2023, VOL. 7, NO:4, 340-348 www.ijastech.

org e-ISSN: 2587-0963

CFD Analysis of Sloshing in the Fuel Tank of a Heavy Vehicle with Emergency
Braking System

Turan Alp Arslan1*, Hüseyin Bayrakçeken1 and Hicri Yavuz2

0000-0003-3259-4854, 0000-0002-1572-4859, 0000-0001-8427-5164

1 Department of Automotive Engineering, Faculty of Technology, Afyon Kocatepe University, Afyon, 03200, Turkey.
2 Department of Engine Vehicles and Transportation Technology, Vocational School of Afyon, Afyon Kocatepe University, Afyon, 03200, Turkey.

Research Article
https://doi.org/10.30939/ijastech..1360466

Liquid sloshing, which occurs in all accelerating and liquid-carrying vehicles, is of great im-
portance, especially in the automotive and aerospace industries. Large-scale fluid sloshing causes
both operational and safety problems in vehicles. In this study, the fuel tank of a heavy vehicle
with an emergency braking system is designed in three dimensions, and liquid sloshing in the
fuel tank is investigated by CFD analysis method. VOF solution method, k-ԑ turbulence model,
and PISO solution algorithm are used in the study. In the analysis of liquid sloshing, it is assumed Received 14.09.2023
that the vehicle is traveling at a certain speed, decelerates and stops with emergency braking, and Revised 27.10.2023
remains stationary for a while. The braking scenario and boundary conditions are based on test Accepted 12.11.2023
data from a heavy vehicle manufacturer. The designed fuel tank with a capacity of 207.6 liters
was analyzed at 25%, 50%, and 60% diesel fuel filling levels in 6 different cases with and without
anti-slosh baffles. Four virtual sensors were placed on the side wall of the fuel tank in the direction
of vehicle movement, and time-dependent pressure changes were analyzed for all cases. In addi-
tion, the fuel volume ratio in all cases is visualized and presented for specific time steps. With
the use of anti-slosh baffles, the maximum pressure, the rate of pressure increase, and the liquid
sloshing were reduced by a factor of 2-3 for different cases. With the design of the fuel tank using * Corresponding author
anti-slosh baffles, instantaneous interruptions in the fuel system are prevented. Reducing the im- Turan Alp Arslan
pact pressures on the tank walls is expected to positively affect noise, vibration, and stability [email protected]
Address: Automotive Engineering De-
problems. partment, Faculty of Technology, Afyon
Kocatepe University, Afyon, Turkey
Keywords: CFD; Emergency braking system; Fuel tank; Sloshing; Tank baffles; VOF Tel:+902722182556

1. Introduction Sloshing occurs in all vehicles that move with acceleration and
The free surface motion or vibration of a fluid in a partially filled carry liquids. The study of this complex fluid-structure interaction
container stimulated by external forces is called sloshing [1-3]. is of great importance, especially in the automotive and aerospace
Sloshing is defined as a complex phenomenon involving nonlinear fields [11]. In addition to space, air, and sea vehicles, partially
properties of the fluid [4]. The free surface motion of the liquid in filled tanks of land vehicles such as cars, trucks, and buses are sub-
the tank depends on the shape and magnitude of the external forces, jected to severe sloshing due to acceleration. Liquid sloshing,
tank geometry, filling level, damping effect of the internal structure, which causes high impact pressures on the tank walls, can cause
and liquid properties [5]. These motions show different character- structural damage and destabilization of the vehicle carrying the
istics such as planar, non-planar, symmetric, asymmetric, turbulent, tank [12]. With inappropriate fuel tank designs, large-scale slosh-
random shock, quasi-periodic and chaotic [6-8]. Liquid sloshing ing occurs, causing sudden interruptions in the fuel system and in-
occurs especially intensively in automotive fuel tanks, liquid trans- creasing fuel vapor, which leads to hydrocarbon (HC) emissions in
portation containers, sea, air, and space vehicles, petrochemical gasoline vehicles. In addition, the amount of noise and vibration
storage, seismically stimulated storage areas, dams, and nuclear transmitted to the passenger compartment increases with the slosh-
power plants and causes both operational and safety problems in ing that occurs during sudden acceleration, sudden deceleration,
all these areas [9,10].
340
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

