energies

Article
Low-Cost Air-Cooling System Optimization on Battery Pack of
Electric Vehicle
Robby Dwianto Widyantara 1 , Muhammad Adnan Naufal 1 , Poetro Lebdo Sambegoro 1 ,
Ignatius Pulung Nurprasetio 1 , Farid Triawan 2 , Djati Wibowo Djamari 2 , Asep Bayu Dani Nandiyanto 3 ,
Bentang Arief Budiman 1, * and Muhammad Aziz 4, *

1 Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha No. 10,
Bandung 40132, Indonesia; [email protected] (R.D.W.); [email protected] (M.A.N.);
[email protected] (P.L.S.); [email protected] (I.P.N.)
2 Department of Mechanical Engineering, Faculty of Engineering and Technology, Sampoerna University, Jl.
Raya Pasar Minggu No. 16, Jakarta 12780, Indonesia; [email protected] (F.T.);
[email protected] (D.W.D.)
3 Departemen Kimia, Universitas Pendidikan Indonesia, Jl. Dr. Setiabudi No. 229, Bandung 40154, Indonesia;
[email protected]
4 Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505, Japan
* Correspondence: [email protected] (B.A.B.); [email protected] (M.A.)

Abstract: Temperature management for battery packs installed in electric vehicles is crucial to ensure
that the battery works properly. For lithium-ion battery cells, the optimal operating temperature is in
the range of 25 to 40 ◦ C with a maximum temperature difference among battery cells of 5 ◦ C. This





 work aimed to optimize lithium-ion battery packing design for electric vehicles to meet the optimal
operating temperature using an air-cooling system by modifying the number of cooling fans and
Citation: Widyantara, R.D.; Naufal,
M.A.; Sambegoro, P.L.; Nurprasetio,
the inlet air temperature. A numerical model of 74 V and 2.31 kWh battery packing was simulated
I.P.; Triawan, F.; Djamari, D.W.; using the lattice Boltzmann method. The results showed that the temperature difference between the
Nandiyanto, A.B.D.; Budiman, B.A.; battery cells decreased with the increasing number of cooling fans; likewise, the mean temperature
Aziz, M. Low-Cost Air-Cooling inside the battery pack decreased with the decreasing inlet air temperature. The optimization showed
System Optimization on Battery Pack that the configuration of three cooling fans with 25 ◦ C inlet air temperature gave the best performance
of Electric Vehicle. Energies 2021, 14, with low power required. Even though the maximum temperature difference was still 15 ◦ C, the
7954. https://doi.org/10.3390/ configuration kept all battery cells inside the optimum temperature range. This finding is helpful
en14237954
to develop a standardized battery packing module and for engineers in designing low-cost battery
packing for electric vehicles.
Academic Editor: Artur Bartosik

Keywords: electric vehicle; battery thermal management system; optimization; lattice Boltzmann
Received: 21 October 2021
method
Accepted: 22 November 2021
Published: 28 November 2021

Publisher’s Note: MDPI stays neutral

1. Introduction
with regard to jurisdictional claims in
published maps and institutional affil- In the last three decades, electric vehicles (EVs) have been developed rapidly and have
iations. brought a massive transformation in the automotive industry due to their low emission
and energy-efficient advantages over internal combustion engine vehicles (ICEV) [1,2].
However, despite its rapid development, the current technology of EV still has several
drawbacks compared to ICEV. One of the main disadvantages is the high battery cost,
Copyright: © 2021 by the authors.
which is more than 40% of the total EV price [3]. This high cost is partly caused by the
Licensee MDPI, Basel, Switzerland.
advanced technological implementation in the battery packing to assure high performance
This article is an open access article
and high safety standards, such as complex cooling systems, massive structural packing
distributed under the terms and protection, and advanced electrical system for energy management and control.
conditions of the Creative Commons The batteries experience high charging and discharging rates during EV operation,
Attribution (CC BY) license (https:// increasing their surface temperature [4]. The frequency of charging and discharging also
creativecommons.org/licenses/by/ impacts battery degradation, decreasing the battery’s lifetime [5]. To realize the best
4.0/). performance and longest lifetime, the batteries need to be operated under the optimal

Energies 2021, 14, 7954. https://doi.org/10.3390/en14237954 https://www.mdpi.com/journal/energies

Energies 2021, 14, 7954 2 of 14

temperature condition. This optimal temperature condition for a lithium-ion battery

