ArticlePDF Available

Increasing the on-road fuel economy by trailing at a safe distance

Authors:

Abstract and Figures

Energy is a driving force for automotive applications. Reducing the energy demand of the vehicle is one method of increasing the fuel economy of a vehicle. Heavy-duty commercial vehicles have large frontal areas that provide large amounts of aerodynamic drag at highway speeds. Reducing the aerodynamic drag lowers the engine demand and therefore increases the fuel economy of the vehicle. This study tested the fuel economy and the front air velocity of a 10.7 m box truck trailing another box truck by distances of 3.1 times the truck length, 4.7 times the truck length, and 6.3 times the truck length at a highway speed of 28 m/s. The distance of 6.3 times the vehicle length was considered 'safe' for trailing another vehicle, whereas the distances of 3.1 times the truck length and 4.7 times the truck length were not considered safe by the United States Fire Administration. The results showed significant reductions in the air velocity in front of the trailing vehicle of 8.5%, 6.5%, and 3.8% for trailing distances of 3.1 times the vehicle length, 4.7 times the vehicle length, and 6.3 times the vehicle length respectively. The fuel economy of the trailing truck increased significantly by 7.4-8.0%, 8.2-9.0%, and 6.5%-7.7%, for trailing distances of 3.1 times the vehicle length, 4.7 times the vehicle length, and 6.3 times the vehicle length respectively. Based on a road load analysis, these fuel economy improvements indicated a reduction in the drag coefficient of the trailing vehicle of 8-10%. Therefore, a box truck trailing another box truck at a safe distance results in a reduction in the aerodynamics drag and a significant increase in the fuel economy.
Content may be subject to copyright.
Automotive Aerodynamics Special Issue
Proc IMechE Part D:
J Automobile Engineering
1–9
ÓIMechE 2017
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0954407017703233
journals.sagepub.com/home/pid
Increasing the on-road fuel economy
by trailing at a safe distance
John Nuszkowski, Harlan Smith, Michael McKinney, Nicholas McMahan,
Benjamin Wilder, Eric Boehringer, Blair Clarkson, Cutler Littleton and
Kyle Parker
Abstract
Energy is a driving force for automotive applications. Reducing the energy demand of the vehicle is one method of
increasing the fuel economy of a vehicle. Heavy-duty commercial vehicles have large frontal areas that provide large
amounts of aerodynamic drag at highway speeds. Reducing the aerodynamic drag lowers the engine demand and there-
fore increases the fuel economy of the vehicle. This study tested the fuel economy and the front air velocity of a 10.7 m
box truck trailing another box truck by distances of 3.1 times the truck length, 4.7 times the truck length, and 6.3 times
the truck length at a highway speed of 28 m/s. The distance of 6.3 times the vehicle length was considered ‘safe’ for trail-
ing another vehicle, whereas the distances of 3.1 times the truck length and 4.7 times the truck length were not consid-
ered safe by the United States Fire Administration. The results showed significant reductions in the air velocity in front
of the trailing vehicle of 8.5%, 6.5%, and 3.8% for trailing distances of 3.1 times the vehicle length, 4.7 times the vehicle
length, and 6.3 times the vehicle length respectively. The fuel economy of the trailing truck increased significantly by 7.4–
8.0%, 8.2–9.0%, and 6.5%–7.7%, for trailing distances of 3.1 times the vehicle length, 4.7 times the vehicle length, and 6.3
times the vehicle length respectively. Based on a road load analysis, these fuel economy improvements indicated a reduc-
tion in the drag coefficient of the trailing vehicle of 8–10%. Therefore, a box truck trailing another box truck at a safe
distance results in a reduction in the aerodynamics drag and a significant increase in the fuel economy.
Keywords
Commercial vehicles, aerodynamic drag, platooning, fuel efficiency, fuel economy, intelligent vehicles
Date received: 13 June 2016; accepted: 9 February 2017
Introduction
Aerodynamic drag is a well-known and well-
documented factor in the overall performance of motor
vehicles. With respect to heavy-duty commercial vehi-
cles with large frontal areas, the air drag can signifi-
cantly affect the fuel consumption of a vehicle. For
example, the air drag of a class 8 tractor-trailer with a
gross vehicle weight of 36,287 kg consumes approxi-
mately 22% of the fuel energy and 53% of the engine
power.