cornering, and driving on rough roads [13]. Various design solu- was analyzed at 25%, 50%, and 60% diesel fuel filling levels, with
tions are put forward to prevent all these problems. The most com- nd without anti-slosh baffles, in 6 different cases. Time-dependent
mon of these is the addition of different geometry and number of pressure changes were obtained for all cases with 4 virtual sensors
anti-slosh baffles in the tank [14]. However, using anti-slosh baf- positioned on the side wall of the tank in the direction of vehicle
fles, which minimize the free surface movement and the adverse movement. In addition, the fuel volume ratio in all cases was vis-
effects caused by dividing the fuel tank into smaller volumes, in- ualized and analyzed for specific time steps. In the study, it is
creases the tank weight and production costs. aimed to minimize the impact pressures, noise, and vibration on
Thanks to the advantages it offers, the finite element method is the tank walls and to prevent instantaneous interruptions in the fuel
frequently used in the design of vehicle and machine parts [15-18]. system by minimizing the sloshing phenomenon with the use of
anti-slosh baffles.
Nowadays, both anti-slosh and weight and cost reduction studies
in fuel tank design are carried out by computational fluid dynamics
2. Materials and Methods
(CFD) method using special software [19,20]. With the CFD
method, all properties and parameters such as acceleration, force, The CFD analysis of the sloshing phenomenon in a heavy vehi-
flow regime, liquid properties, tank geometry, filling level, anti- cle fuel tank with an EBS system has been carried out in Ansys
slosh baffle geometry, number, and location can be tested in dif- Fluent software. The simulation study was completed in four
ferent combinations to determine the optimum design. Unlike ex- stages. In the first stage, the fuel tanks of the heavy vehicle were
perimental studies, these analyses are reproducible, cost-effective, designed in three dimensions, and filling levels and fuel properties
and time-saving [21-23]. Various numerical methods are used for were determined. In the second stage, the mesh structures of the
the sloshing analysis of fuel tanks by the CFD method. The com- tank models were created, and the adequacy of these structures was
mon ones are explicit finite element, arbitrary Lagrangian-Eulerian verified with the literature. In the third stage, the braking scenario
(ALE), hybrid finite element, smoothed particle hydrodynamics and boundary conditions are defined using the test data taken as
reference. In the last stage, analysis methods and solution algo-
(SPH), and volume of fluid (VOF) [24,25]. Developed by Hirt and
rithms suitable for the specified purpose are applied, and the results
Nichols [26] to study the complex free surface motion of incom-
are evaluated.
pressible fluids, VOF is a robust and flexible method compared to
other numerical methods [27]. It is seen that CFD analysis man-
2.1 Governing Equations
agement and the VOF solution method are widely used in the lit-
erature for the simulation of the sloshing of two or more types of A mathematical model is created using partial differential equa-
immiscible fluids in a tank [28-33]. tions, integral equations, and boundary conditions in CFD applica-
Liquid sloshing can occur in the form of laminar or turbulent tions. The performed sloshing analysis provides solutions using the
flow. Numerical methods for solving turbulence problems are volume fraction equation, the momentum equation, and the stand-
based on potential flow theory or Navier Stokes equations [34]. No ard k-ԑ turbulence model. The volume fraction equation, also
clear consensus exists on whether turbulence models should be called the continuity transport equation, is given by Eq. (1).
used for fluid turbulence [35]. Potential flow theory and some an- 1 𝜕 𝑛
alytical methods can be used for laminar flow owing to their rela- [ (𝛼𝑖 𝜌𝑖 ) + ∇(𝛼𝑖 𝜌𝑖 𝑣⃗𝑖 )] = 𝑆𝛼𝑖 + ∑ (𝑚̇𝑝𝑖 − 𝑚̇𝑖𝑝 ) (1)
𝜌𝑖 𝜕𝑡 𝑝=1
tively stable solutions and low computational effort [36]. However,
while acceptable results are obtained from simulation studies with- Where 𝛼, 𝜌, and 𝑣⃗ represent the volume fraction, density, and
out turbulence models, it has been proved that using turbulence velocity for phase i, respectively. 𝑚̇𝑥𝑦 is the mass transfer from
models improves solution accuracy and efficiency [37]. Therefore, phase x to phase y. The momentum equation is shown in Eq. (2).
the inclusion of an appropriate turbulence model in the liquid
sloshing problem is important for simulation success. The standard 𝜕𝑣
𝜌( + 𝑣(∇𝑣)) = −∇𝑃 + ∇𝜏 + 𝜌𝑔 (2)
k- ԑ turbulence model, the RNG k- ԑ turbulence model, the realiz- 𝜕𝑡
