be non-toxic and insulating
EVs use a battery thermal management system (BTMS), which usu­ [23]. Wang et al [24] proposed immersion cooling system using a
ally includes air cooling [13–15], liquid cooling [16,17], and phase transformer oil as the coolant and compared the effectiveness of the

* Corresponding author.
E-mail address: xujunx@[Link] (J. Xu).

[Link]
Received 18 June 2023; Received in revised form 7 October 2023; Accepted 13 October 2023
Available online 14 October 2023
1359-4311/© 2023 Elsevier Ltd. All rights reserved.
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Nomenclature Tsd temperature standard deviation (◦ C)

Uocv open circuit voltage (v)
A battery contact surface area (m2 ) Uc uncertainty
Ac contact surface area of cold plate (m2 ) v velocity of liquid (m .s− 1 )
Ain area of inlet (m2 ) Vb battery volume (m3 )
Q heat generation rate of battery (W) VT total volume of cold plate (m3 )
X distance between hollow fins (mm) Vh volume of cold plate with hollow fins (m3 )
ρ density (kg.m− 3 ) ΔP pressure drop (Pa)
k thermal conductivity (W.m− 1 k− 1 ) µ dynamic viscosity of coolant (pa s)
I current (A)
Subscripts
Cp specific heat capacity (J .kg− 1 k− 1 )
b battery
m mass (g)
c cold plate
pin inlet pressure (Pa)
j joule
pout outlet pressure (Pa)
n number of hollow fins
Pw power consumption (W)
l liquid
q heat generation rate per unit volume (W.m− 3 )
Rj Joule resistance (Ω) Acronyms
T temperature (◦ C) BTMS battery thermal management system
t time (s) LIBs lithium ion battery
Tmax battery maximum temperature (◦ C) PCM phase change materials
ΔTmax maximum temperature difference(◦ C) EV electric vehicles

system with natural air cooling. The results showed that under 0.8 L/ that controlled the battery temperature under normal and extreme
min the Tmax was decreased from 58.3◦ C to 39.4◦ C under 2C discharge. conditions. In normal operation, the cold plate design was able to keep
Guo et al [25] proposed lightweight direct contact liquid cooling system the battery temperature under 25 ◦ C under the flow rate of 0.20 L/min.
of a modular pack with multichannel design. The Tmax , Δ Tmax , and mass Wang et al. [37] designed a novel module design for cylindrical cells and
of the system was considered and the Tmax and Δ Tmax was limited to investigated the effect of serial and parallel cooling. The results showed
36◦ C and 0.65◦ C under 3C discharge rate. In recent years, due to the that under 3C discharge rate Tmax and Δ Tmax is limited to 35.74 ◦ C and
limitations of direct cooling method, many researchers have been 4.17 ◦ C, respectively. The effect of fins plays an important role in heat
attracted to the indirect contact liquid cooling system, which uses in­ transfer with minimal pressure drop [38]. Guo et al. [39] introduced pin
ternal cooling flow channel inside a cold plate which cools down the fin structure in the cooling plate and studied the effects of pressure loss,
battery temperature [26]. The compact structure and easy installation of and flow characteristics. The results indicate that the pin fins structure
liquid cooling system with cold plates makes them an ideal choice for can improve the heat transfer within acceptable pressure loss and the
many applications [27,28]. staggered arrangement of pin fins has lower pressure drop than inline
Implementing cooling plates containing mini channels arranged in arrangement. Saghir et al. [40] studied the effect of water and nano
parallel can significantly decrease the pressure drop within the channels particles for cooling of the battery with pin fins structure. Results
[29]. Huo et al. [30] investigated a BTMS using mini channels. The ef­ showed that by using nanofluids heat transfer improves but resulting in
fects of flow direction, channel number, and ambient temperature on higher pressure drops. Pin fins improves the heat transfer up to 29 % and
temperature distribution were investigated. The findings showed that by water cooling is more suitable in the presence of pin fins. Xu et al [41]
increasing the flow rate and channel number, the batteries Tmax proposed a wrench shaped cold plate design for the cylindrical battery
decreased. Xu et al. [31] investigated the impact of climatic and oper­ modules and compare the performance with straight mini channel cold
ating conditions of a T-shape bifurcation cold plate on the side of the plate. The performance of the cold plate design is evaluated under
large capacity battery module. From the results, it was concluded that temperature uniformity, maximum temperature and power consump­
multiple cold plates could be used for better temperature uniformity tion. The results show that by using bifurcation design the weight and
proper and coolant distribution. Wu et al [32] proposed a mini channel pressure drop of the cold plate can be reduced.
cooling system with step allocated scheme for large battery packs. Re­ Despite major advancements within mini-channel cold plate struc­
sults showed that by using step allocated scheme the temperature of the ture in the previous studies, the need for a cold plate with higher thermal
battery can be controlled under same mass flow rate and higher efficiency, less power consumption, and lighter weight remains chal­
discharge rates for large battery packs. Jin et al [33] proposed oblique lenging. Moreover, conventional cooling plate designs use pin–fin
fin structure which helps to develop the boundary layer and can structures [33,39]. These pin-fins structures are solid structures that can
decrease the pressure drop and maintains the battery surface tempera­ improve heat transfer but occupy more space, and result in increasing
ture under 50◦ C with a flow rate under 0.9 L/min. Deng et al. [34] the weight of the cooling plate. Therefore, cooling plate’s weight is also
extensively investigated the impact of the coolant’s mass flow rate, an important factor to consider. However, there are limited studies that
coolant direction, and several serpentine channels on the aluminum simultaneously take into account power consumption, lightweight
plate’s cooling performance. His results showed that under the mass design, and thermal performance.
flow rate of 1 g/s the temperature of battery can be controlled under To address the problem of cooling plate design considering light
26.67◦ C and 32.17◦ C at 3C and 5C discharge. Tang et al. [35] studied weight structure as mentioned in the above literature, this paper pro­
and improved the three different water cooling techniques for mini- poses a novel cold plate design with uniform circular hollow fins that
channel cold plate battery pack structure. The findings demonstrated can reduce the battery temperature, temperature standard deviation,
that the placement of a cold plate on the adjacent and bottom of the mass of the cooling plate and power consumption. The hollow fin
battery module proves the best cooling. Mohammed et al. [36] proposed configuration is designed to evenly distribute the coolant within the
BTMS with a dual function cooling plate with staggered fin arrangement plate. These circular hollow fins improve the cold plate’s effective heat