1
If the amount of aerodynamic drag experi-
enced by the vehicle can be minimized, then the engine
load can theoretically be reduced, thus leading to an
improvement in the fuel economy. A wide variety of
research and development studies have been carried
out to improve the vehicle aerodynamics of medium-
duty to heavy-duty vehicles.
2–10
Some of these efforts
have led to the use of cab-roof-mounted deflectors, cab
forebody fairings, and trailer skirts. Each of these
methods has shown to improve the aerodynamics of
heavy-duty trucks, but there is an increasing demand
for further minimization of the drag on these vehicles.
One technique to reduce the fuel economy that has
been studied is vehicle platoons. Commonly referred to
as ‘platooning’, this method involves two or more vehi-
cles driving in tandem with each other at very close dis-
tances in an attempt to reduce the coefficient of drag
on the trailing vehicle(s). Many studies have explored
the effectiveness of platooning as it applies to computa-
tional fluid dynamics (CFD) testing,
11–15
wind tunnel
testing,
11–13,16–18
as well as on-road vehicle testing.
19–21
With the advent of autonomous vehicle control, vehicle
platooning may become an accepted strategy owing to
University of North Florida, Jacksonville, Florida, USA
Corresponding author:
John Nuszkowski, School of Engineering, University of North Florida, 1
UNF Drive, Jacksonville, FL 32224-7699, USA.
Email: john.nuszkowski@unf.edu
increased traffic flow
22
and fuel economy.
20,21
The
inter-vehicle spacing can be reduced with computer-
controlled driving because it involves faster reaction
times than human-controlled driving does.
22–24
A large
number of these studies have focused on close inter-
vehicle spacings (less than 1L, where Lis the length of
the vehicle or truck), which is only practical for auton-
omous vehicle control.
Zabat et al.
17,18
used wind tunnel testing to investi-
gate minivan platoons. At an inter-vehicle spacing of
0.2L, the coefficient of air drag was reduced by 56%.
Aerodynamic platoon road tests
19
have shown that, for
vehicles driven in platoons of two, three, and four cars
with inter-vehicle spacings of between 0.5Land 1.2Lat
a speed of 26.8 m/s, the interior vehicles experienced a
fuel economy benefit of about 10%, and the trailing
vehicles experienced a savings of about 7%.
The aerodynamic drag on the trailing vehicle in a
platoon varies not only as a result of the trailing dis-
tance but also as a result of the shape or ‘form’ of the
vehicle.
12
When platooning, the rear slant angle of the
leading vehicle combined with the inter-vehicle spacing
may cause the trailing vehicles to have a higher drag
coefficient.
11,16
In the studies by Rajamani,
11
Pagliarella et al.,
12
and Watkins and Vino,
16
it was
established that the coefficient of drag on the trailing
vehicle increased because of the sharp backlight angles.
Therefore, it is important to note that heavy-duty
trucks do not have these sharp rear angles.
By means of a CFD study, Schito and Braghin
13
found that vehicles with high drag coefficients, when
isolated, have a better performance when operating in a
platoon. These findings indicate that heavy-duty trucks
perform better in a platoon configuration than sedans
and smaller compact cars do. Using CFD, Mihelic et
al.
15
investigated the platooning of tractor–trailers with
an inter-vehicle distance of 0.5L. Ellis et al.
14
showed
that platoons at close distances (less than 0.5L) need to
consider the significant loss of ram air engine cooling.
By simulations, Liang et al.
25
showed that a 7% fuel
savings can result from catching up to another heavy-
duty vehicle that is 10 km away and forming a platoon
on a 350 km trip.
Alam et al.
21
found, by simulations and on-road
experiments, a 4.7–7.7% reduction in the fuel for one
heavy-duty vehicle following another at different time
gaps (the time gaps were not given). In an on-road
study of semitrailer trucks by Bonnet and Fritz,
20
,they
showed that the fuel consumption was reduced by 15–
21% at 22 m/s and by 10–17% at 17 m/s for an inter-
vehicle spacing of less than 1L.
Vehicles (in particular, heavy-duty trucks) are not
typically driven with an inter-vehicle spacing shorter
than their own length. In addition, the necessary stop-
ping distances of heavy trucks are greatly increased at
high speeds. For these reasons, the United States Fire
Administration (USFA)
26
has set guidelines for driving
at safe distances which are dependent upon the vehicle
speeds. To be considered trailing at a ‘safe distance’, a
driver must maintain a distance of 1Lfor every 4.47 m/
s traveled. At safe distances traveling at 22.2 m/s, the
trucks should be no closer than 82.5 m according to the
guidelines. As noted above, the minimum trailing dis-
tance may be decreased with the use of autonomous
vehicle control in comparison with that using human
control.
