International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 115

ISSN 2229-5518

Dynamic modeling and analysis for improving

energy-efficient prototype performance
Renan Luís Knabben, Sérgio Junichi Idehara, Helton da Silva Gaspar

Abstract— Contemporary engineering seeks the optimization of processes, new technologies, and the saving of resources, aiming at
environmental sustainability. In the field of automotive engineering, universities around the world are working on research and extension
projects so that new ideas can emerge in the area. This is the case of the Shell Eco-marathon competition, which aims to design ultra-
efficient automotive prototypes. Thus, this work sought to carry out the dynamic modeling and analysis of the use of a suspension system in
an ultra-efficient automotive prototype, aiming at the use of automotive technologies for the consequent increase in its energy efficiency. The
research was consolidated referencing two virtual models: the original (without a suspension system) and the proposed one (with a
suspension system). With numerical simulations, it was possible to carry out vehicle dynamic analysis for both models, such as rollover and
energy efficiency. With suspension system parameters, the most significant results were obtained in the rollover analysis, which indicated a
reduction of more than 50% in the force acting on the tire in some critical situations. In the case of the energy efficiency, it was evaluated an
energy saving of almost 40% in brake situations compared with original configuration, proving the advantage of the use of this mechanism
in ultra-efficient prototypes.

Index Terms— Automotive, Design, Shell Eco-marathon, Suspension, Ultra-energy-efficient cars, Vehicle rollover, Vibration

——————————  ——————————

1 INTRODUCTION

IJSER
nergy efficiency is a topic widely addressed in the indus- achieve the greatest distance per gallon of fuel. The first official
try, as it seeks maximum performance using the least competition took place in France, in 1985, and was attended by
amount of energy possible. The automotive sector follows thousands of students and scientists from more than 20 Euro-
the same path, as it is based on innovative solutions, such as the pean countries. Today, it takes place annually on 4 different
development of electric and hybrid vehicles (electric and inter- continents and has the presence of more than 100 teams per
nal combustion engines acting simultaneously), the use of light stage [7]. There are two classes of vehicles within the Shell Eco-
materials and the improvement of aerodynamics, for example marathon: prototype and urban concept, which are divided into
described by [1] and [2]. Universities and research institutes three categories: electric battery [8], hydrogen fuel cell [9] and
also work on research projects so that new developments can internal combustion engine (gasoline, diesel or ethanol).
emerge in the automobile world [3]. Among the various exam-
ples, it is possible to mention the energy efficiency prototypes, Energy-efficient prototypes, or also known as ultra-efficient
which use clean energy technologies and have an extremely prototypes, Fig 1, are automotive vehicles designed to achieve
low consumption for its operation. As one of the main projects, the maximum possible energy efficiency. That is, travel long
we can mention the energy efficiency teams, which aim to distances while consuming less fuel [10]. To achieve such re-
model and develop a virtual design [4], [5], and manufacture sults, some innovations must be applied to prototypes, a feature
automotive prototypes [6] that travel the maximum distance that makes them different when compared to commercial urban
while consuming the least amount of fuel possible. One of the vehicles. As main innovations, it is the use of 3 wheels (two
most known competition is Shell Eco-marathon (SEM). In the front and one rear) instead of 4, extremely aerodynamic geom-
case of SEM, the prototypes must travel a fixed distance by con- etry almost without turbulences as described by [11], availabil-
suming the least amount of energy. ity for only one passenger (Fig 2) and the presence of light ma-
terials in its structural composition [12]. Sawulski and
This is a worldwide energy efficiency competition spon- Ławrynczuk [13] and Omar et al. [14] worked in optimized con-
sored by Shell, in which the participants are students from trol strategy related to powertrain low fuel consumption as an
schools and universities in different countries around the automated vehicle engine and motor control, respectively, to
world. The teams aim to develop automotive vehicles that achieve an improved performance of this kind of vehicles.
achieve the highest possible fuel efficiency [7]. The idea came
up in 1939 at Shell's research laboratory, in the United States, In Shell Eco-marathon competitions, most of the teams choose
when scientists established a friendly bet to find out who could for projects that contain three wheels. Even so, there are teams
that prefer to use four wheels in their vehicles [15]. Such config-
————————————————
uration, however, may increase the weight and the rolling re-
 Renan L Knabben, Federal University of Santa Catarina, Rua Dona
Francisca, 8300, 89219-600, Joinville/SC, Brazil, sistance since there will be an extra tire in contact with the
[email protected] ground. According to [16], the configuration with three wheels is
 Sergio J Idehara, Federal University of Santa Catarina, Rua Dona the most used for providing a series of advantages. As main, al-
Francisca, 8300, 89219-600, Joinville/SC, Brazil, [email protected] ready cited aerodynamics, the front and side visibility of the pi-
 Helton S Gaspar, Federal University of Santa Catarina, Rua Dona
Francisca, 8300, 89219-600, Joinville/SC, Brazil, [email protected] lot, stability and resistance to overturning. However, despite be-

IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 116
ISSN 2229-5518

ing very competent, their embedded technology is relatively sim- for a large number of conditions, such as loaded/ unloaded ve-
ple, a condition that may compromise the handling, comfort and hicle, acceleration/ braking, straight line/ curves and regular/
safety of its users [17]. The use of a light structure, like in the chas- irregular streets. Another primary function of the suspension
sis design, may decrease its stiffness and strength [18]. Kral et al. system, according to [25], is its direct contribution to the perfor-
[19], for instance, discuss that a higher structure stiffness is mance of the vehicle on the road. In curve, relieving the efforts
needed to achieve a correct geometry of the steering system from the lateral dynamics and allowing its execution at higher
when cornering, which can increase the overall weight. Some so- speeds, as well as in a straight line, generating greater stability
lutions are studied to improve the vehicle safety, as the approach and control of the car in the face of efforts caused by accelerations
made in [20] and [21]. The authors developed an Advanced and more intense braking (longitudinal dynamics).
Driver Assistance Systems to improve on driver and vehicle
safety predicting possible collision situations. These mechanisms In the case of energy efficiency prototypes, the focus on the
may avoid more serious injuries on the driver as found by [22]. suspension system is not limited to relieving the excitations aris-
ing from the irregularities of the road, but rather maximizing its
performance in terms of handling, especially in cornering situa-
tions, promoting more stability and avoiding rollover to the ve-
hicle, [26], where the driver can work at higher speeds in such
conditions. Therefore, it is necessary to adopt a set of values of
stiffness and damping according to the vehicle dynamics, seek-
ing a gain in efficiency and better control of the prototype by the
pilot. The Fig 3 show a sequence of snapshots taken of an accident
with the EFICEM team prototype during SEM 2019. They exem-
plify the risk of suspension system absence.
Fig. 1. Frontal view of the ultra-efficient prototype design (Federal

IJSER
University of Santa Catarina, Eficem).

Fig. 3. Illustration of a rollover case during cornering movement.

Fig. 2. One-person ultra-efficient prototype (Federal University of Thus, in this work a theoretical studies and discussions of the
Santa Catarina, Eficem). suspension allocation in the prototype is described, regards to
less reduction of vehicle velocity in cornering maneuvers, based
on the velocity limit to avoid the rollover [27], and improve the
Then, seeking a development of a very light ultra-efficient general vehicle energy efficiency.
prototypes, most teams choose to design vehicles that are too
simple, avoiding the use of common components found in com-
mercial cars. However, given that the main objective of the pro-
ject is to achieve the maximum energy efficiency, the perfor- 2 MATERIAL AND METHODS
mance of the vehicle should be analyzed regardless of its weight. The theoretical considerations and bases used to develop the
Thus, the insertion of certain mechanical components in the pro- mathematical model will be detailed. In order to evaluate the
totype, such as a suspension system, can contribute significantly energy efficiency provided by the suspension system, numeri-
to its final efficiency despite the weight increasing. Carlsen and cal simulations of the vehicle dynamics are performed.
Oma [23] built a vehicle prototype for Shell Eco-Marathon using
suspension system that may improve the control of camber and 2.1 Mathematical modeling
caster angles related to the vehicle dynamics. The study of the movement of a vehicle initiates with its dy-
namic modeling, which involves its degrees of freedom (DOF)
The suspension is one of the main sets of an automotive vehi-
and the determination of its coordinate system. The dynamic
cle, it has the role of ensuring tire/ track contact, providing com- equilibrium equations that govern the system's response can be
fort to users and also safety in the curve driving at higher speeds, expressed in the matrix form, according to following equation:
thus ensuring more stability. As defined by [24], the suspension
is responsible for meeting the driveability and comfort required
IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 117
ISSN 2229-5518