consists of two terms: (1) the optimal operating temperature range, which states the
temperature range where a battery cell gives optimal charging and discharging rate while
maintaining the longest life cycle; and (2) the maximum temperature difference, which
states the maximum difference in temperature between each battery cells to one another to
provide relatively uniform charging and discharging rates.
The optimal temperature for lithium-ion battery cells to operate is in the range of
25 to 40 ◦ C, with a maximum temperature difference among battery cells of 5 ◦ C [6].
Operating outside the optimal temperature range can decrease the battery’s performance
significantly [7]. Moreover, safety issues like thermal runaway may arise when the battery
operates in high-temperature conditions [8]. Temperature differences among battery cells
may occur due to an inappropriate cooling system, especially when air cooling is applied.
These differences may cause each battery cell to have different charging–discharging rates.
Therefore, they are not beneficial because the battery management system (BMS) must
work hard in balancing the state of charge (SOC) of the batteries, which can degrade its
lifespan faster.
A battery thermal management system (BTMS) plays an essential role in maintaining
the temperature of batteries at the optimal operating temperature [9]. An optimum BTMS
can also reduce the workload of the BMS by lowering the temperature differences among
battery cells. Therefore, various kinds of BTMS are applied and installed in the battery
packing, such as air cooling [10], liquid cooling [11], heat pipe [12], and phase change
materials (PCMs) [13]. Being a novel medium for BTMS and having high efficiency and
stable performance in extreme conditions, PCMs have gained popularity in recent times.
However, PCMs have the disadvantages of having low conductivity and needing to be
regenerated after completely melted [14].
On the other hand, heat pipe systems do not suffer from these drawbacks, and they
do not require an external power supply. They have high conductivity and efficiency in
reducing battery temperature rise, but the equipment is complicated and not conducive
to the practical applications of EV [15,16]. Another thermal management system, liquid-
cooled BTMS, is also complex with its many supporting devices, like pumps, fans, and
pipes, making it costly and vulnerable to the risk of leakage that may lead to a short
circuit [17].
Another cooling method in BTMS is an air-cooling system. This system has a simple
configuration, low initial and maintenance cost, simple integration, and it possesses no risk
of leakage, making it more favorable in the market compared to the methods mentioned
above. Moreover, air-cooling systems can significantly lower battery manufacturing costs,
which directly reduces the EV price. However, despite its advantages, low heat transfer
coefficient, uneven temperature distribution, and low efficiency are the main drawbacks
of air-cooled BTMS [18,19]. To overcome the low heat transfer coefficient of air as the
cooling medium, a hybrid system of BTMS was developed either by combining air-cooling
with PCMs [20] or by integrating air-cooling with mini-channel liquid cooling [21]. These
studies successfully lower the battery’s temperature, but on the other hand, increase the
BTMS power consumption.
Adjusting the structure and flow configuration of the battery pack could also be
beneficial to lower the temperature difference between battery cells and increase the
efficiency of air-cooled BTMS. Several studies were conducted to achieve this result by
optimizing the shape of battery pack [22,23]. Xu et al. [22] discovered that a horizontal
battery pack with a double U-type duct could improve the heat dissipation performance of
the air-cooling system in various conditions. Zhang et al. [23] minimized the temperature
difference in battery packs for prismatic battery cells for Z-, U-, and I-types air-cooled BTMS
by optimizing the widths of parallel cooling channels and divergence/convergence ducts.
Other studies also have focused on improving the air-cooled BTMS by adding parts to the
battery pack [24,25]. Mohammadian et al. [24] studied that thermal management of air-
cooling systems of high-power lithium-ion batteries could be enhanced by implementing
Energies 2021, 14, 7954 3 of 14

aluminum metal porous. Hong et al. [25] improved air-cooled BTMS performance in
reducing temperature differences by adding a secondary vent. And some studies did both,
optimizing the shape and adding parts to the battery pack. Jiaqiang et al. [26] improved the
performance of the air-cooling strategy by locating the lateral inlet and outlet on different
sides and utilizing the baffle plates. Other studies have tried to improve air-cooled BTMS
performance by optimization. Liu et al. [27] performed manifold size optimization to
improve J-type BTMS thermal performance under varying working conditions, resulting
in the optimal configurations for each battery working condition. To achieve optimal
performance for different battery working conditions, a valve control mechanism was
needed to control the manifold size. Chen et al. [28] found that optimization of airflow
parallel outlet position can improve the performance of J-type BTMS for prismatic battery
cells; although, optimization for the airflow parallel inlet position is not as effective.
The studies mentioned above [22–28] may have successfully improved BTMS perfor-
mance, but their practical implementation on EVs is still challenging, especially in reliability,
energy density, and feasibility due to its complexity. Moreover, there has been no standard
and design guideline regulating the battery packing until now, which leads to complex
designs for battery packing. In addition, to improve the applicability, the optimization
design should consider the air properties, number and position of cooling fans, spacing
between cells, and other related factors [29]. These factors must be determined during the
design process, together with maintaining lower manufacturing costs.
Despite the numerous studies conducted to improve the performance of air-cooling
systems either by combining the air-cooling system with another type of cooling system or
adjusting the structure and airflow configuration of the battery pack, not a lot have been
done by optimizing the number of cooling fans and the inlet air temperature. By optimizing
these parameters, this research aimed to develop a simple yet reliable air-cooling system
that can simultaneously maintain high energy density by conditioning the battery tempera-
ture inside the optimal temperature condition while having low manufacturing cost.
Thermal analysis was necessary to investigate the effect of the number of cooling fans
and inlet air temperature on the BTMS performance for the optimization. Some studies have
performed thermal analysis for electric vehicle batteries using simulation. For example,
Raharjo et al. [30] conducted a thermal analysis of modular battery by computational fluid
dynamics (CFD) simulation to understand its thermal behavior, prevent overheating, and
maintain battery life. Divakaran et al. [31] performed finite element simulation to analyze
18,650 lithium-ion batteries’ thermal behavior under two conditions: with and without
cooling systems.
To understand the cooling phenomenon, the battery packing was built in a 3D numer-
ical model and analyzed using CFD simulation based on the lattice Boltzmann method
(LBM). The effects of the number of cooling fans and inlets on air velocity and temperature
distribution inside the battery pack were revealed, and the simulation results from the
LBM analysis were compared. The effects of the number of cooling fans and inlets along
with the inlet air temperature on the performance of BTMS were then discussed thoroughly.
Furthermore, an optimized cooling strategy for air-cooled BTMS based on the consideration
of temperature distribution and power consumption was developed. The results of this
study are significant to develop a standardized battery packing module and enrich the
literature on electric vehicle battery pack optimization.

