Engineering Science and Technology, an International Journal 21 (2018) 957–969

Contents lists available at ScienceDirect

Engineering Science and Technology,

an International Journal
journal homepage: www.elsevier.com/locate/jestch

Full Length Article

Numerical investigation on cooling performance of Li-ion battery

thermal management system at high galvanostatic discharge
R.D. Jilte, Ravinder Kumar ⇑
Department of Mechanical Engineering, Lovely Professional University, Punjab 14411, India

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

Article history: Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs) are clean energy transportation options
Received 21 March 2018 emerged over traditional internal combustion engine. Li-ion batteries are a viable option for EVs and
Revised 18 June 2018 HEVs due to their advantages of high energy density. At high battery discharging condition, there is a sig-
Accepted 19 July 2018
nificant increase in battery temperature and non-uniform cell temperature. The cooling performance of
Available online 10 August 2018
battery module at constant current discharge rate about 6.94 C (25 A) have presented. The two side wall
of the battery module kept fully open for inflow and outflow of cooling media and better heat dissipation.
Keywords:
Larger inter-cell spacing was considered to provide sufficient circulation of cooling air and removal of
Battery thermal management system
(BTMS)
gases generated by the batteries. The heat generation in the battery cell during discharge process has sim-
Li-ion Battery module ulated with the help of user-defined function (UDF). The paper gives insight into a three-dimensional
Battery temperature transient thermal response, flow field and thermal regimes developed in the battery module. Different
Heat generation air temperature profiles are confirmed in the flow direction and across the width and depth of the battery
Transient three-dimensional CFD study pack. At specific zones, the air temperature rises to 7 °C thus indicating the localized heat spots. In the
considered BTMS, maximum cell-to-cell temperature non-uniformity is restricted to 0.11 °C and battery
temperature is lesser than 28 °C despite the high discharge rate and lower cooling air flow condition.
Ó 2018 Karabuk University. Publishing services by Elsevier B.V. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
The energy is required to drive industrial developments and
other mankind needs. Due to limited reserves of fossil fuels and
their environmental effect, efforts have been given to harnesses
alternative sources of energy. The energy generated using solar
and wind route is required to be stored and supply for off-grid
applications due to its intermittent nature. Li-ion battery packages
can be employed for storing the off-grid electricity [1]. Li-ion bat-
teries are a viable option for Electric vehicles (EVs) and Hybrid
Electric Vehicles (HEVs) due to their advantages of high energy
density as compared to other electrochemical batteries [2]. Cur-
rently, lithium-ion batteries have been widely used in the latest
EV, such as the Tesla Model S, Nissan Leaf, and Chevy Volt.
The use of EVs reduces our dependence on fossil fuels and emis-
sion of pollutants and greenhouse gases. Andersen et al. [3]
depicted that up to 40% reduction of greenhouse gas emission is
possible if the EVs get charged with electric power resources using
renewable route. In case of EVs and HEVs battery efficiency is the
⇑ Corresponding author.
E-mail address: ravchauhan8@gmail.com (R. Kumar).
Peer review under responsibility of Karabuk University.
key parameter. During the discharge mode of the battery, cell tem-
perature increases thus reducing the thermal efficiency.
Several factors affect the lifespan of the battery; among cell
temperature plays a significant role [4,5]. At higher discharge rate,
battery temperature increases due to a higher rate of heat genera-
tion in battery [6]. Sato et al. [7] observed that the charging effi-
ciency and lifecycle will be reduced if operating temperatures are
above 50 °C. Usually, Li-ion batteries operate in the optimal tem-
perature range of 20–40 °C [8]; performance falls drastically at
low temperature [9]. Li-ion batteries operating beyond safe tem-
perature range results in major capacity loss; for example for every
degree rise in operating temperature, the calendar life of the bat-
tery reduces almost by two months [10]. Zang et al. observed that
capacity of Li-ion battery decreases up to 95% at low temperature
(
10 °C) compared to at 20 °C [11]. Wu et al. [12] found that capac-
ity of a fresh Li-ion battery at 3C discharge was decreased from
800 mAh to merely 20 mAh after storing at 60 °C. Giuliano et al.
[13] observed at high-rate discharge, there is a significant increase
in battery temperature and non-uniform surface temperature.
Based on aforementioned studies, along with providing safe
operating cell temperature and cell to cell temperature uniformity,
other issues associated with battery usage in EVs and HEVs can be
listed as:

https://doi.org/10.1016/j.jestch.2018.07.015
2215-0986/Ó 2018 Karabuk University. Publishing services by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
958 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