[𝑀]{𝑧̈ (𝑡)} + [𝐶]{𝑧̇ (𝑡)} + [𝐾]{𝑧(𝑡)} = {𝐹(𝑡)} (1) mass, resulting

(1) in (5) and (6).
1
It contains mass [𝑀], damping [𝐶] and stiffness [𝐾] param- 𝐼𝑥 . 𝜙 = . 𝑡𝑏 . (𝑧𝑠2 . 𝑘𝑡 − 𝑧𝑠1 . 𝑘𝑡 ) − 𝑐𝑡𝑥 . 𝜙̇
̈ (5)
2
eters written in the form of matrices, along with vectors of ac-
celeration {𝑧 ̈(𝑡)}, velocity {𝑧 ̇(𝑡)}, displacement {𝑧(𝑡)} and force 𝐼𝑦 . 𝜃̈ = − 𝑘𝑡 . 𝑙𝑟 . 𝑡𝑎𝑛2 (𝜃) + 2. 𝑘𝑡 . 𝑙𝑓 . 𝑡𝑎𝑛2 (𝜃) − 𝑐𝑡𝑦 . 𝜃̇ (6)
{𝐹(𝑡)}, which are in the time domain. For ultra-efficient auto-
motive prototypes, which have three wheels (two at the front
The variables 𝐼𝑥 and 𝐼𝑦 correspond to the moments of iner-
and one at the rear), two different models will be compared: the
tia of the axes of longitudinal and transversal rotation, respec-
first one (Fig 4) represents the current configuration of ultra-
tively. In turn, 𝑙𝑟 represents the longitudinal distance from the
efficient prototype, without a suspension system (3 DOF
center of gravity to the rear axle of the prototype. The products
model). The second one, Fig 5, will have a suspension set on its
𝑐𝑡𝑥 . 𝜙̇ and 𝑐𝑡𝑦 . 𝜃̇ represent the mechanical friction present in the
front wheels (5 DOF model), enabling in-depth analysis of its
𝑥 and 𝑦 rotations, in sequence.
dynamics depending on the final efficiency.
2.3 Five DOF Model
2.2 Three DOF Model
This model has five degrees of freedom, which are:
The model degree of freedom corresponds to the vertical dis-
1. bounce (𝑧𝑠 ),
placement of the mass in 𝑧, called bounce (𝑧𝑠 ), and its rotations
2. roll (𝜙),
in 𝑥 and 𝑦, called roll (𝜙) and pitch (𝜃), respectively. The dy-
3. pitch (𝜃),
namic equation is:
4. vertical displacement of the sprung mass and vertical dis-
𝑚𝑠 . 𝑧̈𝑠 = − 𝑘𝑡 . 𝑧𝑠1 − 𝑘𝑡 . 𝑧𝑠2 − 𝑘𝑡 . 𝑧𝑠 − 𝑐𝑧𝑠 . 𝑧̇𝑠 (2) placement
(2) of the unsprung mass of the left front wheel
(𝑧𝑚1 ),
5. and right front wheel (𝑧𝑚2 ).

IJSER
The equation of the model is

𝑚𝑠 . 𝑧̈𝑠 = − 𝑘1 . (𝑧𝑠1 − 𝑧𝑚1 ) − 𝑐1 . (𝑧̇𝑠1 − 𝑧̇𝑚1 )

− 𝑘2 . (𝑧𝑠2 − 𝑧𝑚2 ) − 𝑐2 . (𝑧̇𝑠2 − 𝑧̇𝑚2 ) (7)
− 𝑘𝑡 . 𝑧𝑠 − 𝑐𝑧𝑠 . 𝑧̇𝑠

The variables 𝑘1 and 𝑘2 are equivalent to the stiffness con-

stants of the suspension 1 and 2, in that order. Similarly, the
constants 𝑐1 and 𝑐2 refer to the damping of suspension 1 and 2.

