See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/358167615

Numerical evaluation of vehicles aerodynamics in platoon using CFD

simulation

Article in IOP Conference Series Materials Science and Engineering · October 2021
DOI: 10.1088/1757-899X/1220/1/012024

CITATIONS READS

3 126

2 authors:

Iacob-Liviu Scurtu Marius Ioan Gheres

Universitatea Tehnica Cluj-Napoca Technical University of Cluj-Napoca
51 PUBLICATIONS 74 CITATIONS 30 PUBLICATIONS 44 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Iacob-Liviu Scurtu on 28 January 2022.

The user has requested enhancement of the downloaded file.

IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS You may also like

- The Earth radiation balance as driver of
Numerical evaluation of vehicles aerodynamics in the global hydrological cycle
Martin Wild and Beate Liepert
platoon using CFD simulation - Summary of Papers
Serge Gauthier, Snezhana I Abarzhi and
Katepalli R Sreenivasan
To cite this article: I L Scurtu and M I Gheres 2022 IOP Conf. Ser.: Mater. Sci. Eng. 1220 012024
- The global atmospheric water cycle
Lennart Bengtsson

View the article online for updates and enhancements.

This content was downloaded from IP address 193.226.5.148 on 28/01/2022 at 11:14

AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

Numerical evaluation of vehicles aerodynamics in platoon

using CFD simulation

I L Scurtu, M I Gheres
Department of Automotive Engineering and Transports, Technical University of Cluj-
Napoca, Romania

marius.gheres@auto.utcluj.ro

Abstract. Due to increased traffic and new technologies developed to improve road safety, a
new vehicle driving technique is being studied. The vehicles’ platooning driving method has as
objective to minimize the aerodynamic drag and therefore the fuel consumption. The aim of this
study is to evaluate the distance between platoon traveling vehicles and to propose an optimal
travel distance. The CFD evaluation is performed for two categories of vehicles at a given
velocity and distance between them for two cases: in the first simulation scenario, a tractor-trailer
is represented, followed by a SUV positioned at a given distance from the rear of the vehicle
combination; in the second simulation, two tractor-trailers at the same boundary condition as for
the first case are simulated. The vehicles models are made by using a CAD modelling
environment, respecting the overall dimensions of an existing vehicle. The numerical evaluation
of a distance between vehicle models is performed using the CFD method based on the Navier-
Stokes equations averaging. The last part of the paper presents the interpretation of CDF
simulation results, establishing the recommended travel distance between vehicles and
conclusions of this study.

1. Introduction
The growth of the vehicles fleet requires the creation of more efficient transport infrastructure. The
aerodynamic interaction of the vehicles in platoon makes possible a lower fuel consumption depending
on the distance between the vehicles. For conducting this aerodynamic study several papers are studied
with the purpose to determine the state of the art regarding the CFD evaluation of vehicles in platoon.
The vehicles equipped with Intelligent Transport System, like distance sensor, adaptive braking system,
global position system, adaptive cruise control, allow the lower distance between the vehicles than a
vehicle length keeping the road safety of the vehicles in platoon [1]. An experimental investigation study
of a two-truck platoon considering the inter-vehicle distance, lateral offset and yaw is performed by
Törnell et al. [2], where, in experiments, were used two 1:6 scale detailed models in a wind tunnel with
a moving ground. In conclusion, a short inter-vehicle distance leads to a reduction of drag coefficient,
which allows a more efficient running on the road.
Iozsa et al. [3] has performed an CFD study for four truck convoy at 25 m/s air velocity in Ansys
software. This study of aerodynamics consists in streamlines analysis, velocity vector analysis and
pressure map for four identical trucks that follow each other at a 3 m distance. The results show that this
distance between vehicles can reduce by 75% the aerodynamic resistance for the vehicles included in
platoon A CFD comparative experiment is performed by Ebrahim et al. [4] where is studied the platoon
aerodynamics between Ahmed model for two inter-vehicle spacing configurations, with 0.125 L and