2
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Table 1 plate is 3 mm, whereas the height and width of the inlet are 2 mm and 4
Battery parameters and geometry specifications. mm, respectively, as shown in Fig. 1(c). Table 2 shows the aspect ratio
Parameter Values parameters and the thickness of the cold plate. The thickness of cold
plate is varied under different inlet areas and the effect of temperature
Size (Length*Width*Height) 120*67*12 (mm)
Rated voltage 3.7 V distribution is studied. The number of circular hollow fins are kept
Rated capacity 10Ah constant in each design of cold plates while increasing the size of the
Density 2543.9 kg m− 3 diameter.
Specific heat capacity 1679 J kg− 1 K− 1
Thermal conductivity 3.3 W m− 1 K− 1
Mass 0.23 Kg 2.2. Computation model and governing equations

ANSYS (ICEM) is used for the numerical solution and the mesh of the
transfer area by allowing fluid to flow over them with minimal pressure models. Tetrahedron and structured mesh is employed. The final grid
drop, resulting in effective cooling and reduced temperature gradients in structure is shown in Fig. 2 (a). To verify the accuracy of numerical
both the battery and cooling plate. The performance of the cold plate simulations grid independence and time independence test are per­
design is evaluated by four indicators, namely the maximum tempera­ formed which has a great influence on the simulation results [42,43].
ture (Tmax ), temperature standard deviation on the battery contact sur­ The result of the grid test is depicted in Fig. 2 (b), when the grid number
face (Tsd ), power consumption (Pw ), and mass of the cooling plate (m). is increased from 3.8 x 106 to 7.2 x 106 there is no change in the tem­
Furthermore, the effect of fin diameter, mass flow rate, cooling direction perature. Therefore, to save the computational time and ensure the ac­
and thickness of the cold plate is investigated and the optimized cold curacy of simulation results, the grid size 3.8 x 106 was selected. Besides,
plate structure is compared with conventional liquid cooling structures. time step of 1 s, 5 s, 30 s and 60 s are performed and the results are
This design of cold plate is light weight than traditional cooling plate shown in Fig. 2 (c). From the results is it seen that the temperature
designs due to the presence of hollow fins structure, as less material is curves overlaps when the time step is between 1 s and 30 s. When the
used. time step is further increased to 60 s the temperature variation is
The remainder of this paper is as follows: Firstly, the cold plate with different. Therefore, to save the computational time and accuracy, time
circular hollow fins for BTMS is designed and analyzed. Then the step 30 s is selected. All the surfaces are set up for the coupling process.
computational model is built and validated. Further, effects of varying The cold plate is assumed to be homogenous and isotropic. Aluminum is
diameters are analyzed. Ultimately, the main conclusions are drawn at used for the cold plate, and water serves as the coolant. The properties of
the end of this paper. aluminum and water are shown in Table 3. In the simulation, a transient
case is selected, and the battery is discharged at a volumetric heat
2. Model establishment and verification generation rate of 3C.
The discharging current is calculated by multiplying the nominal
2.1. Cold plate design with circular hollow fins for BTMS capacity of the battery by the C-rate value, where the C-rate is the
discharge current concerning the battery’s nominal capacity. The heat is
The battery used in our study is a lithium ion pouch battery with a transferred from the battery to the cooling plates, and then heat is taken
capacity of 10Ah. The battery dimensions and parameters are shown in away by the coolant. The side surface of the battery is set as free con­
Table 1. The size of the cold plate is the same as the size of the battery, vection with a heat transfer coefficient of 5 W m− 2 K− 1 [44]. The focus is
two cold plates are attached on the both sides of the battery. The
schematic of the BTMS model is shown in Fig. 1 (a). Fig. 1(b) shows the
Table 2
section view of cold plate. Whereas Fig. 1(c) shows the dimensions of the
Aspect ratio inlet and cold plate thickness.
cold plate. The distance between circular hollow fins along the x-axis is
Variable Case A Case B Case C
labeled as X, whereas along the y-axis are labeled as Y.
In order to make the BTMS system lightweight, three types of cold Aspect ratio (A/B) 2:4 1:4 0.5:4
plates are designed with non-uniform hollow size diameters ranging Thickness (C) (mm) 3 2 1
Velocity 0.2 m/s
from 4 mm, 6 mm, and 8 mm, respectively. The thickness of the cold

Fig. 1. Schematic of the BTMS System: (a) liquid cooling system (b) section view of cold plate (c) geometry of cold plate.