The equations of displacements at the front left (𝑧𝑠1 ) and

right (𝑧𝑠2 ) ends refer to (3) and (4), already shown in the previ-
ous model. In turn, the velocity equations are:
Fig 4. Model of 3 Degrees of Freedom.
1
𝑧̇𝑠1 = 𝑧̇ 𝑠 − 𝑙𝑓 . 𝜃̇ (𝜃) + . 𝑡 . 𝜙̇. 𝑠𝑒𝑐 2 (𝜙) (8)
The variables 𝑚𝑠 and 𝑧̈𝑠 represent the value of the sprung 2 𝑏
1
mass of the prototype and its acceleration, respectively. 𝑘𝑡 is 𝑧̇𝑠2 = 𝑧̇ 𝑠 − 𝑙𝑓 . 𝜃̇ (𝜃) − . 𝑡𝑏 . 𝜙̇. 𝑠𝑒𝑐 2 (𝜙) (9)
equivalent to the tire stiffness constant. The product 𝑐𝑧𝑠 . 𝑧̇𝑠 rep- 2
resents the mechanical friction present in the displacement of
the suspended mass. The variables 𝑧𝑠1 and 𝑧𝑠2 correspond to the Analogously the unsprung mass equations are
displacement in 𝑧 of the suspended mass in the front left and
right ends of the prototype, respectively, which are written us- 𝑚1 . 𝑧̈𝑚1 = 𝑘1 . (𝑧𝑠1 – 𝑧𝑚1 ) + 𝑐1 . (𝑧̇𝑠1 – 𝑧̇𝑚1 ) − 𝑘𝑡 . 𝑧𝑚1 (10)
ing the equations:
𝑚2 . 𝑧̈𝑚2 = 𝑘2 . (𝑧𝑠2 – 𝑧𝑚2 ) + 𝑐2 . (𝑧̇𝑠2 – 𝑧̇𝑚2 ) − 𝑘𝑡 . 𝑧𝑚2 (11)
1
𝑧𝑠1 = 𝑧𝑠 − 𝑙𝑓 .𝑡𝑎𝑛 (𝜃) + . 𝑡𝑏 . 𝑡𝑎𝑛(𝜙) (3)
2
1 The variables
(3) 𝑚1 and 𝑚2 are equivalent to the values of the
𝑧𝑠2 = 𝑧𝑠 − 𝑙𝑓 .𝑡𝑎𝑛 (𝜃) − . 𝑡𝑏 . 𝑡𝑎𝑛(𝜙) (4) unsprung masses. In turn, 𝑧̈𝑚1 and 𝑧̈𝑚2 correspond to their ac-
2
celerations. The last two equations of the model are obtained
The 𝑙𝑓 symbol is equivalent to the longitudinal distance through the 𝑥 and 𝑦 rotations of the sprung mass:
from the front of the prototype to its center of gravity. 𝑡𝑏 de-
notes the width of the gauge (𝑡𝑟 + 𝑡𝑙 ). Other two equations in
the model come from the 𝑥 and 𝑦 rotations of the suspended
IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 118
ISSN 2229-5518

𝐼𝑥 . 𝜙̈ = 1/2. 𝑡𝑏 . [ 𝑘2 . (𝑧𝑠2 – 𝑧𝑚2 ) + 𝑐2 . (𝑧̇𝑠2 − 𝑧̇𝑚2 ) 𝑀𝑡 . 𝑔. 𝑙𝑓 (18)

𝑁3 =
(𝑙𝑟 + 𝑙𝑓 )
− 𝑘1 . (𝑧𝑠1 – 𝑧𝑚1 ) − 𝑐1 . (𝑧̇𝑠1 – 𝑧̇𝑚1 )] (12)
− 𝑐𝑡𝑥 . 𝜙̇
𝐼𝑦 . 𝜃̈ = [ 𝑘1 . (𝑧𝑠1 – 𝑧𝑚1 ) + 𝑐1 . (𝑧̇𝑠1 – 𝑧̇𝑚1 ) (7)
+ 𝑘2 . (𝑧𝑠2 – 𝑧𝑚2 )
(13)
+ 𝑐2 . (𝑧̇𝑠2 − 𝑧̇𝑚2 ) ] . 𝑙𝑟
− 𝑘𝑡 . 𝑙𝑓 2 . 𝑡𝑎𝑛(𝜃) − 𝑐𝑡𝑦 . 𝜃̇

Thus, with the definition of all the physical equations of the

proposed model, it is possible to determine the complete move-
ment of the vehicle by the numerical differential equation solu-
tion.