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

1 L intervehicle spacing, where L is the body vehicle length. The Ahmed body represents a simplified
vehicles model described originally by S.R. Ahmed in 1984. The Ahmed model geometry was modified
without mounts (generic wheels) at 0.75 L, 0.5 L and 0.4 L, was based on a 25° slant angle. In
conclusion, this study is capable to simulate platoon cases in a virtual wind tunnel without sacrificing
aerodynamic resolution.
Gnatowska and Sosnowski [5] described the CFD analysis inter-vehicles distance effect in platoon
arrangement. The vehicles models are represented by the two identical Ahmed bodies. In terms of
determining the drag coefficient values, the frontal area was 0.007 m2 and the slant angle is to 25 degrees.
The inter-distance between the bodies is in the range of 0÷1.2 L. The CFD experiment is conducted at
two velocities of 8 m/s and 88 m/s. Following the simulations, the complex flow fields have been
visualized and it can be observed the velocity fields and pressure map generated at different regions.
Simeon et al. [6] presented an aerodynamic CFD study for four vehicles in platoon, using TruckSim
software for predicting the fuel consumption efficiency for each vehicle. The vehicles are inter-spaced
in platoon at 30 ft (9.144 m), 50 ft (15.24 m) and 100 ft (30.48 m) by using a cruise control system on
the truck. Another platoon CFD study was performed using OpenFoam software [7]. This presents the
estimation error-based performance of five learning algorithms: Support vector regression, Polynomial
regression, Linear regression and two different models of Neural Networks. The drag coefficient for
each vehicle in two, three and four vehicle platoons are provided at different inter-vehicle distances.
Aerodynamic drag analysis of autonomous electric vehicle platoons was studied by Kaluva et al [8].
The vehicles platoon was studied in urban environments for two vehicles model: a minibus and a
passenger car. The platoon size is in the range of 1÷7 vehicles, running at 10 m/s velocity. The obtained
results show a reduction of drag coefficient of the platoon of up to 24% at a 1 m inter-vehicle distance.
Another CFD study of three platoon trucks with a sensing and control system was performed at 80 km/h
velocity, with an inter-distance between trucks of 10 m. By using the vehicles platooning the fuel
consumption has improved by approx. 14 % [9].

2. Work method
This research concerns a numerical investigation of the airflow velocity distribution, air pressure on the
vehicle body and drag coefficient computed at 25 m/s velocity of the vehicles in platoon. The vehicles
are represented by the CAD models of a tractor-trailer and a medium size SUV. These models were
designed at a real overall dimension.

2.1. CAD modelling

The model of the medium SUV is designed in SolidWorks CAD environment to the real car size. This
model has been used in an aerodynamic study that showed the aerodynamic influence of the bicycle
rack [10].

Figure 1. The CAD model and overall dimensions of the SUV.

The overall dimensions of the SUV model are presented in figure 1.

2
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

Figure 2. The CAD model and overall dimensions of the tractor-trailer.

In figure 2, the designed CAD model of the tractor-trailer is presented. This model presents most
auxiliary elements of a real tractor-trailer. The frontal area of the tractor-trailer model is 9.46 m2.

2.2. Mathematical approach

2.2.1. Force on vehicle body.

When the vehicle is driving on the road, the aerodynamic forces act on the bodywork, which generates
a movement resistance and affects the stability. In this case study, the aerodynamic evaluation of the
vehicles in platoon is performed by the computation of the aerodynamic coefficient and aerodynamic
forces acting on the vehicles in the opposite travelling direction [11]. The expression of the aerodynamic
force is given in relation 1:
1
Fx     V 2  CD  A
2 (1)
where: ρ is the air density, V is the air velocity, A is the frontal area of the vehicle body and CD is the aerodynamic
drag coefficient.
From the expression of the drag force, it is deducted the expression of the drag coefficient presented
in relation 2:
2  Fx
CD 
 V 2  A (2)

2.2.2. Mathematical model.

The airflow around the vehicle body is characterized by the turbulent flow and the fluctuating velocity
fields. The largest fluctuating velocities appear between the space from the cabin and semi-trailer and
between the vehicles in case of vehicles platoon. The mathematical approach used the equations
Reynolds Averaged Navier Stokes, a model used in most engineering applications that use fluid flow
simulations [12]. The equations can be written in the Cartesian tensor form, presented in relations 3 and
4:
U i
0
xi (3)



x j
 
U jU i  u 'j ui'  
P 

xi x j
 2 Sij 
(4)
̅̅̅̅̅
where: Ui is the mean of velocity tensor, xi is vector position, u ′ u′ is the average of fluctuating velocity, P is the
j i
mean static pressure, µ is the molecular viscosity.
In relation 5 the strain rate expression is given:
1  u u j 
Sij   i  
2  x j xi 
(5)