able k- ԑ turbulence model, and the k-ω SST turbulence model are
In this equation, 𝑡 is time, ∇ is divergence, 𝑃 is pressure, 𝜏
the turbulence models commonly used in the analysis of fluid tur-
is stress, and 𝑔 is gravitational acceleration. Finally, the turbu-
bulence [38]. The standard k- ԑ turbulence model stands out with
lence kinetic energy (k) and its dissipation rate (ԑ), which form the
its high prediction accuracy and fast convergence in studies of im-
k-ԑ turbulence model, are given by Eq. (3) and Eq. (4), respectively.
pact pressure. The use of the k- ω SST turbulence model is recom-
mended for studies of free surface motion owing to its high accu- 𝜕 𝜕 𝜕 𝜇𝑡 𝜕𝑘
(𝜌𝑘) + (𝜌𝑘𝑢𝑖 ) = [(𝜇 + ) ]…
racy [39]. 𝜕𝑡 𝜕𝑥𝑖 𝜕𝑥𝑖 𝜎𝑘 𝜕𝑥𝑗 (3)
In this study, the sloshing phenomenon in the fuel tank of a … + 𝐺𝑘 + 𝐺𝑏 − 𝜌𝜀 − 𝑌𝑀 + 𝑆𝑘
heavy vehicle with an emergency braking system (EBS) is inves-
tigated by CFD analysis. VOF solution method and k-ԑ turbulence 𝜕 𝜕 𝜕 𝜇𝑡 𝜕𝜀
model are used in the study. In order to monitor the fluid sloshing, (𝜌𝜀) + (𝜌𝜀𝑢𝑖 ) = [(𝜇 + ) ]…
𝜕𝑡 𝜕𝑥𝑖 𝜕𝑥𝑖 𝜎𝜀 𝜕𝑥𝑗
it is assumed that the vehicle is traveling at a certain speed, decel- (4)
erates and stops with emergency braking, and remains stationary 𝜀 𝜀2
… + 𝐺1𝜀 (𝐺𝑘 + 𝐶3𝜀 𝐺𝑏 ) − 𝐺2𝜀 𝜌 + 𝑆𝜀
for a while. The designed fuel tank with a capacity of 207.6 liters 𝑘 𝑘
341
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Where 𝜇 is the viscosity, 𝜇𝑡 is the turbulent viscosity, 𝜎𝑥 is 50%, and 60% filling levels, with and without anti-slosh baffles, in
the turbulence Prandtl number for x, 𝐺𝑘 is the turbulent kinetic 6 different cases. In these analyses, diesel fuel with a cetane num-
energy due to mean velocity gradients, 𝐺𝑏 is the turbulent kinetic ber of 45, selected from the Ansys library, was preferred. Detailed
energy due to buoyancy, 𝑌𝑀 is the contribution of the fluctuating specifications of the diesel fuel used are presented in Table 1.
dilatation in compressible turbulence to the overall dissipation rate,
𝑆𝑥 is the user-defined source for x, 𝐺1𝜀 , 𝐺2𝜀 , and 𝐶3𝜀 are the Table 1. Fuel properties.
model constants. Fuel Properties (Units) Values
Fuel type Diesel
2.2 Fuel Tank Model and Fuel Properties
Cetane number 45
In the first stage of the study, the fuel tank planned for use in a Density (kg/m3) 846
heavy vehicle, is designed in two different types, with and without Viscosity (kg/m-s) 0.0029
baffles. Solidworks software was used in the three-dimensional de-
Molecular weight (kg/kmol) 180.33
sign phase. The designed fuel tanks are shaped like a rectangular
prism, common in today's heavy vehicles, and have a fuel capacity
of 207.6 liters. It is seen that this capacity is sufficient for a pickup 2.3 Meshing Strategy and Analysis Methods
truck or midi bus. The tanks are equipped with a filling cap, fuel Mesh structure creation on the fuel tank models was performed
transfer pipe, float system, drain valve, and connection brackets. with the Ansys software Mesh module. In order to minimize the
The designed fuel tanks are 1200 mm long, 350 mm high, and 510 deviations in the results in both fuel tank models, a mesh structure
mm wide. Inside the tank, there are two anti-slosh baffles with a with equal properties was used. In this study, where liquid surface
height of 250 mm and a thickness of 3 mm. These baffles, which motion is mainly investigated, tetrahedral elements with 10 nodal
provide fuel flow between the volumes with the help of holes at points, which provide highly accurate solutions in surface contours,
the bottom, are placed symmetrically in the tank at equal distances are preferred. Although the solution time increases slightly with
from the side walls. Some geometric simplifications were made on the use of tetrahedral elements with a large number of nodes, the
both types of fuel tank models in order to shorten the solution time quality, and accuracy of the results obtained improve [40]. Geo-
and improve the quality of the mesh structure. metric changes that occur in the analysis of systems with dynamic
structure cause the quality of the mesh structure to decrease over
time, leading to a decrease in solution accuracy. In order to prevent
these degradations and to keep the mesh quality at a high level dur-
ing the whole analysis process, the adaptive mesh method is used.
The adaptive mesh structure performs 7 resolutions to rebuild the
mesh structure from coarse to fine. Table 2 shows the number of
elements and nodes of the mesh structures used in all filling levels
of the fuel tank models. In the fuel tank model with anti-slosh baf-
fles, the number of elements and nodes increased due to the rela-
tively small hole details at the bottom of the sheet metal baffles.