2. Numerical Model and Simulation

2.1. Heat Generation in Battery Cell
In battery packing, the battery cells generate heat during the charging and discharging
processes. The initial temperature was set according to the ambient temperature. The
heat generation of the lithium-ion battery module consisted of two primary sources: the
irreversible heat caused by the Joule’s heating based on the internal resistance (Qirr ) and
Energies 2021, 14, 7954 4 of 14

the reversible heat inside the battery (Qrev ). From these sources, the battery heat generation
can be represented by Equation (1) [26].
 
dE
Q= I 2 Re − I T (1)
dT

where I is the current flowing through the battery (A), Re is the internal resistance (Ω), T
is the temperature (K), dE/dT is the entropy coefficient (V/K), and Q is the battery heat
generating rate (W). Re and dE/dT are a function of battery temperature (T) and SOC as
shown in Equations (2) and (3), respectively.

−112 × SOC3 − 0.203 × SOC2 × T + 0.000737 × SOC × T 2

 

Re =  +0.00000753 × T 3 + 301 × SOC2 − 0.144 × SOC × T  × 10−3 (2)

−0.0061 × T 2 − 188 × SOC + 1.28 × T + 23.6

dE  
= −0.342 + 0.979 × SOC − 1.49 × SOC2 + 0.741 × SOC3 × 10−3 (3)
dT
To model the battery in an extreme condition, we used a current of 2.6 A for 1 C
discharge rate, battery temperature of 20 ◦ C, SOC of 0.1, along with the parameters shown
in Table 1. The parameters were under the assumption of operation in a tropical region
with the already included ambient temperature. Based on Equations (1) to (3), the lower
the SOC or the battery temperature, the higher the heat generated from the battery. SOC
of 0.1 was the limit value used in the experiment by Jiaqiang et al. according to the
recommendation of the battery manufacturer [26]. Meanwhile, the battery temperature of
20 ◦ C was used as the low temperature in possible operating conditions in a tropical region
with an average ambient temperature of 30 ◦ C. The heat generation rate of the battery
module was calculated with the result of 81.02 W. This value was used for the simulation
input for battery cell heat generation.

Table 1. The specification of the lithium-ion battery investigated in this study.

Characteristics Specifications
Nominal capacity a 2.6 Ah
Nominal voltage a 3.7 V
Cell mass a 0.0475 kg
Thermal conductivity in the axial direction b 37.6 W/m·K
Thermal conductivity in the radial direction b 0.2 W/m·K
Specific capacity c 1200 J/kg·K
Cell diameter 18.3 mm
Cell length 64.5 mm
a Ref. no [26]. b Ref. no [32]. c Ref. no [33].

2.2. System Design

The battery pack design in this study is shown in Figure 1. The interconnection of
the battery cells developed within the battery pack created an electrical and mechanical
connection, and the casing covered the mechanical requirements for the battery pack. The
battery pack also consisted of stiffeners and acrylic to support the battery pack under static
and dynamic loads transferred from an electric frame or chassis. The conductor plate was
used to flow electricity from the battery to other electrical components and vice versa. The
BMS, which could be an active or passive balancer, was placed inside the BMS casing.
The inlet and outlet of the airflow were created on the lateral side of the battery pack
casing to investigate the effect of forced convection by the cooling fan on the temperature
distribution within the battery pack. The battery cells used in this battery pack were lithium
nickel manganese cobalt oxide (NMC) 18,650 cylindrical batteries, and the specification is
presented in Table 1. The 18,650 cylindrical battery type was employed due to its popularity
in its application for electric vehicle battery. This module was suitable for electric trike
Energies 2021, 14, 7954 5 of 14

and city cars. The electric trike, in particular, is currently being developed in our research
facilities and has been presented in several studies, such as by Reksowardojo et al. [34].

Figure 1. Battery pack design and its simplification model.

The original geometry of the battery pack was complex, so it was difficult to configure
the airflow within the battery pack iteratively with the Ansys Discovery Live software.
This situation could also lead to continuous errors and long simulation time. To achieve
the optimum results using the software, the battery pack was modeled as a simplified
form without removing the components sensitive to the heat generation within the battery
pack. The battery cells themselves consisted of layers of cathode, anode, separator, and
current collector. Although these different layers had their thermal properties, the detailed
structure of the cylindrical cell presented in Figure 1 had an insignificant impact on the
thermal performance of the battery, according to many references [35,36]. Therefore, the
battery parameters listed in Table 1 were utilized as the equivalent values representing a
whole battery cell.
The simplified model of the battery pack was created in the form of 240 cells enclosed
by a box, as shown in Figure 1. Each battery cell is represented as a cylinder with a diameter
of 18.3 mm and a height of 64.5 mm. The offset between the 240 cylinders and the box was
1 mm on each surface. This close proximity was applied to ensure the designed battery
pack had a high density and compact geometry as space and weight in EV are the main
constraints to increase energy density [37]. The distance between the center of the two
adjacent cylinders was 20 mm, and the gap between them was 1.7 mm. The pack would
act as the fluid, and the inlet came from one lateral side and the outlet on the other.
Forced convection utilized in this system was carried out using a cooling fan. The
available area for the cooling fan was 240 mm × 40 mm; therefore, the maximum dimension
of the cooling fan was 40 mm × 40 mm. The selected cooling fan was a chip cooler
AP0405MX-J70 from Adda Corp. with a rated power of 0.7 W and a maximum airflow of
4.7 CFM or 7.99 m3 /h.
The effect of fluid flow within the battery pack on the temperature distribution was
studied by conducting LBM-based transient CFD simulations. For the simulation, the
airflow inlet and outlet were positioned at the lateral opposing end of the battery pack,
while the other sides were set as the wall. Each inlet was accompanied by one cooling fan.
For simplicity, from this point forth, the number of cooling fans and inlets will be called
the number of inlets only, without mentioning the cooling fans. The number of inlets was
varied into four configurations of 1-, 2-, 3-, and 4-inlets to investigate the effect of airflow
configuration on temperature distribution inside the battery pack. Afterward, the inlet
air temperature was varied by 20, 25, and 30 ◦ C for each airflow configuration to inves-
tigate the effect of inlet air temperature on temperature distribution and to calculate the
power required for each variation. Finally, the configuration with minimum temperature
distribution and power consumption was chosen as the optimum cooling strategy.
Energies 2021, 14, 7954 6 of 14