3
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 2. Simulation: (a) Mesh of cold plate. (b) Grid independence test. (c) Time step independence.

Therefore, the laminar model is used for all the numerical simulations.
Table 3
The density and specific heat capacity of a battery is assumed to be
Thermophysical properties of the material.
homogenous, and the thermal conductivity is isotropic. The heat gen­
Material Density Specific heat Thermal conductivity Viscosity eration in the battery is assumed to be uniform heat generation. An
ρ (kg m− 3 ) Cp (J kg− 1 k− 1 ) λ(W m− 1 k− 1 ) μ(Pa s)
energy conservation equation (1) is used to compute the temperature of
Aluminum 2719 871 202.4 – a single battery [46].
Water 998.2 4128 0.6 0.001003
∂( )
ρ c T = ∇.(kb ∇T) + Q (1)
∂t b pb
solely on considering the heat generated by the cell body. For the
coolant flow, the velocity inlet and outflow are taken as the boundary where Q, ρb , kb , cpb , and T represents the heat generation rate of the
conditions. Initially, the inlet temperature for the coolant is set at 298 K. battery, density, thermal conductivity, specific heat capacity and tem­
Reynold number is a dimensionless number used to determine the fluid perature, respectively.
flow state. If the Reynold number is less than 2300 laminar flow is used, The battery temperature is calculated by Q in (1), the battery heat
and if it’s greater than 4000 the flow is turbulent and turbulence model generation Q consists of electrochemical heat and joule heat. [47]
is used [45]. So, the maximum flow rate in the channel is 0.06 L/min, 1
(
∂Uocv
)
and the Reynolds number is less than the critical value of 2300. q= I 2 Rj + IT (2)
Vb ∂T

Fig. 3. Experiments: (a) Experimental setup. (b) Thermocouple locations.

4
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 4. Comparison between experiment and simulation.

where q is the heat generation rate per unit volume, Vb is the volume of
the battery, I2 Rj is the joule heat and IT ∂U∂Tocv is the chemical reaction
heat. Rj is the heat generated by Joule resistance.
The mass conservation, momentum conservation, and energy con­
servation of coolant is expressed as:
Fig. 7. Effect of Increasing Diameter on Tmax , mass and pressure drop.

Fig. 5. Temperature contours of battery a surface under varying diameters.

Fig. 6. Curve under different diameters: (a) Tmax and (b) Tsd .

5
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 8. Temperature contours of battery surface under varying flow rates of 4 mm diameter cold plate.

Fig. 9. Curve under different flow rates: (a) Tmax and (b) Tsd .

∂ρl
+ ∇.(ρl v) = 0 (3) Vh = πr2 HF .n (8)
∂t

∂ρl VCP = L*W*H (9)

v + ∇.(ρl v)v = ∇p (4)
∂t
V = VCP − VH (10)
∂T ( )
ρl cpl + ∇. ρl cpl vT = ∇.(kl ∇T) (5) where Vh ,VCP ,V, HF n, represent the volume of the plate with hollow fins,
∂t
the volume of the cold plate without hollow fins, the total volume of the
∂T cold plate, the height of fins, and the number of circular hollow fins,
ρc cpc = ∇.(kc ∇T) (6)
∂t respectively.
The pumping power required for the coolant flow is expressed as the
The energy conservation of the cold plate is expressed as:
power consumption and it is calculated by:
where ρl , cpl , kl and v represents the density, specific heat, thermal
conductivity and velocity of coolant, ρc , cpc and kc denotes the density, Pw = ΔP⋅Ain ⋅vin (11)
heat capacity and thermal conductivity of the cold plate, respectively.
The mass of the cooling plate is calculated by: where Ain is the channel inlet area and vin is the inlet velocity.
The pressure drop is calculated by:
mc = ρAL .VT (7)
ΔP = pout − pin (12)
where mc ,ρAL , VT represents mass, density, and total volume of cold
plate. where pout is the outlet pressure and pin is the inlet pressure.
The volume of the circular hollow fin is calculated by: The Tsd is an important parameters that shows the temperature

6
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 10. Temperature contours of battery surface under varying distance between hollow fins.