2.4 Vehicle dynamics

An extremely important factor for any energy efficient vehicle
Fig. 6. Lateral Dynamic Analysis of an Ultra-efficient Prototype.
is its performance in curves, because, due to the limited geom-
etry and stability, it is subject to overturns, mainly in curves
with small radii that require lower speeds to be performed. In- In dynamic condition, the magnitude of these normal
itially, it is assumed that the vehicle is in uniform circular mo- forces of the prototype (𝑁1 or 𝑁2 , depending on the direction of

IJSER
tion, with a given speed 𝑉 and radius 𝑅. In these circumstances the curve) is compared with the product of the displacement of
there is a force (𝐹𝑐𝑝 ) that keeps the vehicle on its path: the unsprung mass (𝑧𝑚1 or 𝑧𝑚2 ) by the tire stiffness (𝑘𝑡 ). If the
magnitude of the terms 𝑧𝑚1 . 𝑘𝑡 or 𝑧𝑚2 . 𝑘𝑡 exceeds the normal
𝑉2 force (𝑁1 or 𝑁2 ), respectively, it is assumed that the tire starts to
𝐹𝑐𝑝 = 𝑚 . (14)
𝑅 lose contact with the track, where it is a rollover possible situa-
tion.
In the non-inertial frame of reference fixed on the vehicle,
a fictitious force acts over its CG towards the centrifugal direc- 2.4 Numerical simulation setup
tion, with the same magnitude of the centripetal force. As con- The simulations, in Matlab environment, are performed by us-
sequence of the resulting force, a positive resultant moment is ing the real conditions of the Shell Eco-marathon Americas cir-
generated (𝑀𝑟 ), which tends to make the prototype overturn. It cuit (Sonnoma Raceway, adapted), Fig. 6. The vehicle velocity
is calculated by multiplying the resulting force by the distance will be considered constant in the track. The virtual model of an
between the center of gravity and the floor, ℎ𝑐𝑔 , [28]. ultra-efficient automotive prototype, Fig 8, was based to obtain
the parameters of the vehicle, where the data is presented in the
𝑉2 Table 1.
𝑀𝑟 = 𝑚 . . ℎ𝑐𝑔 (15)
𝑅

In the situation shown in Fig 6, it is assumed that the rollo-

ver starts when the vehicle's internal wheel is on the verge of 3 RESULTS
losing contact with the ground. Therefore, in order to perform The vehicle's rollover begins when its wheel inside the curve is
the rollover-related analyzes, the equations corresponding to on the verge of losing contact with the ground. Therefore, the
the normal forces of the prototype must be determined. Ini- magnitude of the normal force of the wheel is compared with
tially, the situation of static equilibrium is assumed. Thus, the the term 𝑧𝑚1 . 𝑘𝑡 or 𝑧𝑚2 . 𝑘𝑡 (depending on the direction of the
prototype (3 and 5 DOF models) is subject to the normal forces curve), where the prior are shown in red lines and the last one
𝑁1 , 𝑁2 e 𝑁3 (one for each wheel), which are equivalent to the in blue line in the plots of Fig 9 and Fig 10. The results, for circuit
weight force (𝑃), found multiplying the total mass of the proto- loop with a constant speed of 7 m/s, for the 3 DOF model can
type (𝑀𝑡 ) by the gravitational constant (𝑔). Applying the sum of be seen in Fig 9 and Fig 10. Analyzing the graphics, at the points
forces in 𝑧 and the sum of moments in 𝑥 and 𝑦, the following where the blue line exceeds the red line, the tire loses contact
equations are found: with the ground, therefore, there are great chances of overturn-
ing. As the model under analysis does not have parameters for
𝑀𝑡 . 𝑔 𝑀𝑡 . 𝑔. 𝑙𝑓 damping and stiffness, it is noted that at certain points the tire
𝑁1 = − (16)
2 2. (𝑙𝑟 + 𝑙𝑓 ) force was higher than the normal force, generating critical rid-
𝑀𝑡 . 𝑔. 𝑙𝑓 ing situations.
𝑀𝑡 . 𝑔 (8)
𝑁2 = − (17)
2 2. (𝑙𝑟 + 𝑙𝑓 )

IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 119
ISSN 2229-5518

TABLE 1 At instant t = 23s, the left wheel loses its normal force at a
PARAMETERS OF AN ULTRA-EFFICIENT VIRTUAL MODEL PROTO- value of 933N (Fig 9), while the normal for the static condition
TYPE. is 235N (red line). This is equivalent to a value close to four
times the allowed limit for the prototype to be able to execute
the curve without losing the contact of the tire with the ground.
The situation was even more worrying in the analyzes of the
prototype's right wheel (Fig 10), as in t = 111s, when the wheel
loses its normal force to a maximum value of 1,024N.