Table 2. Mesh details.

Element Number of Number of

Fuel Tank Models
Type Elements Nodes
Without baffle 65876 339618
Tetrahedral
With baffle 91586 483032

The quality of the mesh structure is directly related to the distri-

bution and uniformity of the elements. Significant dimensional dif-
ferences and shape changes between adjacent elements impair so-
lution stability [41]. After creating the mesh structure of the fuel
tank models, the quality parameters of the mesh structure were
checked, and the validity of this structure was verified with the lit-
Fig. 1. Fuel tank model dimensions and virtual sensor locations erature data. Mesh structure quality parameters, current values,
recommended limits, and adequacy of the mesh structure are pre-
Four virtual sensors coded S1, S2, S3, and S4 were positioned sented in Table 3. In the study where three different quality param-
on the side wall of the tanks in the direction of vehicle motion to eters, namely aspect ratio, orthogonality, and skewness, were con-
investigate the time-dependent pressure variations. The designed sidered, it was determined that the mesh structures created for both
fuel tank model, basic dimensions, and sensor locations are shown tank models were suitable and acceptable for the analysis solution.
in Figure 1. Analyses were carried out on the fuel tanks at 25%,
342
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Table 3. Mesh quality parameters. The reaction time is neglected since the analysis starts with brak-
ing, and acceleration will act on the tank from the start of braking.
Models Parameters Values Recommended Adequacy As in the reference test, the vehicle is assumed to brake and stop
Max. aspect ratio 10.431 <40 Good with minimum slip and maximum efficiency on dry asphalt. The
Without
Min. orthogonality 0.172 >0.15 Acceptable braking scenario and boundary conditions are presented in Table
baffle
Max. skewness 0.827 <0.95 Acceptable 4. The horizontal braking acceleration of 6.1716 m/s2 and vertical
Max. aspect ratio 15.02 <40 Good gravity acceleration of 9.81 m/s2 obtained from the equations are
With applied on the fuel tank. In the analyses performed, the vehicle
Min. orthogonality 0.153 >0.15 Acceptable
baffle brakes for 3.6 seconds and stops and remains stationary for 4.1 sec-
Max. skewness 0.846 <0.95 Acceptable onds for the damping of the sloshing phenomenon. The velocity-
time graph of the heavy vehicle is shown in Figure 2.
2.4 Bounding Conditions and Analysis Setup Fuel tank sloshing analyses were performed in the transient re-
The heavy vehicle in which the fuel tank sloshing event is ana- gime in Ansys Fluent software. In the analysis setup, the VOF
lyzed is traveling at 80 km/h, braking suddenly with EBS and stop- method and the k-ԑ turbulence model were used to analyze the fluid
ping at 40 m. In order to simulate the braking scenario and the surface motions, which provide stable and highly accurate results
sloshing event as closely as possible, the speed and stopping dis- compared to other methods. Although there are many different ap-
tance values are taken from the test data of a heavy vehicle manu- proaches to modeling flow motions in CFD analysis studies, the k-
facturer. Eq. (5) calculates stopping distance using vehicle speed ε turbulence model stands out in solving engineering problems
and stopping time, while Eq. (6) calculates stopping distance using with its practical use and high-accuracy results [43,44]. The PISO
vehicle speed and braking acceleration [42]. algorithm, which is recommended for analysis in the transient re-
gime and where mesh structure distortions are intensively ob-
𝑉0 𝑡𝑠 served, is chosen as a solution method. With this algorithm, stable
𝑆 = 𝑉0 𝑡𝑟 + (5) solutions are obtained in large time steps, and the computational
2
effort is reduced [45]. In the study, each sloshing analysis was gen-
erated with 110 time steps with a magnitude of 0.07 seconds, cov-
𝑉0 2
𝑆 = 𝑉0 𝑡𝑟 + (6) ering a total period of 7.7 seconds. The fuel tank sloshing analysis
2𝑎𝑏 methods are presented in detail in Table 5.
Where 𝑆 is stopping distance, 𝑉0 is vehicle speed, 𝑡𝑟 is reac- Table 5. Sloshing analysis methods.
tion time, 𝑎𝑏 is braking acceleration, and 𝑡𝑠 is stopping time.
Analysis Methods (Units) Values
Table 4. Bounding conditions.
Solver type Pressure-Based
Bounding Conditions (Units) Values Time Transient
Assist braking system EBS Model Multiphase, VOF
Initial velocity (km/h) 80 Viscosity k-ԑ (standart)
Stopping distance (m) 40 Solution method PISO
Braking acceleration (m/s2) 6.1716 Materials Air and Diesel
Braking duration (s) 3.6 Analysis duration (s) 7.7
Time step size (s) 0.07
Number of time steps 110
Maximum iterations / Time step 20

3. Results and Discussion

As a result of the sloshing analyses, time-dependent pressure
values were recorded from virtual sensors placed on the side wall
of the fuel tank in the direction of vehicle movement for all cases.
Figure 3 shows the pressure changes in the tank designs with and
without anti-slosh baffles for 25% diesel fuel filling level. From
the start of the heavy vehicle braking with EBS, the pressure on the
fuel tank wall without baffles increased sharply. At 0.525 seconds,
the pressure on the sensors reached its maximum value. At this
moment, the maximum pressure measured at the sensor coded S1
was 4252.92 Pascal. Afterward, the pressure decreased slightly de-
pending on the braking acceleration and the amount of sloshing
and fluctuated and remained at high values until the end of braking
Fig. 2. Velocity vs time graph of the vehicle
343
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