Fig. 11. Curve under varying distance: (a) Tmax and (b) Tsd .

uniformity on the battery surface which can be expressed as follows: middle of the battery and point 4 is at the bottom. The location of
∫ thermocouples are shown in Fig. 3(b). The battery is charged at a 1C
TdA
Tavg = ∫Ac (13) charge rate and discharged at a 3C rate by the constant current-constant-
Ac
dA current voltage (CC-CV) method. The charge and discharge cut-off
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ voltage is (4.2 V-2.75 V). The battery has a maximum temperature at
√∫ ( )2 point 1 near the positive electrode. From the comparison of experi­
√ T − Tavg dA
T σ = √ Ac ∫ (14) mental and simulated temperatures, the difference at the end of 3C
dA
Ac
discharge is less than 0.42◦ C. This shows that difference between the
experimental results and the simulated result is minor and acceptable.
where Ac is the contact surface area of the battery and cold plate and Tavg
The relative error between the experimental and simulation results does
is the average contact surface temperature.
not exceed 5.5 %, which can fully demonstrate the accuracy and reli­
ability of the numerical model, as shown in Fig. 4.
3. Validation of computational model
3.1. Uncertainty analysis
Validation plays an important role to check the accuracy of numer­
ical method [48]. Hence, in order to verify the numerical solution, the
Arbin Cycler is used for battery charge and discharge. The voltage
experimental setup is built as shown in Fig. 3(a). To maintain a constant
and current accuracy of the battery cycler is ± 0.01 % of full scale.
ambient temperature of 25 ◦ C, the battery is placed in a thermostat
Whereas, the temperature discrepancy is within ± 0.1◦ C. The range of T-
chamber. The battery charge/discharge cycler is used at constant cur­
type thermocouple is − 200 ~ 400◦ C. The uncertainty of the entire
rent mode with high accuracy and transfers the data to the temperature
measure system is calculated using Eq. (15)-(17) [49]. The overall un­
acquisition unit. The temperature of the battery is measured by T-type
certainty of the temperature measurement system is 1.2 %.
thermocouples which are attached on the battery surface. Point 1 and 2
are near the positive electrode and negative electrode, point 3 is at the

7
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 12. Effect of aspect ratio and cold plate thickness.

Fig. 13. Curve under different aspect ratios: (a) Tmax and (b) Tsd .

√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
√∑ investigated. Firstly the cooling performance of the cold plate is
√n 2
√ (xi − x)
√i=1 analyzed by varying the hole size diameters under 0.06 L/min flow rate.
Ua = t (15) The effect of diameters on the temperature difference of the battery is
(n − 1)
studied, and further optimization is done on the cooling plate design.
where t is constant factor and n is number of measurements. The effects of mass flow rate, inlet aspect ratio, thickness of cold plate,
distance along the x-axis between circular follow fins, and flow channel
Ub =
Δe
(16) directions, are studied.
C

where Δe the instrument error and C is constant factor. 4.1. Effect of varying diameters
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
Uc = U 2a + U 2b (17) To study the circular hollow fins effect under different diameters, the
flow rate 0.06 L/min is selected, and the simulation is done under 3C.
4. Results and discussion The distance between the circular hollow fins is kept constant at X-10
mm along the X-axis. Fig. 5 shows the temperature distribution on the
The cold plate design with varying hole size diameters are battery surface under a 3C discharge rate. The battery Tmax reaches
31.34 ◦ C when the diameter of the circular fin is 4 mm, whereas, when

8
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 14. Cold plate design with different cooling directions.