Fig 9. Left Wheel Rollover Analysis of the 3 Degrees of Freedom

Model.

IJSER
Fig 10. Right Wheel Rollover Analysis of the 3 Degrees of Free-
dom Model.

Analogously to the previous case, the possibility of over-

turning for the 5 DOF model with suspension system through-
Fig 7. Computational Mapping of the Track.
out the circuit was evaluated. The graphics of tire force versus
normal force are shown in Fig 11 and Fig 12. As in the previous
case, the 5 DOF model also provides critical riding situations.
Operating under the same initial conditions previously men-
tioned, there are times when the wheel would lose contact with
the ground, which can result in eventual overturns. However,
it is noted that the intensity reached by the tire force was re-
duced considerably when compared to the 3 DOF model. This
can be explained by the insertion of a suspension system in the
front wheels of the prototype, which has the purpose of reduc-
ing the magnitude of this force. In the analysis of the left wheel
(𝑧𝑚1 ), which can be seen in Fig 11, the magnitude of 283N was
obtained as the most critical value of the force applied to the
tire, at the time of approximately t = 29s. Regarding the right
Fig 8. Ultra-Efficient Prototype CAD Model with Driver. wheel (𝑧𝑚2 ), shown in Fig 12, the maximum intensity of the
force reached on the tire was 478N, at time t = 113s.

IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 120
ISSN 2229-5518

minimum average speed to be adopted by the prototype is ap-

proximately 7 m/s. Therefore, this will be the reference speed
(𝑉𝑟 ). Through the simulations, it is estimated the maximum
speed at which both models are able to execute each of the cir-
cuit curves (𝑉𝑛 ) without overturning. The approximate values
of the speeds and parameters of the 3 and 5 DOF models are
shown in the table 2.

Fig 11. Left Wheel Rollover Analysis of the 5 Degrees of Free-

dom Model.

Fig 13. Curves Segmentation of the Sonoma Raceway.

IJSER
Once the variation of the kinetic energy in each of the curves
was calculated for both models, the sum of all values is deter-
Fig 12. Right Wheel Rollover Analysis of the 5 Degrees of Free- mined. Thus, the total energy dissipated in one circuit is ob-
dom Model. tained in Table 3. With the results, it is noted that the 5 DOF
model presents an energy saving of almost 40% referring to the
It is important to emphasize that the main gain of the pro- braking at the entrance of each of the curves shown in Fig 13,
totype with the inclusion of a front suspension system consists when compared to the 3 DOF model. Thus, it is fair to affirm
in the decrease of the magnitude of the force applied on the tire, that the addition of a front suspension system in ultra-efficient
which was reduced by more than 50% in the critical sections of automotive prototypes directly contributes to its energy effi-
the track when compared to the previous model. This allows ciency in curves.
the prototype to drive the curves faster, without the need for
drastic braking, which implies a lower speed resumption at the TABLE 2
PARAMETERS OF 3 AND 5 DOF MODELS.
exit of the curve and, consequently, makes the vehicle more ef-
ficient.

To estimate the energy savings related to the addition of a

front suspension system in the prototype, the energy expendi-
ture of the braking performed at the entrance of each curve is
calculated for both models. Initially, the Shell Eco-marathon cir-
cuit is segmented into four main curves, which are shown in the
figure below.

In order to calculate the dissipated energy as a function of

the braking on each stretch indicated in Fig 13, the kinetic en-
ergy variation equation is used:

𝑀𝑡 . (𝑉𝑟 ² − 𝑉𝑛 ²)
𝐸𝑘 = (19)
2

SEM regulations require vehicles to complete the 8 laps of

the competition in a maximum of 24 minutes. Thus, based on
the total length of the circuit (1,164 m), it is calculated that the

IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 121
ISSN 2229-5518

TABLE 3
DISSIPATED KINETIC ENERGY OF 3 AND 5 DOF MODELS.
REFERENCES
[1] F. Arpino, G. Cortellessa, A. Frattolillo, F. Iannetta and M. Scungio, “Numerical
and experimental investigation of the flow over a car prototype for the Shell
Eco Marathon,” Journal of Applied Fluid Mechanics, vol. 12, no. 1, pp. 207-218,
2019.
[2] J.C. Páscoa, F.P. Brójo, F.C. Santos and P.O. Fael, “An innovative experimental
on-road testing method and its demonstration on a prototype vehicle,” Journal
of Mechanical Science and Technology, vol. 26, no. 6, pp. 1663-1670, 2012.
[3] M. Targosz, W. Skarka and P. Przystalka, “Model-Based Optimization of Ve-
locity Strategy for Lightweight Electric Racing Cars,” Journal of Advanced Trans-
portation, pp. 1-20, 2018.
[4] M. Fabian, M. Puškár, R. Boslai, M. Kopas, S. Kender and R. Huňady, “Design
of experimental vehicle specified for competition Shell Eco-marathon 2017 ac-
cording to principles of car body digitisation based on views in 2D using the
intuitive tool Imagine & Shape CATIA V5,” Advances in Engineering Software,
vol. 115, pp. 413–428, 2018.
[5] C.J.J. Paredis, C. Bishop and D.L. Bodner, “Systems Engineering (LSE): hands-
4 CONCLUSION on experiences in applying LSE to a student eco-car build project,” Procedia
Computer Science, vol. 16, pp. 492-501, 2013.
The present work sought to carry out the dynamic modeling of
[6] V.D. Nguyen and P. Martin, “Product design-process selection-process plan-

IJSER
an ultra-efficient automotive prototype, taking as a reference
ning integration based on modeling and simulation,” Int J Adv Manuf Technol,
the vehicles designed by UFSC's Energy Efficiency team. Based
vol. 77, pp. 187-201, 2015.
on the dynamic model of the vehicle, the study extended to the
[7] Shell, “Shell Eco-marathon History,” Shell, http://www.shell.com/energy-
physical and mathematical analysis (theoretical) of the inclu-
and-innovation/shell-ecomarathon.html, 2016.
sion of a front suspension system in the prototype, with the ob-
[8] J.M.G. Rebelo and M.A.R. Silvestre, “Development of a Coreless Permanent
jective of improving its performance in curves. Thus, the need
Magnet Synchronous Motor for a Battery Electric Shell Eco Marathon Proto-
for drastic braking is minimized, which implies a lower speed
type Vehicle,” Open Eng., vol. 8, pp. 382–390, 2018.
resumption at the exit of the curve and, consequently, makes
[9] J.C. Olivier, G. Wasselynck, S. Chevalier, B. Auvity, C. Josset, D. Trichet, G.
the vehicle more efficient. For conducting the track simulations,
Squadrito and N. Bernard, “Multiphysics modeling and optimization of the
the official Shell Eco-marathon circuit (adapted from Sonoma
driving strategy of a light duty fuel cell vehicle,” International Journal of Hydrogen
Raceway, CA) was mapped.
Energy, vol. 42, pp. 26943-26955, 2017.
[10] S. Jan, “Minimizing the fuel consumption of a vehicle from the shell eco-mara-
Among all the evaluations performed, the results regarding
thon: a numerical study,” ESAIM: Control, Optimisation and Calculus of Varia-
the vehicle's rollover were more satisfactory. Through the
tions, vol. 19, pp. 516-532, 2013.
graphics, it was possible to notice a significant reduction in the
[11] Z. Czyż, P. Karpiński and S. Koçak, “Numerical analysis of the influence of par-
force applied to the tire, when the prototype had a front sus-
ticular parts of the high efficient electric vehicle on the aerodynamic forces,”
pension system. The analysis of rollover possibility during the
Advances in Science and Technology Research Journal, vol. 13, no. 4, pp. 1-7, 2019.
circuit were shown higher for the 3 DOF model (without sus-
[12] A. Messana, L. Sisca, A. Ferraris, A.G. Airale, H.C. Pinheiro, P. Sanfil-
pension) and lower for the 5 DOF model (with suspension). In
ippo, and M. Carello, “From design to manufacture of a carbon fiber
terms of energy efficiency, the 5 DOF model has an energy sav-
monocoque for a three-wheeler vehicle prototype,” Materials, vol. 12,
ing of almost 40% referring to the braking at the entrance of
no. 332, pp. 1-11, 2019.
each curve, when compared to the 3 DOF model, proving that
[13] J. Sawulski and M. Ławrynczuk, “Optimization of control strategy for
the addition of a front suspension system in ultra-efficient pro-
a low fuel consumption vehicle engine,” Information Sciences, vol. 493,
totypes directly contributes to its energy efficiency in curves.
pp. 192–216, 2019.
Therefore, it is safe to affirm that the study of the dynamic of
[14] S.M.H.S. Omar, N.M. Arshad, I.M. Yassin, M.H.A.M. Fakharuzi and
ultra-efficient prototypes is very important for the understand-
T.A. Ward, “Design and optimization of powertrain system for proto-
ing of all critical driving situations that can affect its perfor-
type fuel cell electric vehicle,” Journal of Mechanical Engineering and Sci-
mance. It is possible to make in-depth analysis and develop spe-
ences, vol. 8, pp. 1401-1413, 2015.
cific systems that will improve the performance of the vehicle
[15] M. Omer, “Design and Analysis of Composite Chassis for Shell Eco
dynamics.
Marathon Prototype Vehicle 2016 using Finite Element Analysis,” Ger-
man University of Technology in Oman, Oman, 2015.
[16] J.J. Santin et al “The World’s Most Fuel-Efficient Vehicle: Design and
5 ACKNOWLEDGMENTS Development of PAC-CAR II,” Zürich: vdf Hochschulverlag Ag an der
ETH Zürich, 2007.
The authors thank UFSC Eficem group for provision of proto-
type data and competition information.
IJSER © 2021
http://www.ijser.org
International Journal of Scientific & Engineering Research Volume 12, Issue 1, January-2021 122
ISSN 2229-5518