Nomenclature

As Surface area of a single battery cell, (m2) Tfin Area-weighted average air temperature at the inlet of
Bi Biot number (
) the battery domain, (°C)
C Discharge rate (1C is equal to 3.6 Amp) Tfout Area-weighted average air temperature at the inlet of
Cp, Specific heat, (J/kg K) the battery domain, (°C)
D Diameter of battery cell, (m) Ts Area weighted average surface temperature of the bat-
E Cell potential, (Volt) tery cell (°C)
Eoc Open circuit potential, (Volt) T1 Ambient air temperature, (°C)
g Gravitational acceleration (m/s2) V Air flow velocity, (m/s)
h Convective heat transfer coefficient of cooling air Vb Volume of a single battery cell, (m3)
(W/m2K) Vmax Maximum air flow velocity in module, (m/s)
I Current, (Amp)
K Thermal conductivity, W/mK Greek symbols
L Length of the battery cell, (m) q Density, (kg/m3)
Lc Characteristics length, (m) l Dynamic viscosity of air (Pa-s)
mf Mass flow rate of the supplied air, (kg/s)
Nu Nusselt number (
) Abbreviation List
Pr Prandtl number (
) BTMS Battery thermal management system
Qg Rate of heat generation in a battery cell, (J/s)
EV Electric vehicle
Qentropic Entropic heating, (J/s) HEV Hybrid electric vehicle
QOhmic ohmic heating, (J/s) Li-ion Lithium ion
R The internal resistance of battery cell (Ohm) UDF User defined function
Re Reynold number (
)
ST Transverse pitch, (m)
SL Longitudinal pitch, (m) Subscripts
b Battery material
T Instantaneous cell temperature, (°C)
f Air

1. In application, there are large numbers of cells required for EV
or HEV power capacity.
2. The large amount of heat generated from these cells has to
remove from vehicle body.
3. There should be space and scope for removing hazardous gases
formed during battery operation.
4. The arrangement should be compact and lightweight and scal-
able as per capacity.
5. The arrangement should provide safe battery operation
throughout discharge/charge cycle.
The issues listed above can be resolved with battery thermal
management system which can be grouped under three major cat-
egories based on cooling media (Fig. 1). In general, most of the
commercial EVs use air or liquid cooling strategies for effective
control of battery module temperature [2]; air-cooled BTMS is
the simplest and lighter [14].
There are many studies available on cooling performance of bat-
tery thermal management system. He et al. [15] carried out exper-
imental and numerical study on thermal management of multiple
cells Li-ion modules. The experimental setup consists of an open
wind tunnel to provide controlled air cooling. The battery was
tested experimentally for 0, 1, 2.5 and 5 m/s air flow velocities.
Pesaran et al. [16] compared numerically cooling of battery module
by arranging air direction in series or parallel flow. The observed
maximum temperature difference was 8 °C and 18 °C respectively
in parallel and series flow. Sabbah et al. [17] in his experimental
and numerical study found that an increase in air velocity could
not controlled cell temperature below 55 °C when ambient tem-
perature is 45 °C at 6.67 C discharge rate. A C-rate is a measure
of the rate at which a battery is discharged relative to its maximum
capacity. A 1C rate means that the discharge current will discharge
the entire battery in 1 h. Nelson [18] also pointed that if the cell
temperature rises above 66 °C, then it is difficult to cool it to below
52 °C by air-cooling.
Based on studies available on BTMS, the present work attempt
to address the following issues:
1. Air-cooled battery module working at high discharge rate
necessitates throughout understanding of parameters like a
transient thermal response, development of flow field and tem-
perature regimes, cell-to-cell temperature non-uniformity and
location of heat spots for proper design of BTMS.

Fig. 1. Types of battery thermal management systems.

 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969 959

Fig. 2. (a) Proposed battery pack (b) battery module under consideration (c) cell indices.

2. Any changes to BTMS including selection and flow nature of

cooling media, design of fluid inlets/outlets can be followed.
Table 1
3. This work presents insights into first step considering high gal-
Dimensions and thermophysical properties of battery cell [23,24].
vanostatic discharge rate about 6.94 C (1 C is equal to 3.6 A).
Parameters Details 4. The battery module is fully open on two sides for inflow and
Battery details Li-ion battery, cathode: LiMn2O4, outflow of cooling media for better heat dissipation.
anode: Carbon 5. Larger inter-cell spacing is considered to provide sufficient cir-
Capacity, Ah 3.6
culation of cooling air and removal of gases generated by the
Number of cells 9
Diameter (D), mm 42.4 batteries.
Length (L), mm 97.7
Mass of cell, kg 0.3
Density (qb), kg/m3 2007.7 2. Battery thermal management system
Thermal conductivity (Kb) W/mK 1.0 (radial direction)
Specific heat, (Cp,b), J/kg K 837.4 The battery module contains nine cylindrical Li-ion cells
arranged inline in the direction of air flow as shown in Fig. 2. The
960 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