2.3. Lattice Boltzmann Method

The simulation was conducted using LBM that discerned air at its mesoscopic scale.
LBM has an advantage in its excellent numerical stability and can be solved efficiently on
parallel computers [38]. Furthermore, due to its diffusive particle modeling, LBM could
handle the simulation of complex topology [39]. In LBM, the motion of the air particles
is modeled in a completely disordered direction and random manner whose velocities
are distributed around the mean value. In this study, the structure is introduced as the
Bhatnagar–Gross–Krook (BGK) (Equation (4)) model is discretized in the lattice unit of
space x and time t [40].
Δ Δ − Δ − Δ
τ
1  eq
f i ( x + ei ∆t, t + ∆t) − f i ( x, t) = ∆t f ( x, t) − f i ( x, t) ) + ∆tei Fk (4)
𝑓 τ i ∆
τ
where f i is the distribution function of air particles, ∆t is the lattice time step, Fk is the
external force, ei is the discrete lattice velocity vector of a particle in a link, τ is the single
eq
relaxation time, f i is the local equilibrium distribution function, and i represents the air
particle in specified lattice point.
The LBM simulation was conducted using ANSYS Discovery Live to generate instant
results of the velocity and temperature distribution within the battery pack. The model
used in the LBM simulation was created using ANSYS Space Claim that was integrated
with ANSYS Discovery Live. The dimension of each inlet was 40 mm × 40 mm, and the
outlet dimension was 240 mm × 40 mm. The volume of the fluid was then extracted from
the inlet to the outlet that enclosed the cylinders. After this process, airflow from the inlet
to the outlet could be seen in real-time or live. The flow velocity at the inlet was set to
1.4 m/s according to the cooling fan characteristics, and at the outlet, the relative pressure
was set to 0 MPa. The ambient temperature was 30 ◦ C according to the average ambient
temperature for the tropical region where the BTMS was designed to be used. The air
was assumed to be dry air having low thermal conductivity to represent the worst cooling
condition. The thermal condition of the battery cells was set by adjusting the heat flow
for each cylinder with the calculated heat generation rate of 0.338 W. The example of the
model in ANSYS Discovery Live, consisting of battery cells and the volume of the fluid
within the battery pack for four inlets, is shown in Figure 2.

(a) (b)
Figure 2. Example of LBM model (a) battery cells configuration (b) battery pack for 4 inlets.

The solution was created by assigning the inlet region and the outlet region enclosing
the whole model. This region that enclosed the cylinders was extracted as the volume
of fluid, which was the volume of air flowing. After the fluid volume was extracted, the
velocity contour of the air flowing could be investigated from the maximum value or
the small particles flowing within the battery packs. The number of particles flowing
within the battery pack could be adjusted by setting the solution’s speed or fidelity, where
high-speed results reduced the fidelity, and the high fidelity reduced the speed results.
This fidelity was based on the distance between the lattice points. The high speeds could
generate the temperature contour within seconds, albeit not very accurately, because when
Energies 2021, 14, 7954 7 of 14

the high-speed result was chosen, fewer particles of air were flowing within the cells.
In this simulation for the 3D geometry model, the setting was adjusted to have higher
fidelity than the speed, so the contour of the temperature distribution would resemble
the real condition more accurately. The simulation in this work was performed in two
configurations. The former was varying the number of inlets into 1, 2, 3, and 4 inlets with
a constant inlet air temperature of 30 ◦ C to study the effect of airflow configuration on
temperature distribution. The latter was varying both the number of inlets (1, 2, 3, and
4 inlets) and the inlet air temperature (20, 25, and 30 ◦ C) to find the optimum cooling
strategy for the air-cooled BTMS.

3. Results and Discussion

3.1. LBM Simulation Results
The effect of airflow configuration on temperature distribution within the battery
pack was studied by varying the number of inlets and simulated using LBM-based CFD. A
constant air inlet temperature of 30 ◦ C was used to represent the average air temperature of
the tropical region where the BTMS was intended to be used. The performance of air-cooled
BTMS could be represented by the air velocity and the temperature inside the battery pack.
The air velocity affected the convective heat transfer of the air and indicated the amount of
air flowing and taking the heat generated by the batteries. Meanwhile, the temperature
was the main objective of BTMS and represented its performance in providing an optimal
operating condition for the batteries. In this section, the distributions and the average
values of air velocity and the temperature inside the battery pack were compared.
The velocity and temperature contour, along with the batteries whose temperatures
were maximum and minimum of one, two, three, and four-inlets configuration, are pre-
sented in Figures 3–6 respectively. The velocity contour described the velocity distribution
of the air traveling across the battery pack and indicated how the air could reach every spot
inside the battery pack. The temperature contour showed the temperature distribution
inside the battery pack after being cooled by the air.