Trucks in the study by Bonnet and Fritz
20
yielded a
reduction in the fuel consumption of up to 20% when
platooning at close distances. Fuel economy benefits
from a reduction in the aerodynamic drag may still
exist when traveling at larger inter-vehicle spacings
(greater than 1L), which are necessary for human-
controlled vehicles. The goal of this study was to inves-
tigate the change in the on-road fuel economy during
safe-distance human driving for a 10.7 m box truck
trailing another 10.7 m box truck at highway speeds.
Theory
The aerodynamic drag force which a vehicle experiences
is affected by the ambient air density r, the frontal area
Aof the vehicle, the drag coefficient C
d
of the vehicle,
and the square of the freestream air velocity v
r
accord-
ing to
FD=1
2rCdAv2
rð1Þ
For a vehicle trailing closely behind a leading vehicle, a
change in the drag coefficient and/or the freestream air
velocity may occur, whereas the air density and the
frontal area can be assumed to be constant.
The on-road forces experienced by the vehicle along
the vehicle’s forward direction include the inertia of the
vehicle mass due to the vehicle mass mand the vehicle
acceleration a, the rolling resistance from the tires due
to the acceleration gdue to gravity, the rolling resis-
tance mof the tires, the road grade udue to the varying
terrain altitude, and the drag force F
D
according to
27
FR=ma +mmg+mmgsin u+FDð2Þ
The rate of energy required is proportional to the
product of the on-road force to move the vehicle and
the velocity v
v
of the vehicle, as expressed by
_
ER=FRvvð3Þ
This energy comes from the vehicle’s engine and origi-
nates from the energy content of the fuel (i.e. the lower
heating value (LHV)); it is proportional to the mass
flow rate _
mFof the fuel, the losses h
a
due to auxiliary
devices, the overall transmission efficiency h
g
, and the
engine efficiency h
e
, as given by
_
ER=hghahe
_
mFLHV ð4Þ
2Proc IMechE Part D: J Automobile Engineering
Experimental setup
The procedure for measuring the vehicle-to-vehicle fuel
economy during safe-distance driving took place in two
phases. The first phase of experiments was performed
using one class 6 Freightliner Business Class M2 Ryder
Cummins diesel truck (Figure 1(a)) to establish a base-
line. This provides a means of comparing the fuel econ-
omy of a single vehicle with the fuel economy of a
trailing vehicle traveling at a safe distance. Upon com-
pletion of the experiments on the first vehicle, an identi-
cal Freightliner Business Class M2 Ryder diesel truck
was incorporated as the lead vehicle for safe-distance
testing along the same route (Figure 1(b)). The box
truck had a curb vehicle weight of 7911 kg with a fron-
tal area of 10.1 m
2
.
The testing route was from outside the University of
North Florida (UNF) campus merging on to I-295 S
headed south and then exiting at US-1 and returning
on I-295 N headed north back to UNF (Figure 2). The
road terrain consisted of a flat road with overpasses.
The route from start to finish was approximately 31
km, and the speed limit for I-295 was 29 m/s; however,
the testing speed was 28 m/s because of the difference
between the vehicle’s speedometer display and the
actual vehicle speed.
The recommendation for safe-distance driving, as
provided by the USFA,
26
is 1Lfor every 4.47 m/s that a
vehicle is traveling. This recommendation was the basis
for determining a safe trailing distance for the safe-
distance driving experiments. To observe any possible
trends in the data, distances shorter than those consid-
ered ‘safe’ were implemented. The truck length was
measured at 10.7 m. As a result, the required trailing
distances for the route were 3.1L(33 m), 4.7L(50 m),
and 6.3L(67 m) respectively. For repeatability, the sin-
gle truck was tested six times, and each trailing distance
was tested five times, on the route. The testing was con-
ducted over five consecutive nights from 9 pm to 5 am
with clear skies or partly cloudy skies. The ranges for
the ambient temperature, the humidity, the pressure,
and the wind speed were 23.9–28.0 °C, 69–88%, 101.2–
101.5 kPa, and 0.0–4.3 m/s respectively.