inter-cell spacing are characterized by the transverse pitch (ST) and

longitudinal pitch (SL). The dimensions of the module can be repre-
sented in terms of the radius of a battery cell (R) to facilitate any
addition or deletion of cells for varying the capacity of the battery
module. The present study involves symmetrically placed cells; ST
and SL both are taken as 3  R. The width and breadth of the bat-
tery module are kept as 10  R whereas the height of the module
is taken as L + R. Here L is the length of the battery cell. The spacing
kept above the cells allows good air ventilation and space for cir-
cuit connection. Larger inter-cell spacing is considered to provide
sufficient circulation of cooling air and removal of gases generated
by the batteries [16]. Two sides of the battery module are kept fully
open for inlet and outlet of cooling media. The dimensions of the
Li-ion cell and their thermo-physical and chemical properties are
given in Table 1. The battery material considered as isotropic in
nature [19,20–22]. Therefore battery cell components (cathode,
anode, separator, current collector tabs) can be taken as a homoge-
nous body with constant values of thermal conductivity and speci-
fic heat. This assumption simplifies numerical model with the
lumped system as discussed in the subsequent sections. Fig. 4. Mesh sensitivity analysis.

Fig. 3. Geometry modeled with different mesh size.

 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969 961

Table 2
Constants used in Eq. (10).

Inlet air velocity (m/s) Reynold number, Re C m n

0.1 100–1000 0.52 0.5 0.36
0.5 103 to 2  105 0.27 0.63 0.36

Table 3
Correction factor (F) to be used in Eq. (10).

Number of tubes in longitudinal direction 1 2 3 4 5 7 10 13

Correction factor (F) 0.70 0.80 0.86 0.90 0.93 0.96 0.98 0.99

Table 4
Validation of CFD study.

V (m/s) Vmax (m/s) Reylold number (ReD) Nu Eq (10) [37] Correction factor (F) [29] Nucorrected Eq. (11) Present numerical scheme % difference
h (W/m2K) Eq. (13) Nu Eq. (12)
0.1 0.3 799.30 13.02 0.86 11.20 6.92 11.16
0.37
0.5 1.5 3996.53 44.45 0.86 38.22 23.77 38.34 0.27

3. Battery heat generation model
The development of flow field and temperature pattern in bat-
tery module depends on heat generation in cells, heat dissipation
from cell to surrounding air. The heat generation in a cell involves
electrochemical reaction that raises battery surface temperature.
The heat generation in a battery cell (Qg) can be considered to orig-
inate from two ways: (i) entropic heating (Qentropic) and (ii) ohmic
heating (QOhmic) [25,26].
Q g ¼ Q entropic þ Q Ohmic
system model. In case of lumped capacitance model, the tempera-
ture of the battery cell is uniform throughout the volume. Thermal
gradients within the cell can be neglected. Hence, battery temper-
ature considered as an only time dependent parameter. The selec-
tion of lumped capacitance model can be verified based on a
limiting condition specified in terms of Biot number (Bi). Biot num-
ber is the ratio of convection at the surface of the body to conduc-
tion within the body and can be expressed as:
hLc
Bi ¼ ð5Þ
ð1Þ kb

Ohmic heating or over-potential heat generation is always
taken as positive and produced due to the internal resistance of cell
and associated electrochemical reactions [27]. Entropic heating is
caused due to entropy change during charging/discharging condi-
tions and can be positive or negative based on endothermic or
exothermic reactions. The heat generated due to Ohmic heating
can be computed as:
Q ohmic ¼ IðE
Eoc Þ ¼ I2 R
where I represent the charging or discharging current, E and Eoc are
cell potential and open circuit potential respectively. The internal
resistance (R) depends on the cell temperature (T) [23,15] and can
be expressed by
R ¼
0:0001T 3 þ 0:0134T 2
0:5345T þ 12:407
where resistance is substituted milliohms and cell temperature is in
°C.
The entropic heat generation is calculated using [23]
where h is the convective heat transfer associated with cooling
media (air), kb is the thermal conductivity of battery material, Lc
is the characteristics length calculated by
Vb
Lc ¼ ð6Þ
As
where Vb and As are respectively volume and surface area of a single
battery cell. The surface area is calculated neglecting the top and
ð2Þ bottom surface of the battery cell. This consideration is based on
the assumption that major heat dissipation from the battery cells
occurs in radial directions [28]. Based on air flow rate used in pre-
sent work, the maximum value of heat transfer coefficient is limited
to 25 W/m2K. The computed Biot number is found to be less than
0.1 (criteria to apply lumped capacitance model). Hence lumped
ð3Þ model is used in the present case [29]. Heat generation source term
(Eq. (1)–(4)) is directly modeled as a volumetric phenomenon.
4.2. Numerical model

dEOC There are many experimental and numerical studies reported

Q entropic ¼
IT
dT
A user-defined function (UDF) is written based on Eq. (1)–(4) for
heat generation in the battery cell
4. Numerical procedure and validation
4.1. Lumped capacitance analysis
A simplified model is used for incorporating battery heat gener-
ation. The selected isotropic nature of battery material allows us to
use constant values of thermal conductivity and other physical
properties. This helps in analyzing the applicability of lumped
ð4Þ on battery thermal management system (BTMS). In case of numer-
ical investigations, finite volume approach are mostly preferred
[15,30–34]. Three dimensional model of a battery module is cre-
ated and meshed using Gambit software of CFD package Fluent
6.3 [35]. A pressure based, laminar/.k-epsilon turbulent, incom-
pressible, transient solver is used in the simulation. The solver uses
SIMPLE algorithm for pressure and velocity coupling.
4.3. Governing equations
4.3.1. Fluid domain
The air flow in fluid domain considered as laminar and
incompressible. The computational domain is solved with
962 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