(a) (b)

Figure 3. The contour of (a) velocity and (b) temperature of 1 inlet model.

(a) (b)

Figure 4. The contour of (a) velocity and (b) temperature of 2 inlet model.
Energies 2021, 14, 7954 8 of 14

(a) (b)

Figure 5. The contour of (a) velocity and (b) temperature of 3 inlet model.

(a) (b)

Figure 6. The contour of (a) velocity and (b) temperature of 4 inlet model.

The air entered the battery pack through the inlet on the left side and exited through
the outlet on the right side while taking the heat generated by the batteries in the process.
The arrows on the left side of the figure indicate the number and locations of the inlet
and the airflow direction. The air velocity had its maximum value around the inlet as the
nearest area to the cooling fan. It gradually decreased as it reached the outlet due to the
decreasing kinetic energy. On the contrary, the battery temperature was at the minimum
value around the inlet and gradually increased as it got further. This happened because the
convective heat transfer decreased due to the rising temperature of the air as it traveled
through the battery pack while taking the heat from the batteries. This phenomenon led to
the temperature difference that resulted in different charging/discharging rates between
battery cells, causing a heavy workload for BMS. Excessive workload caused inefficient
equalization, especially for BMS with an active balancing method, leading to non-optimal
power generation and reducing the batteries’ lifespan faster. Ideally, the simulation result
needs to be validated with an experiment, but obtaining velocity and temperature contour
from an experiment might be challenging. However, the current simulation result was
sufficient to show the temperature distribution, which represents the performance from
each configuration.
The overall velocity and temperature results were represented by their average value.
The average value of air velocity and batteries temperature from each configuration are
plotted in Figure 7. The results generated showed that air velocity increased while the
temperature decreased with the increasing number of inlets. The increase in the number of
inlets meant an increase in the number of cooling fans. This led to more power pushing
the air from the inlet to the outlet, resulting in higher air velocity and more air particles
transferring the heat from the batteries. This condition, in turn, caused a lower temperature
as the number of inlets increased. The lowest average temperature of 36.4 ◦ C was achieved
by the four inlets configuration; however, the maximum battery temperature was still
42.2 ◦ C, which is higher than the maximum optimal temperature range of 40 ◦ C. Therefore,
an additional optimization strategy was needed for the air-cooled BTMS to meet the optimal
Energies 2021, 14, 7954 9 of 14

temperature requirement. In the next section, we varied the inlet air temperature and the
number of inlets to obtain the optimum cooling strategy.

2.5 60

2.0 50

Temperature (°C)
Velocity (m/s)
40
1.5
30
1.0
20
0.5 10

0.0 0
1 inlet 2 inlet 3 inlet 4 inlet

Velocity Temperature

Figure 7. The average values of velocity and temperature vs. the number of inlets.

3.2. Optimization of Battery Packing Design

In addition to the number of inlets, the inlet air temperature was varied to 20, 25,
and 30 ◦ C to find the optimum temperature distribution inside the battery pack. Before
entering the battery pack, the inlet air temperature was the conditioning of air temperature
from its ambient temperature. The simulation is meant to model the battery operation in
a tropical region, with no winter and summer, and where a temperature drop is unlikely
to happen. Therefore, the ambient temperature does not change drastically throughout
the year and can be considered a constant. We set the ambient temperature at 30 ◦ C as it
was the average temperature for the tropical region. Therefore, the inlet air temperature of
20 ◦ C and 25 ◦ C was obtained by cooling the ambient temperature. The number of inlet
and inlet air temperature were fundamental factors that affected the BTMS performance,
which was also simple and easy to manufacture once the design was optimized.
The simulation results of varying the inlet air temperature to each inlet configuration
are presented in Figure 8 as a plot of temperature difference to the average temperature.
It was fair to use average temperature as the representative parameter because one of
the main objectives of the optimization was to obtain the lowest temperature difference
that was still located within the optimal temperature range. The horizontal shaded area
represents the optimal operating temperature, while the vertical one represents the optimal
temperature difference for the lithium-ion battery, and the intersection between the two
areas represents the optimal criteria for an optimized cooling strategy. From Figure 8, it can
also be seen that for constant inlet air temperature, the temperature difference decreases as
the number of inlets increases. The temperature difference was significantly affected by the
number of inlets because higher inlet numbers had a wider inlet area that made it easier
for air to reach every spot inside the battery pack.
Energies 2021, 14, 7954 10 of 14

55
Tinlet 30°C
50
Tinlet 25°C
45
Tinlet 20°C

Average temperature (°C)

40

35 Optimal
operating
30 temperature
25
1 inlet
Maximum temperature

20
2 inlet
difference

15
3 inlet
10
4 inlet
5

0
0 5 10 15 20 25 30 35 40 45
Temperature difference (°C)

Figure 8. Temperature distribution of LBM model.