For all experimental testing, the fuel weight was
measured from an auxiliary fuel tank, which was
mounted on top of a Brecknell bench base model
3700LP fuel scale base and a Brecknell digital fuel indi-
cator model 200ES located in the rear truck box. This
allowed the fuel weight to be measured on board in real
time. The Dearborn Group’s ‘Dearborn Protocol 5’
data link monitor was installed and connected directly
to a laptop to collect the engine control unit (ECU)
data (primarily, the ECU fuel flow rate). To guarantee
an appropriate trailing distance, an MDL Laser
Systems ILM 150R range finder was mounted on to
the back of the leading vehicle. The range finder was
used to record the distance between the two trucks on
the on-board laptop as well as to give an indication to
the driver of the trailing distance. In addition, two
Figure 1. Truck tests: (a) single truck of length L: (b) leading truck and trailing truck.
Figure 2. Vehicle testing route.
28
Nuszkowski et al. 3
anemometers were installed on the trailing truck
(Figure 3) to investigate any velocity field changes. One
anemometer was installed 0.9 m in front of the truck’s
bumper and 4.3 m from the ground. The boom for this
anemometer was capable of extending along the hori-
zontal axis and the vertical axis of the truck to allow
the anemometer to be adjusted outside the vehicle’s
flow field. The other anemometer was installed 0.9 m
in front of the truck’s bumper and 2.5 m from the
ground. Both anemometers were RM Young 85000
ultrasonic anemometers which measure the wind velo-
city in two dimensions with an accuracy of 62% at up
to 30 m/s.
All the sensors were connected to an on-board lap-
top for real-time collection of data at a collection rate
of 10 Hz throughout each test run. During testing, a
wireless radio network was created between the two
vehicles. Two US GlobalSat BU 353 Global
Positioning System (GPS) receivers were used to
acquire the GPS velocities and locations between the
vehicles. The lead vehicle received the GPS information
from the trailing vehicle; in order to maintain consis-
tency with control tests, the trailing test truck applied
cruise control at its assigned speed, and the leading
truck controlled the trailing distance.
Results
The cruise control of the vehicle kept the vehicle speed
within 60.5 m/s at highway speeds (Figure 4). A test
from each set (single truck; truck with a trailing dis-
tance of 3.1L; truck with a trailing distance of 4.7L;
truck with a trailing distance of 6.3L) is shown in the
figure. The vehicle traveled southbound, exited,
re-entered traveling northbound, and returned to the
starting location. It should be noted that the figure
depicts the vehicle speed from the GPS sensor.
The distance between the leading truck and the trail-
ing truck (Figure 5) was controlled by the driver of the
front truck. The set distance from the laser range finder
was controlled to within 65 m at highway speeds.
Spikes occurred in the data when there was a curve in
the road or when the leading truck left the highway
and the laser missed the trailing truck.
As mentioned with respect to the experimental setup,
two anemometers were placed at the front of the trail-
ing truck. Because of the different placements of the
two anemometers, the air velocities were different
(Figure 6). The freestream anemometer was placed suf-
ficiently far in front of and above the vehicle to read
the freestream air velocity and shows an air velocity
close to the value of 28 m/s at which the vehicle was tra-
veling. The freestream air velocity and the vehicle speed
would be identical if there were no other effects such as
the ambient wind, the influence of other vehicles, and
Figure 3. Anemometer locations on the trailing truck (scaled
drawing).
Figure 4. Vehicle speed from a single test run at each trailing
distance for the trailing truck.
Figure 5. Trailing distance between the leading vehicle and the
trailing vehicle from a single test run.
Figure 6. Anemometer air velocities for the first single-truck
test.
4Proc IMechE Part D: J Automobile Engineering
the obstructions along the road. The front anemometer
was influenced by the flow field of the trailing truck
and was well below the vehicle speed.
Figure 7 and Figure 8 show the continuous air velo-
city for the freestream and the front of the trailing truck
respectively for each of the trailing distances from a sin-
gle test run. Little difference was noticed between the
continuous freestream air velocity at different trailing dis-
tances and that for the single-truck runs (Figure 7). For
the 4.7Ltest, the influence of a cross-wind can be seen
during 1–7 km and then in the opposite direction for 24–
30 km. By operating the vehicle in both directions, the
influence of the cross-wind on the fuel economy was
minimized. The air velocity at the front of the truck was
lower for the three different trailing distances than for
the single-truck test, thereby demonstrating that the lead-
ing truck influences the air stream of the trailing truck
even at trailing distances of 3.1L–6.3L(Figure 8).