Fig. 5. Transient thermal response of battery module (air velocity = 0.1 m/s, T1 = 22 °C).

three-dimensional governing equations for the conservation of The fluid domain (air flow) and solid domain (heat conduction
mass, momentum, and energy during battery discharge condition. with internal volumetric heat generation) are coupled for simulat-
ing conjugate heat transfer.
4.3.2. Solid domain
In case of pure heat conduction in cell volume with heat gener-
ation, governing equation is written as [36]: 4.4. Boundary conditions

 " #
@T @2T @2T @2T The battery module (Fig. 2b) is based on the uniform flow of
qb C p;b ¼ ks þ þ þ Qg ð7Þ cooling air at the module entry. The battery module is completely
@t @x2 @y2 @z2
open at two sides for inflow and outflow of cooling media. The left
where heat generation in the cell (Qg) is expressed in Eq. (1). qb, Cp,b opening of the battery domain is assigned as velocity inlet whereas
and kf are density, specific heat and thermal conductivity of battery the exit is treated as pressure outlet. The atmospheric air of 22 °C
material. (T1) is used as the cooling medium. CFD model of battery module
 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969
Fig. 6. The temporal cell temperature variation (air velocity = 0.1 m/s, T1 = 22 °C).
comprised of air as fluid zone and battery as a solid zone. The bat-
tery surfaces separating the air zone are considered as coupled wall
condition. No slip boundary condition is applied to battery
surfaces.
4.5. Mesh sensitivity analysis
The battery module was modeled to different mesh sizes (mesh
1: 96928 nodes, mesh 2: 214914 nodes, mesh 3: 376176 nodes and
mesh 4: 741447 nodes) as shown in Fig. 3 to check for grid sensi-
tivity analysis. The computational results are shown in Fig. 4. It can
see that solutions are grid independent after refining mesh. There-
fore mesh 3 (376176 nodes) was adopted to avoid the long compu-
tational time and solution instability.
4.6. Validation of numerical model
The numerical scheme followed in the present study has been
validated considering the case of flow across the bank of the tubes.
Based on the arrangement of cells, air flow velocity (V) supplied at
the inlet of the battery domain and air flow velocity occurring in
inter-cell spacing are different. The maximum velocity occurs at
the minimum flow area between the cells. In the case of an in-
line arrangement, transverse pitch (ST) is used to calculate the
maximum velocity.
ST
V max ¼ V ð8Þ
ST
D
qf V max D
ReD ¼ ð9Þ
lf
where D is the battery cell diameter (42.4 mm) and lf is the
dynamic viscosity of the air. In the present study, inlet air velocity
is taken as 0.1 m/s and 0.5 m/s. The laminar/turbulent model was
selected based on the range of Reynold number. A user-defined
function (UDF) is written based on Eq. (1)–(4) for heat generation
in the battery cell. Second order upwind scheme is used for the
momentum and energy equations. Convergence criteria for the
residuals of continuity and momentum equation were set as 10
3
and 10
6 for energy equation. The convergence is judged by exam-
ining residual levels and also by monitoring relevant integrated
quantities such as heat transfer coefficient. The solutions are
obtained once the convergence criteria are satisfied.
963
For the transient simulation time step of 0.25 s and 20 iterations
per step were chosen.
4.6.1. Evaluation of correlation based Nusselt number
The validation of the CFD model is carried out with empirical
correlations proposed by [37].
0:25
NuD ¼ CRem
D Pr ðPr=Pr s Þ
n
ð10Þ
Eq. (10) is also being used in for validation of CFD results in sim-
ilar forced convection application [38,39,24,23]
The values of the constants used in Eq. (10) are given in Table 2.
Area-weighted average air temperature is determined at the
inlet (Tfin) and outlet of the battery domain (Tfout). All the proper-
ties of air except the Prandtl number Prs were determined at the
mean air temperature (Tfin + Tfout)/2. Prs is determined at battery
surface temperature. The constants used in Eq. (10) are based on
the consideration that air flow is occurring over the sixteen tubes
in the longitudinal direction. In case of a lesser number of tubes,
the correction factor is given in Table 3 [29].
Therefore the corrected Nusselt number is given
Nucorrected ¼ 0:86 NuD ð11Þ
4.6.2. Evaluation of CFD based Nusselt number
The computations results obtained on simulation under both
laminar and turbulent conditions are were used for estimating
CFD based Nusselt number as per procedure outlined below:
i) Nusselt number is calculated taking battery diameter (D) as
characteristic dimension
hD
Nu ¼ ð12Þ
kf
where h is the convective heat transfer coefficient, kf is the thermal
conductivity of air and D is the diameter of the battery cell.
ii) Thermal conductivity of air is determined at the mean air
temperature (Tfin + Tfout)/2. For this purpose, Area-
weighted average air temperature is determined at the inlet
(Tfin) and outlet of the battery domain (Tfout) is measured
iii) In Eq. (12) the unknown heat transfer coefficient (h) are cal-
culated based on the energy balance applied to battery
module
Q ¼ mf Cpf ðT f ;out
T f ;in Þ ¼ hAs DT lm ð13Þ
The area As is the total surface area of the battery cells that dis-
sipates heat due to circulated air. Neglecting the heat dissipated
from the top surface and the bottom surface of the cell, the area
As is calculated considering the cylindrical surface area of the nine
cells.
Based on velocity of air considered and inlet area of domain,
mass flow rate of air (mf) are calculated. The earlier measured air
temperature in step ii is used to get heat transferred from battery
surfaces to cooling air from Eq. (13).
The other unknown DT lm are determined by Eq. (14)
ðT s
T fo Þ
ðT s
T fi Þ
DT lm ¼ ð14Þ
ln½ðT s
T fo Þ
ðT s
T fi Þ
where Ts is the area weighted average surface temperature of the
battery cell
iv) Once all variables involved in Eq. (13) is known, h is calcu-
lated and substituted in Eq. (12) to get Nusselt nulmber.
964 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