Meanwhile, the average temperature was affected by the inlet air temperature more
than by the inlet number. This gave the idea of adjusting the inlet number and air tem-
perature parameters to achieve any desired performance. The best airflow configuration
was the four inlets configuration with 25 ◦ C air inlet temperature. This configuration
gave the lowest average temperature and the temperature differences that were still in
the optimum range compared to all other configurations. However, using four cooling
fans meant that the configuration cost more power to operate while potentially being over
ρ ∀
capacity. It is unjustified to have a high-performance BTMS that consumes a large amount
of power since the objective of BTMS is to optimize the power generation for the vehicle
operation. Therefore, the power required for each cooling fan and the inlet air temperature
should also be considered to obtain the most energy-efficient cooling strategy giving the
best performance. ρ
The required power was ∀ calculated as the sum of the power required for operating
the cooling fans and generating the sensible heat. The sensible heat represented the power
needed to obtain the inlet air temperature of 20 ◦ C and 25 ◦ C by cooling the air from its
ambient temperature of 30 ◦ C before entering the battery pack. The sensible heat and the
power required were calculated from Equations (5) and (6), respectively.

Hs = C p × ρ × ∀ × dT (5)

P = n × Pf + Hs (6)
where Hs is the sensible heat (kW), Cp is the specific heat of air (1.006 kJ/kg·◦ C), ρ is the
density of air (1.202 kg/m3 ), ∀ is air volume flow (m3 /s), dT is the temperature difference
(◦ C), P is the power required (W), n is the number of cooling fans, and Pf is the cooling fan
rated power (W).
The power required and the resulting temperature of all cooling strategies are plotted
in Figure 9. Each strategy is represented by a marker that indicates its required power in x
axis, and average temperature in y axis. The marker shape represents the inlet number of
one, two, three, and four inlets by triangle, square, diamond, and circle, respectively. The
color represents the inlet air temperature of 20 ◦ C, 25 ◦ C, and 30 ◦ C by blue, green, and red,
respectively. The dashed vertical line extending from the top and bottom of the markers
represent the temperature range of the battery cells resulting in each corresponding config-
Energies 2021, 14, 7954 11 of 14

uration. The green shaded area represents the area of optimal operational temperature for
the lithium-ion battery.

75 No inlet; T (˚C)
1; 20
65 1; 25

1; 30
Temperature (˚C)

55
2; 20

45 2; 25

2; 30
35 3; 20

3; 25
25
3; 30

15 4; 20
0 20 40 60 80 100 120
4; 25
Required Power (W)
4; 30

Figure 9. Temperature versus required power.

The optimal cooling strategy has the lowest median temperature and minimum
temperature difference that meets the optimal temperature condition while having the
least power required. Therefore, the optimal cooling strategy in Figure 9 is represented
by the closest marker to the objective point (0 W, 25 ◦ C), whose marker and dashed line
are located inside the green shaded area. The optimization utilized the Qhull algorithm
to find the nearest marker to the objective point. As a result, it was found that the three
inlets configuration with the inlet air temperature of 25 ◦ C was the best cooling strategy in
terms of performance and power consumption and was chosen as the optimized cooling
strategy for air-cooled BTMS. The configuration gave a mean temperature of 33.1 ◦ C,
which met the optimal temperature condition for a lithium-ion battery. However, the
maximum temperature difference in this configuration was 14.9 ◦ C, which did not meet the
requirement. In fact, the temperature differences in all simulated configurations did not
meet the requirement. This is the challenge of air-cooling applications for a battery pack.
Compared to the available BTMSs, the optimized design from this study has several
advantages. Firstly, the optimized cooling strategy is low cost and easy to manufacture
since there are no complex parts needed and it only consists of a few parts. In addition,
the optimized design can meet the optimal temperature condition for a lithium-ion battery
with low power consumption compared to other cooling methods such as water cooling,
heat pipe, and PCMs. Therefore, the three inlets configuration with the inlet air temperature
of 25 ◦ C can be proposed as a standard design for a battery pack air-cooling system. For
future work, when the simplicity is not the main objective, employing variable cooling
fan speed might improve the performance while increase the system complexity for the
additional control system. This study only discusses the optimization from the temper-
ature distribution perspective. Therefore, further research is required to evaluate other
important aspects, such as structural strength and water protection, to complement the
design optimization of the battery pack air-cooling system.
Energies 2021, 14, 7954 12 of 14

4. Conclusions
The effect of the number of inlets and the inlet air temperature on air-cooled BTMS
performance was studied using a LBM-based CFD simulation. To find the optimum cooling
strategy, the power required to operate the configuration was also considered. In the end, an
optimized cooling strategy for an air-cooled BTMS consisting of three inlets configuration
with an inlet air temperature of 25 ◦ C is recommended to achieve optimum performance
in terms of temperature distribution and power consumption. The configuration gave
the average temperature of 33.1 ◦ C and the maximum temperature difference of 14.9 ◦ C.
Although the temperature difference is wider than the requirement of 5 ◦ C, the chosen
cooling strategy kept all the batteries inside the range of optimum operating temperature.
Therefore, the optimized design can be proposed as a standard for a battery packing cooling
system. For future work, an investigation on the effect of employing variable speed for
cooling fans and a study for structural strength and water protection for the air-cooling
systems are recommended.