3
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

The standard k-ε turbulence model is applied in this study and is given in equation 6:
k U i   T   
U j   ij        
x j xi x j  k  x j  (6)
where: τij is the Reynolds stress, µT is turbulent viscosity, σk is closure coefficient, k is turbulent kinetic energy
and ε is the dissipation rate of turbulent energy.
The dissipation rate of turbulence kinetic energy equation is given in the next relation:
  U i 2   T   
U j  C 1  ij  C 1       
x j k x j k x j    x j  (7)
The turbulent viscosity expression is given in the relation 8:
k2
T   C
 (8)

2.2.3. Reynolds number.

To perform the aerodynamic simulation is necessary to establish the airflow regime. This is established
by calculating the Reynolds number that is an important dimensionless variable and represents the ratio
between inertial and viscous forces [13]. When the number value is high, the inertial force dominates
and viscous forces can be neglected; if the number value is lower, the viscous forces dominate and the
inertial forces are usually neglected. The expression of the Reynolds number is presented in the relation
9:
 V  L
Re 
 (9)
To determine the flow rate around the truck and semi-trailer, the Reynolds number is calculated
following the relation 1 at V=25 m/s airflow velocity, the air density is ρ = 1.2041 kg/m3, corresponding
to a temperature T=20°C, the length of the truck assembly is L = 17.185 m and the dynamic air viscosity
µ = 1.802 ∙ 105 kg/(m∙s). To establish the external flow regime in this case, the airflow is considered
over a plane surface. The error that can be highlighted between reality and simulation is given by the
fact that the undisturbed air does not move in relation to the ground while in the case of simulation the
air is moving in relation to the ground. If the value of the Reynolds numbers is Re<5x105 the airflow is
laminar. The transient flow regime is situated in the range of 5x105< Re<107, while after Re>107 the flow
is turbulent. The obtained value of the Reynolds number in the studied case is 2.87x107, which indicate
a turbulent airflow.

3. Boundary conditions and numerical simulation

The CFD domain is designed to allow the change of the distance between the vehicles and permits to
consider the road in the simulation. For better visualisation of the airflow around the vehicle body’s, the
computational domain has a width of 6 m, a height of 7 m and a length of 80 m. The first vehicle is
placed into the CFD domain at a distance of 8 m from the air inlet. In this study, two simulations are
performed: the first CFD simulation case consists of a tractor-trailer model that has a 25 m/s velocity
followed with the same speed by the SUV at 5 m, 10 m, 15 m, 20 m, 25 m and 30 m; the second
simulation case is performed at the same initial conditions, but a similar tractor-trailer type is replacing
the SUV. The velocity value was chosen due to the European speed limitation for the tractor-trailer
vehicles. The computational domain for the first case is meshed into 93266 fluid cells, of which 16397
fluid cells are in contact with the solid surface; for the second case, the considered volume is meshed
into 107005 cells, of which 25940 fluid cells are in contact with the vehicles surface. For both cases, the
mesh is generated automatically at a ratio factor of the initial mesh, which means that the nearby of the
fluid layer is denser, to increase the accuracy of the results.

4
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

Figure 3. The CFD computational domain and mesh grid.

In figure 3, the meshed grid of the CFD domain is presented. It can be observed that the mesh density
is high in the nearby area of the vehicle surface.
To have a comparative term of the results, separate tractor-trailer and SUV simulations are performed
to compute the pressure on bonnet and windscreen surfaces, and also the drag coefficient. For the
interpretation of these results, the surface of the bonnet and windscreen are chosen as target surfaces.
All simulations are performed at a velocity of 25 m/s and with parameters values applied to the
calculation of the Reynolds number. Figure 4 shows the bonnet and windscreen surface for both vehicles
from which the pressure are taken.

Windscreen surface

Bonnet surface

Figure 4. The reference surfaces chosen for both vehicles.

4. Results
This section presents the simulation results. The results are discussed in terms of global drag forces and,
respectively, bonnet and windscreen relative pressure obtained for different varied inter-vehicle
distances.