[17] B. Fréchède, A.S. McIntosh, R. Grzebieta, M.R. Bambach, “Character-

istics of single vehicle rollover fatalities in three Australian states
(2000–2007),” Accident Analysis and Prevention, vol. 43, pp. 804-812,
2011.
[18] E.Ch. Tsirogiannis, G.E. Stavroulakis and S.S. Makridis, “Electric Car
Chassis for Shell Eco Marathon Competition: Design, Modelling and
Finite Element Analysis,” World Electric Vehicle Journal, vol. 10, no.8,
pp. 1-13, 2019.
[19] J. Kral, M. Palko, M. Palko and L. Pavlikova, “Design and develop-
ment of ultra-light front and rear axle of experimental vehicle,” Open
Eng., vol. 10, pp. 232–237, 2020.
[20] K. Jezierska-Krupa, M. Krysiak and W. Skarka, “Advanced driver
safety support systems for the urban type vehicle,” Transport Problems,
vol. 10, no. 4, pp. 138-148, 2015.
[21] K. Jezierska-Krupa and W. Skarka, “Design method of ADAS for ur-
ban electric vehicle based on virtual prototyping,” Journal of Advanced
Transportation, pp. 1-18, 2018.
[22] M.R. Bambach, R.H. Grzebieta and A.S. McIntosh, “Thoracic injuries
to contained and restrained occupants in single-vehicle pure rollover
crashes,” Accident Analysis and Prevention, vol. 50, pp. 115-121, 2013.
[23] B.D. Carlsen and O. Oma, “Development, Building and Competing
with a New Car in Shell Eco-marathon,” Master of Science in Mechanical
Engineering, Norwegian University of Science and Technology, 81p,

IJSER
Jul, 2017.
[24] J. Reimpell, H. Stoll and J.W. Betzler, “The Automotive Chassis: Engi-
neering Principles,” Butterworth-Heinemann, 2. ed, 2001.
[25] T.D. Gillespie, “Fundamentals of Vehicle Dynamics,” Society of Auto-
motive Engineers Inc, 1992.
[26] M.P. Czechowicz and G. Mavros, “Analysis of Vehicle Rollover Dy-
namics Using a Highfidelity Model,” Vehicle System Dynamics, vol. 52,
no. 5, pp. 608-636, 2014.
[27] S-K Chen, N. Moshchuk, F. Nardi, and J. Ryu, “Vehicle rollover avoid-
ance,” IEEE Control Systems Magazine, pp. 70-85, 2010.
[28] G. Genta and L. Morello, “The Automotive Chassis: Volume 2 System
Desing,” Berlin: Springer, Mechanical Engineering Series, 2009.

IJSER © 2021
http://www.ijser.org