at 3.6 seconds. As the braking ended and the vehicle came to a design with and without anti-slosh baffles. In the tank model with-
standstill, the diesel fuel moved towards the opposite wall of the out the anti-slosh baffle, a sudden accumulation of fuel on the tank
tank, causing a sudden decrease in the pressure. However, in this wall and a rapid pressure increase occurred upon braking. At 0.42
tank, where there were no anti-slosh baffles, the pressure measured seconds, the pressure values reached a maximum and were meas-
from the sensors continued fluctuating at high values throughout ured as 6120.43 Pascal on the sensor coded S1. Then, with contin-
the analysis due to the large-scale sloshing effect. In the fuel tank ued braking, the pressure decreased slightly and remained at high
where anti-slosh baffles were used, there was a significant im- values until 3.6 seconds in a partially fluctuating manner. With the
provement in the rate of pressure rise and maximum pressure. In end of braking, a rapid pressure drop followed by a high amount
this tank, the initial peak pressure measured from the sensor coded of sloshing was observed. At this filling level, severe sloshing per-
S1 was 1508 Pascal at 0.7 seconds. Until the end of the braking sisted even at the end of the analysis at 7.7 seconds. In the fuel tank
event, the pressure in this tank increased gradually due to the fuel where the anti-slosh baffle was used, there was no sudden increase
passage between the volumes. However, it is clear from the graphs in pressure with increasing filling level as in the 25% filled tank.
that this increase is balanced, and the sloshing is minimized. At the A reasonably balanced pressure increase and fuel passage between
end of braking at 3.6 seconds, the pressure on the wall where the the volumes occurred at this design and fill level. Until approxi-
sensors were positioned decreased sharply with the reverse move- mately 3.22 seconds at the end of braking, the fuel smoothly tran-
ment of the fuel. Then, until the end of the analysis, small sloshing sitioned to the volume in the direction of travel and gradually in-
events were damped, and the fuel in the tank stabilized. creased the pressure. At this moment, the maximum pressure was
measured as 3000.43 Pascal via the sensor coded S1. With the end
of the braking, the fuel moving to the opposite volumes in the tank
decreases the pressure on the sensors. Again, it is seen that the
sloshing is minimized until the end of the analysis with the steady
flow of fuel over time.

Fig. 3. The pressure history curves of sloshing for a) 25% fill level with-
out baffle and b) 25% fill level with baffle

Figure 4 shows the pressure changes in the tank designs with

and without anti-slosh baffles for 50% diesel fuel filling level.
When the graphs are analyzed, it is seen that with increasing filling Fig. 4. The pressure history curves of sloshing for a) 50% fill level with-
level, the liquid movement in the tank is more stable for both the out baffle and b) 50% fill level with baffle
344
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Figure 5 shows the pressure changes in the tank designs with

and without anti-slosh baffles for 60% diesel fuel filling level. In
the tank design without anti-slosh baffles, a sudden fuel pressure
hits the tank wall, and the pressure partially fluctuates until 3.6 sec-
onds and remains at high values. Here, the maximum pressure was
measured as 6229.75 Pascal at 0.35 seconds via the sensor coded
S1. With the increase in the filling level, the sloshing decreased
slightly, and the stability of the fluid movements increased. After
braking, when the vehicle became stationary, a high rate of slosh-
ing occurred, and the fluid stability decreased. In the fuel tank de-
sign with anti-slosh baffles, there are improvements in the pressure
increase and the rate of pressure increase. The maximum pressure
reached 3729.75 Pascal in 2.1 seconds on the sensor coded S1 and
remained approximately at this value until the end of braking. Af-
ter braking, minimal fluid movements and sloshing with minimal
irregularity were observed. Until the end of the analysis, the pres-
sure gradually decreased slightly with regular fuel flow between
the volumes. It was determined that the pressure changes measured
over the sensors coded S1, S2, S3, and S4 were in similar forms
and that the amount of pressure increased only with the increase in
height due to the sensor positions. It is seen that the use of anti-
slosh baffles has significant effects on impact pressure and liquid
sloshing in all 25%, 50%, and 60% diesel fuel filling levels. The
maximum pressure, rate of pressure increase, and liquid sloshing
decreased by about 2-3 times for different cases using anti-slosh
baffles. It is thought that this will prevent damage to the tank walls,
noise, vibration, instability on the vehicle carrying the tank, fuel
system interruptions, and increase in HC emissions in gasoline ve-
hicles.
The fuel volume fraction results for all diesel fuel fill levels and
fuel tank designs are shown in Figure 6, where the liquid sloshing
is visualized and presented. The results cover the entire analysis
period of 7.7 seconds with 0.7 second time step intervals. The fluid Fig. 5. The pressure history curves of sloshing for a) 60% fill level with-
movements seen in the figure support the pressure changes meas- out baffle and b) 60% fill level with baffle
ured by the virtual sensors. In tank designs without anti-slosh baf-
fles, the braking acceleration acting with the start of the analysis 4. Conclusions
causes large-scale liquid sloshing on the fuel. Although the amount
of sloshing decreases with the increase in the filling level, the fuel In this study, the fuel tanks of a heavy vehicle with an EBS brak-
is completely withdrawn from the tank center as all the fuel accu- ing system were analyzed in 6 cases with and without anti-slosh
mulates on the side wall. With this situation, interruptions in the baffles at 25%, 50%, and 60% diesel fuel filling levels. Using anti-
fuel system become inevitable. With the end of braking, the fuel slosh baffles reduced the high impact pressures, sudden pressure
suddenly hits the opposite wall again. Fuel movement with sudden rises, and large-scale liquid sloshing on the fuel tank walls. This
acceleration changes at the beginning and end of braking can cause situation positively affects noise and vibration generation, tank
damage, noise, and vibration on the tank walls. Liquid sloshing is damage, and stability disturbances. The anti-slosh baffles pre-
minimized in tank designs where anti-slosh baffles are used. The vented high accelerations acting on the tank, pulling fuel from the
division of the fuel tank into three separate volumes with baffles tank bottom and interrupting the fuel system. This design is also
reduces the sudden pressure increase and maximum pressure on thought to reduce HC formation in gasoline vehicles. In future
the side walls. Fuel passage through the holes at the bottom of the studies, it is aimed to minimize liquid sloshing, tank weight, and
baffles is smooth and gradual. This situation ensures that there is cost by changing the fuel type, fuel tank design, number, location,
always some fuel in the tank bottom and prevents interruptions in and geometry of the anti-slosh baffles.
the system. In addition, the decrease in pressure rise is expected to
prevent damage to the tank walls, noise, and vibration. Consider-
ing the fuel flow, pressure values, amount of sloshing, production
costs and fuel tank weight, the partitioned tank design is consid-
ered sufficient for a heavy vehicle moving with low accelerations.