the diameter of the circular fins is increased to 6 mm the Tmax reaches from 8 mm to 10 mm, the Tmax reaches 30.13◦ C and 31.34 ◦ C, respec­
31.29 ◦ C. Similarly the battery Tmax under 8 mm diameters reaches tively. Fig. 11(a),(b) shows the Tmax and Tsd curves. The battery has a
31.08 ◦ C. Tmax and the Tsd curves are shown in Fig. 6(a) and 6(b). It is lower temperature when the distance is kept 6 mm, this is because when
apparent that the battery’s temperature decreases as the diameter of the the hollow fins are closer to each other more heat is conducted and taken
circular hollow fins increases. By increasing the diameter, the cold away by the coolant. This suggests that a distance of 6 mm between the
plate’s mass also decreases. However, at the same period, the pressure circular hollow fins along the X-direction is the most effective in
drop increases, which results in increasing power consumption, as maintaining a lower Tmax . Whereas, the temperature distribution on the
shown in Fig. 7. As a result, the plate with 4 mm circular hollow fins is battery surface is better when the distance of 10 mm is used.
chosen, and additional parametric analysis is carried out on it. This is
because the battery temperature under a 4 mm plate is within the safe 4.4. Effect of aspect ratio and thickness of the cold plate
temperature limit, and it has a minimum pressure drop and Tsd than a
cold plate with 6 mm and 8 mm circular hollow fins. In order to study the aspect ratio of the cold plate inlet and the
thickness of the cold plate, as shown in Table 2, is studied in this section.
4.2. Effect of mass flow rate Fig. 12 shows the contour of the temperature distribution on the battery
surface. The velocity of the coolant is kept constant for all three cases.
This section examines the impact of mass flow rate on the cold plate Case A has a cold plate thickness of 3 mm and the Tmax reaches 30.10 ◦ C.
design using a cold plate with a 4 mm circular hollow fin. With higher The temperature distribution in Case A is also much better than the Case
flow rates, a more uniform temperature distribution can be achieved. B and Case C. When the thickness of the cold plate and the aspect ratio of
The flow rate ranges are 0.02 L/min, 0.04 L/min, and 0.06 L/min, and the inlet is decreased the Tmax reaches 31.36 ◦ C and 34.27 ◦ C for Case B
the coolant temperature is fixed at 298 K. Fig. 8 shows the temperature and C, respectively. This is because when the thickness of the cold plate
contours of the battery surface under different flow rates. When the cold is increased, more heat is carried and taken away by the cold plate.
plate with 4 mm diameter is used, and the flow rate is 0.02 L/min the Fig. 13(a),(b) shows the Tmax and Tsd curves under different aspect ratios
Tmax , reaches 34.7 ◦ C. As the flow rate increases, the maximum tem­ and cold plate thickness. The Tsd of Case A is the minimum than Case B
perature decreases to 31.87 ◦ C and 30.89 ◦ C at flow rates of 0.04 L/min and Case C,which means that the temperature distribution in case A is
and 0.06 L/min, respectively. The Tmax reduces by 10.9 % when the flow better. It shows that when the aspect ratio and cold plate thickness are
rate of 0.06 L/min is used. The Tmax and Tsd curves are shown in Fig. 9 increased, the Tmax can be reduced, but it is limited to some extent.
(a),(b). The Tsd and Tmax under 0.06 L/min is minimum, hence 0.06 L/ Further increase in the cold plate thickness may cause an increase in the
min is used in further calculations. weight of the BTMS. So Case A with a thickness of 3 mm is selected to be
a better choice.
4.3. Effect of distance between hollow fins
4.5. Effect of cooling directions
In order to study the effect of distance between circular hollow fins
on the temperature distribution of the battery along the X-direction, is The flow direction has an important impact on the overall perfor­
studied in this section. The cold plate with a 4 mm diameter is selected mance of the thermal management system. Four cases of different flow
under the flow rate of 0.06 L/min. The distance between hollow fins is directions, as shown in Fig. 14, were considered and simulated. Model
varied from 6 mm to 10 mm, and the temperature distribution on the (A) has one inlet at the top and one outlet at the bottom. Model (B) has 1
battery surface is shown in Fig. 10. When the distance is kept at 6 mm inlet at the top and two outlets at the bottom. Model (C) has 2 inlets at
the Tmax reaches 30.89 ◦ C. Whereas when the distance further increased the top and 2 outlets at the bottom. In contrast, Model (D) has two bi-

9
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 15. Temperature contours of battery surface, Temperature distribution and velocity stream lines in channel.

direction inlets and outlets at the top and bottom, respectively. The plate distribution in the channel and velocity stream lines under different
with a 4 mm circular hollow fin is selected under a flow rate of 0.06 L/ cooling directions. When model A is used the Tmax reaches 30.85◦ C.
min, and the inlet temperature is kept at 25 ◦ C. When model B is used, which has one inlet at the center of the battery
Fig. 15 shows the temperature contours of the battery, temperature and 2 outlets at the bottom of the battery, the battery Tmax reaches

10
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

Fig. 16. (a) Tmax and (b) Tsd comparison curves with different models.

Fig. 17. Temperature distribution contours in a battery module: (a) Battery Module (b).

31.58◦ C. When model C is used, which has 2 inlets at the top and 2 battery temperature distribution in a module. Battery number 3 has the
outlets at the bottom of the battery, the battery Tmax reaches 30.08 ◦ C. highest temperature, and battery number 1 has the lowest temperature.
Model C has a better temperature distribution as compared to Model D, This is because battery 3 is placed in the middle of the module, and heat
where the battery Tmax decreases by 29.87 ◦ C. Fig. 16(a),(b) shows the dissipation is more in the module’s center. Fig. 18 shows the battery Tmax
battery Tmax and Tsd curves. Model D, which has bidirectional inlet and and ΔTmax curves of the module. Battery 3 reaches the temperature of
outlet flow channels, has the minimum Tmax and Tsd among the other 29.61◦ C, and the temperature difference of the module is limited to 4 ◦ C.
cooling direction models. Furthermore, a battery module is built
considering 6 battery cells and 7 cold plates to study the effect of tem­ 4.6. Comparison with different cooling structures
perature distribution in terms of Tmax and Tsd of Model D is studied. The
schematic of battery module is shown in Fig. 17(a). Fig. 17(b) shows the In order to evaluate and compare the effectiveness of the proposed

11
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

make a meaningful comparison, the models are reconstructed as shown

in Fig. 19(a),(b), and (c), and the size of the battery and cold plate is the
same as discussed in this work. Then the designed models are simulated
under the same simulation parameters and compared with the optimized
model D, as described in section 4.5.
Fig. 20. Shows the comparison results of the optimized model with
different cooling structures. Fig. 20 (a) show the Tmax and ΔTmax curves,
whereas Fig. 20 (b) shows the power consumption and mass of the
cooling plate. It can be seen that Tmax and ΔTmax is limited to 29.87 ◦ C
and 3.52 ◦ C respectively. It means that the optimal model can lower the
battery temperature and the temperature distribution more uniformly.
The Tmax , ΔTmax , mass and power consumption is reduced when
compared with these structures except the power consumption in [50],
this structure has the minimum power consumption only. The optimal
model can reduce the Tmax and ΔTmax by 6 % and 29 %, whereas the
overall mass and power consumption are reduced 55.2 % and 85.9 %
when compared with other cooling structures.