Author Contributions: Conceptualization, P.L.S., F.T., B.A.B. and M.A.; methodology, P.L.S., I.P.N.,
B.A.B. and M.A.; modelling and design, M.A.N., F.T. and D.W.D.; validation, P.L.S. and I.P.N.; formal
analysis, R.D.W., M.A.N. and D.W.D.; writing—original draft preparation, R.D.W., M.A.N., A.B.D.N.
and B.A.B.; writing—review and editing, R.D.W., P.L.S., F.T., D.W.D., A.B.D.N., B.A.B. and M.A.;
illustration, R.D.W. and I.P.N.; supervision, P.L.S., I.P.N., A.B.D.N. and B.A.B.; funding acquisition,
B.A.B. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Indonesia Endowment Fund for Education (LPDP) under
Research and Innovation Program (RISPRO) for electric vehicle development with contract no.
PRJ-85/LPDP/2020.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Conflicts of Interest: The authors declare that they have no known competing financial interests or
personal relationships that could have appeared to influence the work reported in this paper. The
funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in
the writing of the manuscript, or in the decision to publish the results.

References
1. Yuan, X.; Li, X. Mapping the technology diffusion of battery electric vehicle based on patent analysis: A perspective of global
innovation systems. Energy 2021, 222, 119897. [CrossRef]
2. Aziz, M.; Oda, T.; Mitani, T.; Watanabe, Y.; Kashiwagi, T. Utilization of Electric Vehicles and Their Used Batteries for Peak-Load
Shifting. Energies 2015, 8, 3720–3738. [CrossRef]
3. Tao, Y.; Huang, M.; Chen, Y.; Yang, L. Orderly charging strategy of battery electric vehicle driven by real-world driving data.
Energy 2020, 193, 116806. [CrossRef]
4. Peng, X.; Cui, X.; Liao, X.; Garg, A. A Thermal Investigation and Optimization of an Air-Cooled Lithium-Ion Battery Pack.
Energies 2020, 13, 2956. [CrossRef]
5. Huda, M.; Koji, T.; Aziz, M. Techno Economic Analysis of Vehicle to Grid (V2G) Integration as Distributed Energy Resources in
Indonesia Power System. Energies 2020, 13, 1162. [CrossRef]
6. Malik, M.; Dincer, I.; Rosen, M.; Fowler, M. Experimental Investigation of a New Passive Thermal Management System for a
Li-Ion Battery Pack Using Phase Change Composite Material. Electrochim. Acta 2017, 257, 345–355. [CrossRef]
7. Gandoman, F.; Jaguemont, J.; Goutam, S.; Gopalakrishman, R.; Firouz, Y.; Kalogiannis, T.; Omar, N.; Mierlo, J.V. Concept of
reliability and safety assessment of lithium-ion batteries in electric vehicles: Basics, progress, and challenges. Appl. Energy 2019,
251, 113343. [CrossRef]
8. Sambegoro, P.L.; Budiman, B.A.; Philander, E.; Aziz, M. Dimensional and Parametric Study on Thermal Behaviour of Li-ion
Batteries. In Proceedings of the 2018 5th International Conference on Electric Vehicular Technology (ICEVT), Surakarta, Indonesia,
30 October 2018; pp. 123–127. [CrossRef]
9. Al-Zareer, M.; Dincer, I.; Rosen, M. Heat transfer modeling of a novel battery thermal management system. Numer. Heat Transf.
Part A Appl. 2018, 73, 277–290. [CrossRef]
10. Liu, Y.; Zhang, J. Self-adapting J-type air-based battery thermal management system via model predictive control. Appl. Energy
2020, 263, 114640. [CrossRef]
Energies 2021, 14, 7954 13 of 14