345
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Fig. 6. Fuel volume fraction at various time steps for a) 25% fill level without baffle, b) 25% fill level with baffle, c) 50% fill level without baffle, d)
50% fill level with baffle, e) 60% fill level without baffle, and f) 60% fill level with baffle

346
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Nomenclature References
ALE : Arbitrary Lagrangian-Eulerian [1] Zhang E, Zhu W, Wang L. Influencing Analysis of Different Baffle
CFD : Computational Fluid Dynamics Factors on Oil Liquid Sloshing in Automobile Fuel Tank. Proceed-
EBS : Emergency Braking System ings of the Institution of Mechanical Engineers, Part D: Journal of
FVF : Fuel Volume Fraction Automobile Engineering. 2020; 234(13): 3180-3193.
HC : Hydrocarbon [2] He R, Zhang E, Fan B. Numerical Analysis on the Sloshing of Free
Oil Liquid Surface Under the Variable Conditions of Vehicle. Ad-
PISO : Pressure Implicit with Split Operator
vances in Mechanical Engineering. 2019; 11(2): 1-13.
RNG : Renormalization Group
[3] Topçu EE, Kılıç E. A Numerical Investigation of Sloshing in a 3D
SPH : Smoothed Particle Hydrodynamics
Prismatic Tank with Various Baffle Types, Filling Rates, Input Am-
SST : Shear Stress Transport plitudes and Liquid Materials. Applied Sciences. 2023; 13(4): 2474.
VOF : Volume of Fluid [4] Al-Zughaibi AI, Hussein EQ, Rashid FL. Studying and Analysis of
𝒂𝒃 : Braking acceleration Nonlinear Sloshing (Vibrating) of Interaction Fluid Structure in
𝜶 : Volume fraction Storage Tanks. Journal of Mechanical Engineering Research and
ԑ : Turbulence dissipation rate Developments. 2021; 44(7): 180-191.
𝒈 : Gravitational acceleration [5] Akyıldız H, Ünal NE. Sloshing in a Three-Dimensional Rectangular
𝑮𝒃 : Turbulent kinetic energies due to the buoyancy Tank: Numerical Simulation and Experimental Validation. Ocean
: Turbulent kinetic energies due to the mean veloc- Engineering. 2006; 33(16): 2135-2149.
𝑮𝒌 [6] Topçu EE, Kılıç E, Çavdar K, Kuzucu İ. Investigation of Sloshing
ity gradients
𝑮𝟏𝜺 , 𝑮𝟐𝜺 , in a Prismatic Tank Filled with Different Liquid Level. The Online
: Model constants Journal of Science and Technology. 2017; 7(4): 115-120.
𝑪𝟑𝜺
k : Turbulent kinetic energy [7] Ibrahim RA. Liquid Sloshing Dynamics: Theory and Applications.
𝒎̇𝒙𝒚 : Mass transfer from phase x to phase y Cambridge University Press: New York, USA, 2005.
𝝁 : Viscosity [8] Akyıldız H, Ünal NE, Aksoy H. An Experimental Investigation of
the Effects of the Ring Baffles on Liquid Sloshing in a Rigid Cylin-
𝝁𝒕 : Turbulent viscosity
drical Tank. Ocean Engineering. 2013; 59: 190-197.
𝑷 : Pressure
[9] Ingle WP, Nalawade D, Jagadale M. Comparative Study of Sloshing
𝝆 : Density
Phenomenon in a Tank Using CFD. International Engineering Re-
𝑺 : Stopping distance
search Journal. 2017; SE PGCON-MECH-2017: 1-5.
𝑺𝒙 : User-defined source for x [10] Mahmud MA, Khan RI, Wang S, Xu Q. Comprehensive Study on
𝝈𝒙 : Turbulent Prandtl numbers for x Sloshing Impacts for an Offshore 3D Vessel via the Integration of
𝒕 : Time Computational Fluid Dynamics Simulation, Experimental Unit, and
𝒕𝒓 : Reaction time Artificial Neural Network Prediction. Industrial & Engineering
𝒕𝒔 : Stopping time Chemistry Research. 2020; 59(51): 22187–22204.
𝝉 : Internal stress forces [11] Eswaran M, Saha UK, Maity D. Effect of Baffles on a Partially
⃗⃗
𝒗 : Velocity Filled Cubic Tank: Numerical Simulation and Experimental Valida-
𝑽𝟎 : Vehicle speed tion. Computers & Structures. 2009; 87(3-4): 198-205.
ω : Turbulence dissipation rate [12] Rajagounder R, Mohanasundaram GV, Kalakkath P. A Study of
: Contribution of the fluctuating dilatation in com- Liquid Sloshing in an Automotive Fuel Tank under Uniform Accel-
𝒀𝑴 eration. Engineering Journal. 2016; 20(1): 71-85.
pressible turbulence to the overall dissipation rate
𝛁 : Divergence [13] Singal V, Bajaj J, Awalgaonkar N, Tibdewal S. CFD Analysis of a
Kerosene Fuel Tank to Reduce Liquid Sloshing. Procedia Engineer-
Conflict of Interest Statement ing. 2014; 69: 1365-1371.
[14] Hasheminejad SM, Mohammadi MM. Effect of Anti-Slosh Baffles
The authors declare that there is no conflict of interest in the on Free Liquid Oscillations in Partially Filled Horizontal Circular
study. Tanks. Ocean Engineering. 2011; 38(1): 49-62.
[15] Duran S, Coşkun Y, Kılıç D, Uyumaz A, Zengin B. Determination
CRediT Author Statement
of the Load Applied to Bale Wrapping Machine Rotary Arm and
Turan Alp Arslan: Software, Methodology, Writing-original Performing of Design Optimization. Engineering Perspective. 2021;
draft, Conceptualization, Hüseyin Bayrakçeken: Conceptual- 1(4): 110-114.
ization, Supervision, Hicri Yavuz: Writing-original draft, Vis- [16] Binboğa F, Şimşek HE. Design and Optimization of a Semi-Trailer
ualization Extendable RUPD According to UNECE R58. Engineering Per-
spective. 2022; 2(2): 13-20.
[17] Karaman M, Öztürk E. Analysis of the Behavior of a Cross-Type
Hydraulic Outrigger and Stabilizer Operating Under Determined
347
Arslan et al. / International Journal of Automotive Science and Technology 7 (4): 340-348, 2023