Fig. 7. The cooling media (air) temperature contours inside battery module at t = 500 s (air velocity = 0.1 m/s, T1 = 22 °C).

Table 4 shows the comparison between Nusselt number (Eq. (10)). It shows good agreement; hence present numerical
obtained by the CFD model (Eq. (12)) and published correlation scheme is used for analyzing thermal behavior of battery module.
 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969
Fig. 8. Air temperature variation in flow direction.
5. Result and discussions
Numerical simulations have been carried out for constant cur-
rent battery discharge condition at 6.94 C where 1 C is equal to
3.6 A. The flow field is analyzed in battery module based on heat
generations from all nine cells. The results of computations are
presented in terms of transient thermal response module, flow
field and temperature pattern inside battery module.
5.1. Transient thermal response of battery module
The battery cell is subjected to heat generation and its rate can be
computed based on Eq. (1). During the startup of the process, battery
surface temperature is approximated to that of the atmospheric con-
dition and all the cells are at a temperature of 22 °C. The battery
module is also occupied with cooling air at the same temperature.
The contours of temperature developed with time are plotted in
Fig. 5. Differential surface temperature can be seen during the tem-
poral progression inside the battery module. It shows due to heat
generation within the cell, the battery surface temperature
increases and can be notably seen within 50 s past the discharge pro-
cess. This can be confirmed based on the different colors in contours
clearly indicating the temperature difference between battery sur-
faces and surrounding air. The air flowing through the battery mod-
ule absorbs the heat released by batteries and in a flow path become
warmer. The transient thermal response of battery cell is also con-
firmed by plotting cell temperature (Ts in °C) variation with time
(t in second) as shown in Fig. 6. The sample plots correspond to rep-
resentative cell (cell 5) which is centrally located in battery module.
5.2. Air temperature regimes
The temperature developed inside the battery module affect the
battery working condition. The cooling air circulated at lower supply
965
temperature carries away heat generated in the battery module. The
arrangement of battery cells, inter-cell spacing, state, and condition
of cooling media are the major factors responsible for developing
thermal regimes inside battery module. In the present paper, com-
plete temperature regimes and flow fields in the module are visual-
ized with the help of three-dimensional contours of temperature as
depicted in Fig. 5. Thermal gradients and localized heat spots can be
observed within the module. The local heat spots are detrimental to
safe working of the battery. For analyzing localized heat spots, three
planes were chosen at 25 mm (Fig. 7a), 50 mm (Fig. 7b) and 75 mm
(Fig. 7c) from battery base surface at discharge time 500 s and supply
air condition 22 °C and 0.1 m/s. It can be seen that air temperature
remains unaffected till it approaches first row of battery cells Also,
the zones adjacent to the two side wall of the module are noticed
unaffected by heat dissipation from cells. Local heat spots here refer
to the localized hot air temperature. As depicted in Fig. 7, local heat
spots are observed behind the cells. The quantitative measurement
of air temperature throughout the battery module helps in identify-
ing such data. The higher thermal gradients are observed in areas
behind the cells present in the second row as compared to first
row cells. The thermal gradient is seen to be varied with the depth
of the module; higher being near to the base surface.
5.2.1. Temperature profile in flow direction
The air temperature change (Ta,x
T1) within the module is
indicative of heat dissipation from battery surface to cooling
media. Here, Ta,x is the local air temperature measured in flow
direction and T1 is the air temperature at supply condition.
Fig. 8 show such plots of temperature regimes in the flow direction
at discharge time 500 s and supply air condition 22 °C and 0.1 m/s.
The module occupies relatively higher air temperature and experi-
ences up to 2 °C air temperature change. The air temperature
developed in the system has resulted in a thermal gradient across
the height of the module. The air temperature rises by 1.76 °C,
1.95 °C and 2.06 °C respectively at 25 mm, 50 mm and 75 mm
966 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