4.1. Single truck and single SUV

In this section are presented the first reference results of separated simulations for single tractor-trailer
and SUV (table 1). The bonnet and windscreen surface has been chosen because their shape can
influence the vehicle aerodynamics.
Table 1. Results for single truck and single SUV
Tactor-trailer SUV
Drag coefficient 0.728 0.433
Drag force (N) 2594.13 442.61
Relative pressure on the bonnet surface (Pa) 101633 101235
Relative pressure on the windscreen surface (Pa) 101568 101320

The values of the aerodynamics coefficient, aerodynamic force, relative pressure of the bonnet
surface and windscreen are determined at an initial air speed of 25 m/s for each vehicle. Relative pressure
values are related to atmospheric pressure.

5
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

4.2. Truck followed by SUV

Table 2 presents the numerical results of the truck followed by the SUV in the range of 5-30 m.
For the validation, the results for the leading tractor-trailer followed by SUV are presented in figure
5, in ascending order of the distance between the vehicles. The airflow velocity distribution is in the
range of 5-30 m. Following the visual distribution of the velocity, which is presented in figure 5 and the
numerical results from table 2. With increasing the distance between vehicles, the drag coefficient and
drag force increase. The relative pressure on the SUV bonnet decreases with the increase of inter-
distance from vehicles. The longitudinal view of the velocity streamlines distribution is represented on
the middle plane of the truck body and the top view is represented on the plane situated at 1 m distance
from the ground.
5m

10m

15m

20m

25m

30m

Figure 5. Airflow velocity distribution at different distances of a tractor-trailer followed by SUV.

6
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

The value of the drag coefficient and aerodynamic force corresponds to the vehicles assembly
consisting of the truck followed by SUV.

Table 2. Numerical results of the truck followed by SUV.

Distance between truck and SUV (m)
5 10 15 20 25 30
Drag coefficient 0.757 0.770 0.781 0.784 0.785 0.788
Drag force (N) 2697.27 2744.93 2781.56 2795.05 2793.03 2809.04
Relative pressure on
the SUV bonnet 101341 101303 101291 101284 101280 101278
surface (Pa)
Relative pressure on
the SUV windscreen 101338 101314 101316 101315 101312 101322
surface (Pa)

The relative pressure on the surface of the SUV bonnet is higher in the case of two vehicles than the
relative pressure value obtained in the individual SUV simulation, decreasing with the increase in the
distance between the vehicles and the pressure on the surface of the windscreen is less in the range of
10-25 m.

4.3. Tractor-trailer followed by another similar tractor-trailer

The airflow velocity distributions between the identical tractor-trailer combinations are presented in
figure 6, where it can be observed the simulation results in the same boundary condition as in the
previous simulations.
Table 3 presents the numerical results of the simulation between the similar two tractor-trailers at the
same boundary conditions.

Table 3. Numerical results for the two tractor-trailer combinations.

Distance between trucks (m)
5 10 15 20 25 30
Drag coefficient 0.950 1.041 1.080 1.119 1.126 1.133
Drag force (N) 3385.44 3708.40 3846.34 3984.71 4009.36 4032.82
Relative pressure
on the bonnet 101393 101441 101465 101489 101495 101500
surface (Pa)
Relative pressure
on the windscreen 101368 101394 101435 101453 101471 101476
surface (Pa)

In the case of driving in platoon the aerodynamic resistance value of the second truck is less than in
the case of a single truck and the aerodynamic drag on the leading truck increases.

7
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

5m

10m

15m

20m

25m

30m

Figure 6. Airflow velocity distribution at different distances of two same tractor-trailers.

4.4. Tractor-trailer during the overtaking operations by the second vehicle

This section presents the case when the tractor-trailer is overtaking by the second vehicle: in the first
case, the tractor-trailer is followed by the SUV and in the second case by the tractor-trailer. The
overtaking scenario is composed by the trailer-tractor as a leading vehicle followed by the second
vehicle positioned at a 5 m distance and the half part of the vehicle body is placed on the right side of
the tractor-trailer.
Figure 7 presents the airflow distribution in overtaking intention in case of tractor-trailer followed
by SUV, the 7a present the pressure on the vehicle surface and 7b show streamlines distribution-top

8
AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

view at 1 m. In figure 8, it is represented the airflow distribution in overtaking intention in the case of
tractor-trailer followed by tractor-trailer, the 8a presents the pressure on the vehicle surface and 8b show
streamlines distribution-top view at 1 m.

a) Pressure distribution on the vehicles surface

b) Streamlines distribution -top view at 1 m

Figure 7. Airflow distribution in overtaking intention.