Loads. Engineering Perspective. 2021; 1(1): 22-29. [33] Sances D, Gangadharan S, Sudermann J, Marsell B. CFD Fuel Slosh
[18] Theogene N, Mathieu N, Gebre A. Optimization of Concrete Sleep- Modeling of Fluid-Structure Interaction in Spacecraft Propellant
ers Subjected to Static and Impact Loadings. Engineering Perspec- Tanks with Diaphragms. 51st AIAA/ASME/ASCE/AHS/ASC
tive. 2021; 1(3): 92-98. Structures, Structural Dynamics, and Materials Conference. Flor-
[19] Micheli GB, Fogal MLF, Scalon VL, Padilha A, Ismail KAR. Anal- ida/USA, April 2010.
ysis and Reduction of the Sloshing Phenomena Due to Sudden [34] Zhang C. Application of an Improved Semi–Lagrangian Procedure
Movement of Spray Mixture Tanker. Journal of Applied Fluid Me- to Fully–Nonlinear Simulation of Sloshing in Non–Wall–Sided
chanics. 2022; 15(2): 399-413. Tanks. Applied Ocean Research. 2015; 51: 74-92.
[20] Scurtu IL, Gheres M, Jurco AN. Sloshing Simulation of the Liquid [35] Mahfoze OA, Liu W, Longshaw SM, Skillen A, Emerson DR. On
From the Truck Tanker. International Symposium, ISB-INMA TEH' the Efficacy of Turbulence Modelling for Sloshing. Applied Sci-
2018, Agricultural and Mechanical Engineering. Bucharest/Roma- ences. 2022; 12(17): 8851.
nia, November 2018. [36] Liu D, Tang W, Wang J, Xue H, Wang K. Comparison of Laminar
[21] Arslan TA, Solmaz H, İpci D. Rhombic Mekanizmalı Beta Tipi Bir Model, RANS, LES and VLES for Simulation of Liquid Sloshing.
Stirling Motorunun Adyabatik Şartlarda CFD Analizi. 2nd Interna- Applied Ocean Research. 2016; 59: 638-649.
tional Symposium on Automotive Science and Technology. An- [37] Brown SA, Greaves DM, Magar V, Conley DC. Evaluation of Tur-
kara/Turkey, September 2021. bulence Closure Models Under Spilling and Plunging Breakers in
[22] Arslan TA, Solmaz H, İpci D, Aksoy F. Investigation of the Effect the Surf Zone. Coastal Engineering. 2016; 114: 177–193.
of Compression Ratio on Performance of a Beta Type Stirling En- [38]Tahmasebi MK, Shamsoddini R., Abolpour B. Performances of Dif-
gine with Rhombic Mechanism by CFD Analysis. Environmental ferent Turbulence Models for Simulating Shallow Water Sloshing in
Progress & Sustainable Energy. 2023; 42(4): e14076. Rectangular Tank. Journal of Marine Science and Application. 2020;
[23] Halis S, Solmaz H, Polat S, Yücesu H. Numerical Study of the Ef- 19(3): 381–387.
fects of Lambda and Injection Timing on RCCI Combustion Mode. [39]Ha M, Kim D, Choi HI, Cheong C, Kwon SH. Numerical and Exper-
International Journal of Automotive Science and Technology. 2022; imental Investigations into Liquid Sloshing in a Rectangular Tank.
6(2): 120-126. The 2012 World Congress on Advances in Civil, Environmental, and