Fig. 9. Air temperature variation in transverse direction.

from the base surface of the module. Similar kind of temperature 5.2.2. Temperature profile in transverse direction
profile is expected in between cells 4 and 7 as the flow is symmet- The thermal regimes developed in the battery module also
rical in the flow direction. studied across the width of the module (transverse direction).
 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969 967

Fig. 10. Flow field in the module.

Results are analyzed considering two cases separately: tempera-
ture profile in spacing between row 1 cells & row 2 cells (case 1)
and spacing between row 2 cells & row 3 cells (case 2). Fig. 9 shows
such plots for case 1 at discharge time 500 s and supply air condi-
tion 22 °C and 0.1 m/s. The air temperature variations are very
peculiar in the transverse direction. Temperature increases from
supplied cooling condition (22 °C) to the local temperature max-
ima 26.06 °C for case 1 and 26.37 °C for case 2 at certain zones
and thus confirming the local heat spots as seen in Fig. 9. The study
on air temperature profile is further extended to include its varia-
tion at different depth of the module. The zones of local tempera-
ture maxima are in the middle of the module depth (50 mm from
the base surface of the module).
5.3. Flow fields in battery module
External flow over the battery cell is important for designing the
BTMS. Heat dissipated from cells is based on forced circulation. For
understanding the flow fields in the module, contours of velocity
are used at a different depth of the module as shown in Fig. 10a
968 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969

at discharge time 500 s and supply air condition 22 °C and 0.1 m/s. The maximum attainment of air velocity within the module is
The air velocity is zero at the battery surface based on the no-slip around 0.27 m/s for supply air velocity 0.1 m/s.
condition boundary condition as discussed in Section 4.4. The air
approaching the battery cell branches out at separation point 5.4. Cell-to-cell heat dissipation
(Fig. 10a) and encircles the cell. This resulted in the formation of
the boundary layer that wraps around the cell. The complete Temperature uniformity within the module is important for
reduction in air velocity at stagnation point raises the local pres- minimum long-term degradation [40]. The study on cell-to-cell
sure at that point. The pressure gets reduced in the flow direction heat dissipation within the module is important for avoiding local-
as a result of increased air velocity. It can be observed that air ized heat spots detrimental to the safe operating condition of a
velocity attains maximum value at the spacing between the cells. specific cell. The heat dissipation rates are found to be varying from
Air while flowing over the battery surface gets detached from the cell-to-cell (Fig. 12). As expected, it is observed that cells present in
surface (the boundary layer that wraps around the cell gets row 1 (cell: 1, 4, 7) are at relatively cooler conditions as they are
detached) and results in the formation of separation zone behind near to the airflow inlet. The cells present in row 2 (cell: 2, 5, 8)
the battery cell. In the separation zones, air recirculation and back- are dissipating the heat to the slightly warmer air. This is because
flows found to be varying across the depth of the module. The areas cooling media (air) gets heated as it flows across row 1 and absorbs
occupied by separated flow are lesser in Fig. 10 (The separated flow
regains faster and flows stream gets reattached with the main flow
stream). The extent of flow separation is different across the depth
of the battery cell due to differential flow velocities at the lower
flat bottom surface and spacing above the cells (Fig. 10b).
As seen in Fig. 10a, there is a symmetrical nature of flow in cell
spacing formed by cell 1, 4 and 7. Therefore airflow patterns in bat-
tery module are investigated on representative cell spacing formed
by cell 1 and cell 4 as illustrated in Fig. 11 at discharge time 500 s
and supply air condition 22 °C and 0.1 m/s. The air flow variations
are plotted in the flow direction. The air velocity increases rapidly
from supplied value to local maximum. The separation zones
formed behind the cell affects the continuity of airflow and reduces
air velocity in a flow path near to the separation zone. The flow
stream regains velocity while passing through cell spacing. Periodic
flow patterns are observed in the battery module. The observed
flow field is further extended to include it across the depth of the
cavity. The higher flow field is observed near the lower area of
the battery module surface although the similar periodic nature is
observed at the central and upper zone of the battery module. We
have provided around 21.2 mm spacing from the upper surface of
the cells for better circulation of cooling media (Fig. 2). The typical
Fig. 12. Cell-to-cell temperature variation in battery module at discharge time 500
flow nature is also seen at higher air velocity as shown in Fig. 10.
s and supply air condition 22 °C and 0.1 m/s.