5. Conclusion
Fig. 18. Tmax and Δ Tmax curves of battery module.
In this paper a novel cold plate design with uniform circular hollow
model. A comparison is made by considering Tmax , Δ Tmax , power con­ fins is proposed to improve the Tmax , Δ Tmax , power consumption and
sumption and mass of cooling plate. It’s important to note that battery mass of the cold plate. The performance analysis of the cold plate design
type, size, dimension, cooling structures and operating conditions in the is done through numerical simulation. Studies were done on the impact
existing literature [50–52] are different from each other. Therefore to of changing the diameter of the circular hollow fins, distance between

Fig. 19. Redesign model of different cold plate structures: (a) [50], (b) [51], (c) [52].

Fig. 20. Model comparison with different cooling structures: (a) Tmax and Δ Tmax , (b) Power consumption and mass.

12
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

hollow fins, thickness of cold plate, and cooling directions. The results [5] X. Li, F. He, G. Zhang, Q. Huang, D. Zhou, Experiment and simulation for pouch
battery with silica cooling plates and copper mesh based air cooling thermal
shows that fins size has less impact on the temperature distribution, but
management system, Appl. Therm. Eng. 146 (2019) 866–880.
it have significant impact on the pressure drop and mass of cooling plate. [6] J. Xu, X. Mei, X. Wang, Y. Fu, Y. Zhao, J. Wang, A Relative State of Health
Furthermore, the optimized cooling structure is compared with other Estimation Method Based on Wavelet Analysis for Lithium-Ion Battery Cells, IEEE
liquid cooling [Link] key findings are as follows. Trans. Ind. Electron. 68 (2021) 6973–6981.
[7] Z. Rao, S. Wang, M. Wu, Z. Lin, F. Li, Experimental investigation on thermal
management of electric vehicle battery with heat pipe, Energy Convers. Manag. 65
i) The effect of diameter on the maximum temperature is very less. (2013) 92–97.
However, when the diameter of the hollow fin increases, the mass [8] J. Zhao, Z. Rao, Y. Huo, X. Liu, Y. Li, Thermal management of cylindrical power
battery module for extending the life of new energy electric vehicles, Appl. Therm.
of the cold plate decreases, and pressure drop increases, which Eng. 85 (2015) 33–43.
results in more power consumption. [9] J. Xu, X. Mei, H. Wang, J. Wang, A Hybrid Self-Heating Method for Batteries Used
ii) The temperature uniformity and the cooling effect of the battery at Low Temperature, IEEE Trans. Ind. Informatics. 17 (2021) 4714–4723.
[10] A.A. Pesaran, Battery thermal models for hybrid vehicle simulations, J. Power
improve with an increase in flow rate. When the flow rate of 0.06 Sources 110 (2002) 377–382.
L/min was used, the battery maximum temperature reached [11] Y. Zhang, X. Song, C. Ma, D. Hao, Y. Chen, Effects of the structure arrangement and
30.89 ◦ C. spacing on the thermal characteristics of Li-ion battery pack at various discharge
rates, Appl. Therm. Eng. 165 (2020), 114610.
iii) The effect of increasing the distance between circular hollow fins [12] T. Hempijid, C. Kittichaikarn, Effect of heat sink inlet and outlet flow direction on
may cause an increase in temperature, but the temperature dis­ heat transfer performance, Appl. Therm. Eng. 164 (2020), 114375.
tribution remains uniform. When the distance between circular [13] K. Chen, Y. Chen, Z. Li, F. Yuan, S. Wang, Design of the cell spacings of battery pack
in parallel air-cooled battery thermal management system, Int. J. Heat Mass Transf.
hollow fins reduces the temperature of the battery reduces.
127 (2018) 393–401.
iv) The effect of cooling direction has an important effect on the [14] Y. Xie, X. Wang, W. Li, Y. Zhang, D. Dan, K. Li, F. Feng, C. Wu, P. Wang,
battery temperature. Four designs were simulated, and case 4 A resistance-based electro-thermal coupled model for an air-cooled battery pack
showed the best performance. The Tmax reached 29.87 ◦ C and the that considers branch current variation, Int. J. Therm. Sci. 159 (2021), 106611.
[15] J. Xie, Z. Ge, M. Zang, S. Wang, Structural optimization of lithium-ion battery pack
Δ Tmax was 3.52 ◦ C. with forced air cooling system, Appl. Therm. Eng. 126 (2017) 583–593.
v) When compared with other liquid cooling structures, the opti­ [16] S. Lin, L. Zhou, Thermal performance of rectangular serpentine mini-channel
mized structure can reduce the Tmax , Δ Tmax , mass of cooling plate cooling system on lithium battery, J. Clean. Prod. 418 (2023), 138125.
[17] J. Xu, Z. Guo, Z. Xu, X. Zhou, X. Mei, A systematic review and comparison of liquid-
and power consumption by 6 %, 29 % and 55.2 % and 85.9 % based cooling system for lithium-ion batteries, Etransportation. 17 (2023), 100242.
respectively. [18] X. Liu, L. Yao, C. Su, X. Xiong, Y.-P. Wang, A hybrid battery thermal management
system coupling with PCM and optimized thermoelectric cooling for high-rate
discharge condition, Case Stud. Therm. Eng. 49 (2023), 103269.
CRediT authorship contribution statement [19] B. Xu, F. Xia, Y. Wang, X. Xie, W. Gan, A battery thermal management scheme