Table 4. Numerical results for the overtaking intention case.

Relative pressure on Relative pressure
Drag Drag force
the bonnet surface on the windscreen
coefficient (N)
(Pa) surface (Pa)
Tractor-trailer followed by SUV 0.783 2787 101302 101349
Tractor-trailer followed by
1.136 4047 101523 101496
tractor-trailer

a) Pressure distribution on the vehicles surface

b)Streamlines distribution -top view at 1 m

Figure 8. Airflow distribution in overtaking intention

9
 AITS 2021 IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1220 (2022) 012024 doi:10.1088/1757-899X/1220/1/012024

The drag coefficient and the drag force it is higher in the case of the tractor-trailer followed by a
tractor-trailer in overtaking intention because the tractor-trailer has a larger front surface in comparison
with SUV.

5. Conclusions
In this paper, the CFD based numerical evaluation of the aerodynamic behaviour in platoon of tractor-
trailer and SUV is determined. Accurate numerical simulations are performed to describe the airflow
structures around the vehicles body and to obtain the numerical values of the pressure on the bonnet and
windscreen surface. For better road safety, 20 m between the vehicles can be considered as a safe choice.
The movement of vehicles in convoy is recommended to be applied on the highway so as not to affect
the traffic on the two-way public road. The results of this study can be used as a reference for keeping
the optimal distance between vehicles through the adaptive cruise control.

References
[1] Schito, P. 2012 Numerical and experimental investigation on vehicles in platoon SAE
International Journal of Commercial Vehicles, 5(2012-01-0175), 63-71
[2] Törnell, J., Sebben, S., & Elofsson, P. 2021 Experimental investigation of a two-truck platoon
considering inter-vehicle distance, lateral offset and yaw Journal of Wind Engineering and
Industrial Aerodynamics, 213, 104596
[3] Iozsa, D., Stan C., and Ilea L. 2017 Study on the influence of the convoy rolling over aerodynamic
resistance IOP Conf. Series: Materials Science and Engineering 252 (2017) 012035
[4] Ebrahim, H. M., Dominy, R. G., & Leung, P. S. 2016 Evaluation of vehicle platooning
aerodynamics using bluff body wake generators and CFD In 2016 International Conference
for Students on Applied Engineering (ICSAE) (pp 218-223) IEEE
[5] Gnatowska, R., & Sosnowski, M. 2018 The influence of distance between vehicles in platoon on
aerodynamic parameters. In EPJ Web of Conferences (Vol. 180, p. 02030) EDP Sciences.
[6] Siemon, M., Smith, P., Nichols, D., Bevly, D., & Heim, S. 2018 An integrated CFD and truck
simulation for 4 vehicle platoons (No. 2018-01-0797). SAE Technical Paper
[7] Jaffar, F., Farid, T., Sajid, M., Ayaz, Y., & Khan, M. J. 2020 Prediction of Drag Force on Vehicles
in a Platoon Configuration Using Machine Learning IEEE Access, 8, 201823-201834
[8] Kaluva, S. T., Pathak, A., & Ongel, A. 2020 Aerodynamic drag analysis of autonomous electric
vehicle platoons Energies 13(15) 4028
[9] Tsugawa, S., Kato, S., & Aoki, K. 2011 An automated truck platoon for energy saving IEEE/RSJ
international conference on intelligent robots and systems (pp. 4109-4114) IEEE
[10] Scurtu I.L. 2021 Aerodynamic performance evaluation for a vehicle structure equipped with a
bicycle rack Automotive engineering ISSN 1842-4074 vol 58 pp 10-14
[11] Bodea S.M., Prodan C.V., Scurtu I.L. 2019 The Aerodynamic Study of a Body Truck In: Burnete
N., Varga B. (eds) Proceedings of the 4th International Congress of Automotive and Transport
Engineering (AMMA 2018). AMMA2018 2018. Proceedings in Automotive Engineering.
Springer, Cham.
[12] Jurco A.N. 2021 Study of the influence of roof luggage box on a vehicle aerodynamics Automotive
engineering ISSN 1842-4074 vol 58 pp 15-18
[13] Balcau M.C. 2021 Aerodynamic study of a car towing a motorcycle trailer Automotive
engineering ISSN 1842-4074 vol 58 pp 22-26

10

View publication stats