Fig. 11. Air velocity variation in flow direction.

 R.D. Jilte, R. Kumar / Engineering Science and Technology, an International Journal 21 (2018) 957–969
heat from cells 1, 4, 7. Therefore cells at row 2 dissipate heat con-
vectively to the warmer air. The similar nature of heat exchange is
observed in cells present at row 3 (cell: 3, 6, 9). All the cells present
at the outlet ends are at a slightly higher temperature as compared
to their counterpart preceding cells. In the considered BTMS,
maximum cell-to-cell temperature non-uniformity is restricted to
0.11 °C and battery temperature is lesser than 28 °C despite the
high discharge rate and lower cooling air flow condition.
6. Conclusions
In the present study, the three-dimensional numerical investi-
gation has been carried out on the transient thermal behavior of
Li-ion battery module. A user-defined function (UDF) is used for
formulating heat generation in the battery cell. The results are
visualized in terms of the three–dimensional transient thermal
response of battery cell. In-depth analysis of localized heat spots
has been presented based on thermal regimes and flow field. The
following conclusions can be drawn:
 Different air temperature profiles are seen in the flow direction
and across the width and depth of the module.
 The air temperature variations are found in the transverse
direction. Temperature increases from supplied cooling condi-
tion (22 °C) to the local temperature maxima 26.37 °C at cer-
tain zones and thus confirming the local heat spots.
 The zones of local temperature maxima are in the middle of the
module depth.
 Typical flow patterns are observed in the battery module. The
extent of flow separation is different across the depth of the
battery.
 The battery cell temperature differs from each other and is a
function of position in the battery module. In the considered
BTMS, maximum cell-to-cell temperature non-uniformity is
restricted to 0.11 °C and battery temperature is lesser than
28 °C despite the high discharge rate and lower cooling air flow
condition.
References
[1] Z. Rao, S. Wang, A review of power battery thermal energy management,
Renew. Sustain. Energy Rev. 15 (9) (2011) 4554–4571.
[2] H. Liu, Z. Wei, W. He, J. Zhao, Thermal issues about Li-ion batteries and recent
progress in battery thermal management systems: a review, Energy Convers.
Manage. 150 (August) (2017) 304–330.
[3] P.H. Andersen, J.A. Mathews, M. Rask, Integrating private transport into
renewable energy policy: The strategy of creating intelligent recharging grids
for electric vehicles, Energy Policy 37 (7) (2009) 2481–2486.
[4] S. Panchal, S. Mathewson, R. Fraser, R. Culham, and M. Fowler, ‘‘Experimental
Measurements of Thermal Characteristics of LiFePO4 Battery,” no. April, 2015.
[5] S. Panchal, S. Mathewson, R. Fraser, R. Culham, M. Fowler, Thermal
management of lithium-ion pouch cell with indirect liquid cooling using
dual cold plates approach, SAE Int. J. Alt. Power. 4 (2) (2015) 293–307.
[6] Z. Rao, S. Wang, M. Wu, Z. Lin, F. Li, Experimental investigation on thermal
management of electric vehicle battery with heat pipe, Energy Convers.
Manage. 65 (2013) 92–97.
[7] N. Sato, Thermal behavior analysis of lithium-ion batteries for electric and
hybrid vehicles, J. Power Sources 99 (2001) 70–77.
[8] F. He and L. Ma, ‘‘Thermal Management in Hybrid Power Systems Using
Cylindrical and Prismatic Battery Cells,” Heat Transf. Eng., no. September,
2015.
[9] Z. Lei, Y. Zhang, X. Lei, Temperature uniformity of a heated lithium-ion battery
cell in cold climate, Appl. Therm. Eng. 129 (September) (2018) 148–154.
969
[10] C.G. Motloch, J.P. Christopheresen, J.P. Belt, R.B. Wright, G.L. Hunt, R.A. Sutula,
High-power battery testing procedures and analytical methodologies for
HEV’s, SAE Trans. 111 (2002) 797–802.
[11] S.S. Zhang, K. Xu, T.R. Jow, The low temperature performance of Li-ion
batteries, J. Power Sources 115 (1) (2003) 137–140.
[12] M.-S. Wu, P.-C.J. Chiang, High-rate capability of lithium-ion batteries after
storing at elevated temperature, Electrochim. Acta 52 (11) (2007) 3719–3725.
[13] S. C. Chen, C. C. Wan, and Y. Y. Wang, ‘‘Thermal analysis of lithium-ion