suited for cold regions based on PCM and aerogel: Demonstration of performance
Muhammad Mubashir: Methodology, Writing – original draft, Data and availability, Appl. Therm. Eng. 227 (2023), 120378.
[20] R.C. Sahin, S. Gocmen, E. Cetkin, Thermal management system for air-cooled
curation, Validation. Jun Xu: Conceptualization, Methodology, Inves­ battery packs with flow-disturbing structures, J. Power Sources 551 (2022),
tigation, Writing – review & editing, Supervision, Funding acquisition. 232214.
Zhechen Guo: Methodology, Formal analysis, Data curation, Valida­ [21] K. Chen, Z. Zhang, B. Wu, M. Song, X. Wu, An air-cooled system with a control
strategy for efficient battery thermal management, Appl. Therm. Eng. 121578
tion. Xingzao Wang: Formal analysis, Investigation, Data curation. (2023).
Haitao Wang: Resources, Data curation, Validation. Fengpu Qiao: Data [22] G. Murali, G.S.N. Sravya, J. Jaya, V. Naga Vamsi, A review on hybrid thermal
curation, Investigation, Validation. Enhu Li: Investigation, Validation. management of battery packs and it’s cooling performance by enhanced PCM,
Renew, Sustain. Energy Rev. 150 (2021), 111513.
Xuesong Mei: Supervision, Funding acquisition. [23] P. He, H. Lu, Y. Fan, H. Ruan, C. Wang, Y. Zhu, Numerical investigation on a
lithium-ion battery thermal management system utilizing a double-layered I-
shaped channel liquid cooling plate exchanger, Int. J. Therm. Sci. 187 (2023),
Declaration of Competing Interest 108200.
[24] H. Wang, T. Tao, J. Xu, H. Shi, X. Mei, P. Gou, Thermal performance of a liquid-
immersed battery thermal management system for lithium-ion pouch batteries,
The authors declare that they have no known competing financial J. Energy Storage. 46 (2022), 103835.
interests or personal relationships that could have appeared to influence [25] Z. Guo, J. Xu, Z. Xu, M. Mubashir, H. Wang, X. Mei, A Lightweight Multichannel
the work reported in this paper. Direct Contact Liquid-Cooling System and Its Optimization for Lithium-Ion
Batteries, IEEE Trans. Transp. Electrif. 8 (2022) 2334–2345.
[26] Y. Huang, P. Mei, Y. Lu, R. Huang, X. Yu, Z. Chen, A.P. Roskilly, A novel approach
Data availability for Lithium-ion battery thermal management with streamline shape mini channel
cooling plates, Appl. Therm. Eng. 157 (2019), 113623.
Data will be made available on request. [27] M.M. Hamed, A. El-Tayeb, I. Moukhtar, A.Z. El Dein, E.H. Abdelhameed, A review
on recent key technologies of lithium-ion battery thermal management: External
cooling systems, Results Eng. 16 (2022), 100703.
Acknowledgements [28] G. Zhao, X. Wang, M. Negnevitsky, C. Li, An up-to-date review on the design
improvement and optimization of the liquid-cooling battery thermal management
system for electric vehicles, Appl. Therm. Eng. 219 (2023), 119626.
This work was supported by the National Natural Science Foundation [29] C. Wang, G. Zhang, X. Li, J. Huang, Z. Wang, Y. Lv, L. Meng, W. Situ, M. Rao,
of China (52075420), the National Key Research and Development Experimental examination of large capacity liFePO4 battery pack at high
temperature and rapid discharge using novel liquid cooling strategy, Int. J. Energy
Program of China (2020YFB1708400).
Res. 42 (2018) 1172–1182.
[30] Y. Huo, Z. Rao, X. Liu, J. Zhao, Investigation of power battery thermal management
References by using mini-channel cold plate, Energy Convers. Manag. 89 (2015) 387–395.
[31] X. Xu, W. Li, B. Xu, J. Qin, Numerical study on a water cooling system for prismatic
LiFePO4 batteries at abused operating conditions, Appl. Energy 250 (2019)
[1] W. Li, D. Zhao, L. Zhang, X. Chen, Proper orthogonal and dynamic mode
404–412.
decomposition analyses of nonlinear combustion instabilities in a solid-fuel ramjet
[32] N. Wu, X. Li, N. Ouyang, W. Zhang, Mini-channel liquid cooling system for large-
combustor, Therm. Sci. Eng. Prog. 27 (2022), 101147.
sized lithium-ion battery packs by integrating step-allocated coolant scheme, Appl.
[2] Y. Xi, X. Li, Y. Wang, B. Xu, N. Wang, D. Zhao, Experimental study of transition to
Therm. Eng. 214 (2022), 118798.
instability in a Rijke tube with axially distributed heat source, Int. J. Heat Mass
[33] L.W. Jin, P.S. Lee, X.X. Kong, Y. Fan, S.K. Chou, Ultra-thin minichannel LCP for EV
Transf. 183 (2022), 122157.
battery thermal management, Appl. Energy 113 (2014) 1786–1794.
[3] A. Greco, X. Jiang, D. Cao, An investigation of lithium-ion battery thermal
[34] T. Deng, G. Zhang, Y. Ran, P. Liu, Thermal performance of lithium ion battery pack
management using paraffin/porous-graphite-matrix composite, J. Power Sources
by using cold plate, Appl. Therm. Eng. 160 (2019), 114088.
278 (2015) 50–68.
[35] A. Tang, J. Li, L. Lou, C. Shan, X. Yuan, Optimization design and numerical study
[4] C. Lin, J. Xu, J. Hou, Y. Liang, X. Mei, Ensemble Method with Heterogeneous
on water cooling structure for power lithium battery pack, Appl. Therm. Eng. 159
Models for Battery State-of-Health Estimation, IEEE Trans. Ind. Informatics. (2023)
(2019), 113760.
1–9.