11. Dong, F.; Song, D.; Ni, J. Investigation of the effect of U-shaped mini-channel structure on the thermal performance of liquid-cooled
prismatic batteries. Numer. Heat Transf. Part A Appl. 2019, 77, 105–120. [CrossRef]
12. Smith, J.; Singh, R.; Hinterberger, M.; Mochizuki, M. Battery thermal management system for electric vehicle using heat pipes.
Int. J. Therm. Sci. 2018, 134, 517–529. [CrossRef]
13. Talluri, T.; Kim, T.H.; Shin, K.J. Analysis of a Battery Pack with a Phase Change Material for the Extreme Temperature Conditions
of an Electrical Vehicle. Energies 2020, 13, 507. [CrossRef]
14. Lazrak, A.; Fourmigué, J.; Robin, J. An innovative practical battery thermal management system based on phase change materials:
Numerical and experimental investigations. Appl. Therm. Eng. 2018, 128, 20–32. [CrossRef]
15. Chi, R.G.; Rhi, S.H. Oscillating Heat Pipe Cooling System of Electric Vehicle’s Li-Ion Batteries with Direct Contact Bottom Cooling
Mode. Energies 2019, 12, 1698. [CrossRef]
16. Putra, N.; Ariantara, B.; Pamungkas, R. Experimental investigation on performance of lithium-ion battery thermal management
system using flat plate loop heat pipe for electric vehicle application. Appl. Therm. Eng. 2016, 99, 784–789. [CrossRef]
17. Fang, G.; Huang, Y.; Yuan, W.; Yang, Y.; Tang, Y.; Ju, W.; Chu, F.; Zhao, Z. Thermal management for a tube–shell Li-ion battery
pack using water evaporation coupled with forced air cooling. RSC Adv. 2019, 9, 9951–9961. [CrossRef]
18. Li, X.; Zhao, J.; Yuan, J.; Duan, J.; Liang, C. Simulation and analysis of air cooling configurations for a lithium-ion battery pack. J.
Energy Storage 2021, 35, 102270. [CrossRef]
19. Qin, P.; Sun, J.; Yang, X.; Wang, Q. Battery thermal management system based on the forced-air convection: A review. eTransporta-
tion 2021, 7, 100097. [CrossRef]
20. Ling, Z.; Wang, F.; Fang, X.; Gao, X.; Zhang, Z. A hybrid thermal management system for lithium ion batteries combining phase
change materials with forced-air cooling. Appl. Energy 2015, 148, 403–409. [CrossRef]
21. Yang, W.; Zhou, F.; Zhou, H.; Wang, Q.; Kong, J. Thermal performance of cylindrical lithium-ion battery thermal management
system integrated with mini-channel liquid cooling and air cooling. Appl. Therm. Eng. 2020, 175, 115331. [CrossRef]
22. Xu, X.; He, R. Research on the heat dissipation performance of battery pack based on forced air cooling. J. Power Source 2013, 240,
33–41. [CrossRef]
23. Zhang, J.; Wu, X.; Chen, K.; Zhou, D.; Song, M. Experimental and numerical studies on an efficient transient heat transfer model
for air-cooled battery thermal management systems. J. Power Source 2021, 490, 229539. [CrossRef]
24. Mohammadian, S.; Rassoulinejad-Mousavi, S.; Zhang, Y. Thermal management improvement of an air-cooled high-power
lithium-ion battery by embedding metal foam. J. Power Source 2015, 296, 305–313. [CrossRef]
25. Hong, S.; Zhang, X.; Chen, K.; Wang, S. Design of flow configuration for parallel air-cooled battery thermal management system
with secondary vent. Int. J. Heat Mass Transf. 2018, 116, 1204–1212. [CrossRef]
26. Jiaqiang, E.; Yue, M.; Chen, J.; Zhu, H.; Deng, Y.; Zhu, Y.; Zhang, F.; Wen, M.; Zhang, B.; Kang, S. Effects of the different air cooling
strategies on cooling performance of a lithium-ion battery module with baffle. Appl. Therm. Eng. 2018, 144, 231–241. [CrossRef]
27. Liu, Y.; Zhang, J. Design a J-type air-based battery thermal management system through surrogate-based optimization. Appl.
Energy 2019, 252, 113426. [CrossRef]
28. Chen, K.; Wu, W.; Yuan, F.; Chen, L.; Wang, S. Cooling efficiency improvement of air-cooled battery thermal management system
through designing the flow pattern. Energy 2019, 167, 781–790. [CrossRef]
29. Tete, P.; Gupta, M.; Joshi, S. Developments in battery thermal management systems for electric vehicles: A technical review. J.
Energy Storage 2021, 35, 102255. [CrossRef]
30. Raharjo, J.; Wikarta, A.; Sidharta, I.; Yuniarto, M.N.; Rusli, M.R. Thermal analysis simulation of parallel cell in modular battery
pack for electric vehicle application. J. Phys. Conf. Ser. 2020, 1517, 012023. [CrossRef]
31. Divakaran, A.M.; Hamilton, D.; Manjunatha, K.; Minakshi, M. Design, Development and Thermal Analysis of Reusable Li-Ion
Battery Module for Future Mobile and Stationary Applications. Energies 2020, 13, 1477. [CrossRef]
32. Saw, L.H.; Ye, Y.; Tay, A.A.O. Electrochemical–thermal analysis of 18650 Lithium Iron Phosphate cell. Energy Convers. Manag.
2013, 75, 162–174. [CrossRef]
33. Rao, Z.; Qian, Z.; Kuang, Y.; Li, Y. Thermal performance of liquid cooling based thermal management system for cylindrical
lithium-ion battery module with variable contact surface. Appl. Thermal Eng. 2017, 123, 1514–1522. [CrossRef]
34. Reksowardojo, I.K.; Arya, R.R.; Budiman, B.A.; Islameka, M.; Santosa, S.P.; Sambegoro, P.L.; Aziz, A.R.A.; Abidin, E. Energy
Management System Design for Good Delivery Electric Trike Equipped with Different Powertrain Configurations. World Electr.
Veh. J. 2020, 11, 76. [CrossRef]
35. Lu, Z.; Yu, X.; Wei, L.; Qiu, Y.; Zhang, L.; Meng, X.; Jin, L. Parametric study of forced air cooling strategy for lithium-ion battery
pack with staggered arrangement. Appl. Therm. Eng. 2018, 136, 28–40. [CrossRef]
36. Wu, Y.; Li, K.; Wang, J.; Ji, S.; Wang, S. Experimental study and numerical modeling on cylindrical lithium-ion power battery
thermal inertia. Energy Procedia 2019, 158, 4396–4401. [CrossRef]
37. Wahid, M.R.; Budiman, B.A.; Joelianto, E.; Aziz, M. A Review on Drive Train Technologies for Passenger Electric Vehicles. Energies
2021, 14, 6742. [CrossRef]
38. Arumuga Perumal, D.; Dass, A.K. A Review on the development of lattice Boltzmann computation of macro fluid flows and heat
transfer. Alex. Eng. J. 2015, 54, 955–971. [CrossRef]
Energies 2021, 14, 7954 14 of 14

39. Zhang, C.; Fakhari, A.; Li, J.; Luo, L.; Qian, T. A comparative study of interface-conforming ALE-FE scheme and diffuse interface
AMR-LB scheme for interfacial dynamics. J. Comput. Phys. 2019, 395, 602–619. [CrossRef]
40. Sheikholeslami, M.; Hayat, T.; Alsaedi, A. Numerical simulation for forced convection flow of MHD CuO-H2 O nanofluid inside a
cavity by means of LBM. J. Mol. Liq. 2018, 249, 941–948. [CrossRef]