batteries,” vol. 140, no. May 2004, pp. 111–124, 2005.
[14] M.R. Giuliano, A.K. Prasad, S.G. Advani, Experimental study of an air-cooled
thermal management system for high capacity lithium-titanate batteries, J.
Power Sources 216 (2012) 345–352.
[15] F. He, X. Li, L. Ma, Combined experimental and numerical study of thermal
management of battery module consisting of multiple Li-ion cells, Int. J. Heat
Mass Transf. 72 (2014) 622–629.
[16] A. A. Pesaran, S. Burch, and M. Keyser, ‘‘An Approach for Designing Thermal
Management Systems for Electric and Hybrid Vehicle Battery Packs Preprint,”
no. January, 1999.
[17] R. Sabbah, R. Kizilel, and J. R. Selman, ‘‘Active (air-cooled) vs . passive (phase
change material) thermal management of high power lithium-ion packs :
Limitation of temperature rise and uniformity of temperature distribution,”
vol. 182, pp. 630–638, 2008.
[18] P. Nelson, D. Dees, K. Amine, G. Henriksen, Modeling thermal management of
lithium-ion PNGV batteries, J. Power Sources 110 (2) (2002) 349–356.
[19] S. Al Hallaj, H. Maleki, J.S. Hong, J.R. Selman, Thermal modeling and design
considerations of lithium-ion batteries, J. Power Sources 83 (1–2) (1999) 1–8.
[20] E. Barsoukov, J.H. Jang, H. Lee, Thermal impedance spectroscopy for Li-ion
batteries using heat-pulse response analysis, J. Power Sources 109 (2) (2002)
313–320.
[21] J. Hong, H. Maleki, S. Al-Hallaj, L. Redey, J.R. Selman, Electrochemical-
calorimetric studies of lithium-ion cells, J. Electrochem. Soc. 145 (5) (1998)
1489.
[22] A. Mills, S. Al-Hallaj, Simulation of passive thermal management system for
lithium-ion battery packs, J. Power Sources 141 (2) (2005) 307–315.
[23] R. Mahamud, C. Park, Reciprocating air flow for Li-ion battery thermal
management to improve temperature uniformity, J. Power Sources 196 (13)
(2011) 5685–5696.
[24] X. Li, F. He, L. Ma, Thermal management of cylindrical batteries investigated
using wind tunnel testing and computational fl uid dynamics simulation, J.
Power Sources 238 (2013) 395–402.
[25] Y. Chen, Three-dimensional thermal modeling of lithium-polymer batteries
under galvanostatic discharge and dynamic power profile, J. Electrochem. Soc.
141 (11) (1994) 2947.
[26] A. Pesaran, A. Vlahinos, and S. D. Burch, ‘‘Thermal Performance of EV and HEV
Battery Modules and Packs,” in: Fourteenth Int. Electr. Veh. Symp., no.
September, pp. 540–23527, 1997.
[27] H.J. Bergveld W.S. Kruijt P.H.L. Notten Battery management systems: design by
modelling 2002, 2002.
[28] David Linden and Thomas Reddy, Handbook of batteries. 2002.
[29] Y. A. Cengel, ‘‘Heat Transfer: A Practical Approach,” pp. 757–789, 2002.
[30] S.A. Khateeb, M.M. Farid, J.R. Selman, S. Al-Hallaj, Design and simulation of a
lithium-ion battery with a phase change material thermal management
system for an electric scooter, J. Power Sources 128 (2) (2004) 292–307.
[31] N. Javani, I. Dincer, G.F. Naterer, G.L. Rohrauer, Modeling of passive thermal
management for electric vehicle battery packs with PCM between cells, Appl.
Therm. Eng. 73 (1) (2014) 307–316.
[32] T. Wang, K.J. Tseng, J. Zhao, Z. Wei, Thermal investigation of lithium-ion
battery module with different cell arrangement structures and forced air-
cooling strategies, Appl. Energy 134 (2014) 229–238.
[33] X.M. Xu, R. He, Research on the heat dissipation performance of battery pack
based on forced air cooling, J. Power Sources 240 (2013) 33–41.
[34] S. Chacko, Y.M. Chung, Thermal modelling of Li-ion polymer battery for
electric vehicle drive cycles, J. Power Sources 213 (2012) 296–303.
[35] ‘‘UDF Manual,” Fluent 6.1, no. February, 2003.
[36] L. Jiji, Heat Conduction, Springer-Verlag, Berlin Heidelberg, 2009.
[37] A. Žukauskas, Heat transfer in tube banks in crossflow, Adv. Heat Transf. 18
(1987) 87.
[38] I. Voicu, H. Louahlia, H. Gualous, R. Gallay, Thermal management and forced
air-cooling of supercapacitors stack, Appl. Therm. Eng. 85 (2015) 89–99.
[39] 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. Eng. 85 (2015) 33–43.
[40] R. Kizilel, R. Sabbah, J.R. Selman, S. Al-Hallaj, An alternative cooling system to
enhance the safety of Li-ion battery packs, J. Power Sources 194 (2) (2009)
1105–1112.