13
M. Mubashir et al. Applied Thermal Engineering 237 (2024) 121791

[36] A.H. Mohammed, R. Esmaeeli, H. Aliniagerdroudbari, M. Alhadri, S.R. Hashemi, [44] X.M. Xu, R. He, Research on the heat dissipation performance of battery pack based
G. Nadkarni, S. Farhad, Dual-purpose cooling plate for thermal management of on forced air cooling, J. Power Sources 240 (2013) 33–41.
prismatic lithium-ion batteries during normal operation and thermal runaway, [45] P. Linyuan, D. Jian, Effects of staggered opposed hot jets on the initiation and
Appl. Therm. Eng. 160 (2019), 114106. propagation of gaseous detonation in a supersonic combustible inflow, Phys. Fluids
[37] H. Wang, T. Tao, J. Xu, X. Mei, X. Liu, P. Gou, Cooling capacity of a novel modular 35 (2023) 14102.
liquid-cooled battery thermal management system for cylindrical lithium ion [46] P. Amiribavandpour, W. Shen, D. Mu, A. Kapoor, An improved theoretical
batteries, Appl. Therm. Eng. 178 (2020), 115591. electrochemical-thermal modelling of lithium-ion battery packs in electric vehicles,
[38] C. Ji, D. Zhao, Lattice Boltzmann investigation of acoustic damping mechanism and J. Power Sources 284 (2015) 328–338.
performance of an in-duct circular orifice, J. Acoust. Soc. Am. 135 (2014) [47] D. Bernardi, E. Pawlikowski, J. Newman, A General Energy Balance for Battery
3243–3251. Systems, J. Electrochem. Soc. 132 (1985) 5.
[39] Z. Guo, Q. Xu, S. Zhao, S. Zhai, T. Zhao, M. Ni, A new battery thermal management [48] D. Zhao, E. Gutmark, A. Reinecke, Mitigating self-excited flame pulsating and
system employing the mini-channel cold plate with pin fins, Sustain. Energy thermoacoustic oscillations using perforated liners, Sci. Bull. 64 (2019) 941–952.
Technol. Assessments. 51 (2022), 101993. [49] Y. Ding, M. Wei, R. Liu, Channel parameters for the temperature distribution of a
[40] M.Z. Saghir, M.M. Rahman, Y. Bicer, Investigation of channel materials toward battery thermal management system with liquid cooling, Appl. Therm. Eng. 186
better cooling lithium-ion batteries in the presence of nanofluid and pin-fins, Int. J. (2021), 116494.
Thermofluids. 18 (2023), 100349. [50] D. Zhao, Z. Lei, C. An, Research on battery thermal management system based on
[41] Q. Xu, Y. Xie, Y. Huang, X. Li, H. Huang, S. Bei, H. Wang, K. Zheng, X. Wang, L. Li, liquid cooling plate with honeycomb-like flow channel, Appl. Therm. Eng. 218
Enhancement of thermal management for cylindrical battery module based on a (2023), 119324.
novel wrench-shaped design for the cold plate, Sustain. Energy Technol. [51] T. Deng, G. Zhang, Y. Ran, Study on thermal management of rectangular Li-ion
Assessments. 59 (2023), 103421. battery with serpentine-channel cold plate, Int. J. Heat Mass Transf. 125 (2018)
[42] D. Zhao, Y. Guan, A. Reinecke, Characterizing hydrogen-fuelled pulsating 143–152.
combustion on thermodynamic properties of a combustor, Commun. Phys. 2 [52] A. Jarrett, I.Y. Kim, Design optimization of electric vehicle battery cooling plates
(2019) 44. for thermal performance, J. Power Sources 196 (2011) 10359–10368.
[43] Y. Sun, D. Zhao, X. Zhu, Generation and Mitigation Mechanism Studies of
Nonlinear Thermoacoustic Instability in a Modelled Swirling Combustor with a
Heat Exchanger, Aerospace 8 (2021).

14