[24] Nicolici S, Bilegan RM. Fluid Structure Interaction Modeling of Materials Research (ACEM’ 12). Seoul/ Korea, August 2012
Liquid Sloshing Phenomena in Flexible Tanks. Nuclear Engineering [40] Kadkhodapour J, Schmauder S, Sajadi F. Quality Analysis of Addi-
and Design. 2013; 258: 51-56. tively Manufactured Metals. Elsevier, 2022.
[25] Calderon-Sanchez J, Gonzalez LM, Marrone S, Colagrossi A, Gam- [41] Arslan TA. Rhombic Mekanizmalı Beta Tipi Bir Stirling Motorunda
bioli F. A SPH Simulation of the Sloshing Phenomenon Inside Fuel Sıkıştırma Oranının Motor Performansına Etkilerinin Sayısal Olarak
Tanks of the Aircraft Wings. 14th International SPHERIC Work- İncelenmesi. Master Thesis, Gazi University, Ankara, Turkey, 2021.
shop. Exeter/United Kingdom, June 2019. [42] Bayrakçeken H, Düzgün M. Vehicles Brake Efficiency and Braking
[26] Hirt CW, Nichols BD. Volume of Fluid (VOF) Method for the Dy- Distance Analysis. Journal of Polytechnic. 2005; 8(2): 153-160.
namics of Free Boundaries. Journal of Computational Physics. 1981; [43] Latif H, Hultmark G, Rahmana S, Maccarini A, Afshari A. Perfor-
39(1): 201-225. mance Evaluation of Active Chilled Beam System for Office Build-
[27] Hosain ML, Sand U, Fdhila RB. Numerical Investigation of Liquid ings - A Literature Review. Sustainable Energy Technologies and
Sloshing in Carrier Ship Fuel Tanks. IFAC-PapersOnLine. 2018; Assessments. 2022; 52: 101999.
51(2): 583-588. [44] Ermiş K, Çalışkan M, Okan A. The Effects of Surface Roughness
[28] Rebollo XV, Sadeghi E, Kusano I, Garcia-Granada A-A. Study of on the Aerodynamic Drag Coefficient of Vehicles. International
the Sloshing Dynamics in Partially Filled Rectangular Tanks with Journal of Automotive Science and Technology. 2022; 6(2): 189-
Submerged Baffles Using VOF and LES Turbulence Methods for 195.
Different Impact Angles. Computation. 2022; 10(12): 225. [45] Seif MS, Asnaghi A, Jahanbakhsh E. Implementation of PISO Al-
[29] Wei G, Zhang J. Numerical Study of the Filling Process of a Liquid gorithm for Simulating Unsteady Cavitating Flows. Ocean Engi-
Hydrogen Storage Tank Under Different Sloshing Conditions. Pro- neering. 2010; 37: 1321-1336.
cesses. 2020; 8(9): 1020.
[30] Wright MD, Gambioli F, Malan AG. CFD Based Non-Dimensional
Characterization of Energy Dissipation Due to Verticle Slosh. Ap-
plied Sciences. 2021; 11(21): 10401.
[31] Sanapala VS, Rajkumar M, Velusamy K, Patnaik BSV. Numerical
Simulation of Parametric Liquid Sloshing in a Horizontally Baffled
Rectangular Container. Journal of Fluids and Structures. 2018; 76:
229-250.
[32] Siraye W, Wakjira A, Nallamothu RB. An Analysis of Fuel Oil
Sloshing in Partially Filled Cargo Tanker Trucks under Cornering
Conditions Using Various Baffle Systems. Journal of Engineering.
2023; 9941864.

348