To investigate the overall influence of the leading
truck, the mean vehicle speed, the fuel economy, the air
velocity, and the trailing distance were calculated when
the trailing truck was in cruise control and reached the
set cruise control speed of 28 m/s. For each test run, a
mean value was calculated for the southbound highway
cruise speed section (approximately 460 s in length)
and a mean value was calculated for the northbound
cruise speed section (approximately 470 s in length).
These two values were then averaged to obtain the
mean value of each test run. For each set of tests (single
truck; truck with a trailing distance of 3.1L; truck with
a trailing distance of 4.7L; truck with a trailing distance
of 6.3L), the mean
x, the standard deviation s, and the
coefficient of variation (CV) were calculated. The indi-
vidual test results are provided in Appendix 1. The
mean, the standard deviation, and the CV for each set
of tests and the percentage difference compared with
the single truck are shown in Table 1. The CVs were
0.3–1.1%, 0.3–2.2%, 0.4–1.5%, and 1.4–3.2% for the
vehicle speed, the fuel economy, the air velocity, and
the distance respectively, which provides an indication
of test-to-test repeatability. A Student’s ttest was per-
formed in which the trailing truck was compared with
the single truck. Significant differences were determined
with a Student’s ttest at a pvalue less than 0.05, and
the percentage differences which did not show a signifi-
cant difference are noted in Table 1. The high values
for fuel economy shown in Appendix 1 and Table 1
occur because the vehicle was tested at the curb vehicle
weight with the cruise control setting engaged. Vehicles
with heavy-duty (class 8) applications are not typically
operated under the vehicle cargo loading presented in
this study. Further testing would be needed to deter-
mine the differences in the fuel economy for these high-
cargo-load applications.
With trailing distances of 3.1L, 4.7L, and 6.3L, the
fuel economy from the fuel scale showed significant
increases of 8.0%, 9.0%, and 7.7% respectively. In
addition, the increases in the fuel economy measured
by the ECU correlated with the fuel scale. The fuel
economy increases ranged from 7.7% to 9.0% with the
fuel scale, and from 6.5% to 8.2% with the ECU. The
air velocity in front of the trailing truck showed signifi-
cantly lower air velocities by 8.5%, 6.5%, and 3.8% for
trailing distances of 3.1L, 4.7L, and 6.3Lrespectively
than without the leading truck. It should be noted that
the air velocity in front of the trailing truck showed sig-
nificant differences when different trailing distances are
compared with each other, whereas the fuel economy
showed no significant difference. This arose because
the test-to-test variability showed a CV of 0.3–2.2% for
the fuel economy, whereas the differences between the
fuel economy values for different trailing distances were
only 0.3–1.7%.
Discussion
Any reduction in the drag coefficient which resulted
because the box truck trailed the second box truck
reduces the road load from the aerodynamic drag. The
total road load power is reduced owing to the reduction
in the aerodynamic drag and, as shown in equation (4),
the resulting fuel consumption is be reduced. The results
of Bonnet and Fritz
20
showed fuel economy savings of
Figure 7. Freestream air velocity from a single test run at each
trailing distance for the trailing truck.
Figure 8. Front air velocity from a single test run at each
trailing distance for the trailing truck.
Nuszkowski et al. 5
up to 21% when trailing at less than 1L. The fuel econ-
omy data collected from the box diesel truck in this
study showed benefits of 6.5–7.7% when traveling at a
highway speed of 28 m/s for a trailing distance of 6.3L.
Figure 9 shows the rate of energy, using equation (2)
and equation (3), for each of the road loads for the
route tested. The values for the drag coefficient and the
tire rolling resistance were derived experimentally from
coastdown testing.
29
During the cruise control section,
the highway vehicle speed caused the majority (about
85%) of the vehicle work to occur from the aerody-
namic drag, with the remainder being from the rolling
resistance. Significant vehicle inertia loading occurred
as the vehicle was slowing down or accelerating owing
to the on ramps and the off ramps but, during the
cruise control sections, the integrated inertia loading
was negligible. In addition, although there was signifi-
cant instantaneous road power from the road grade as
the vehicle traveled on overpasses, the integrated vehi-
cle work from the road grade was negligible during the
cruise control sections.
For the curb vehicle weight tested, a reduction in the
aerodynamic drag coefficient of 8–10% results in a
reduction in the vehicle work of 6.5–9.0% (Figure 10).
On the assumption that the losses from the auxiliaries,
the fuel conversion efficiency, and the transmission effi-
ciency (equation (4)) remain constant as the road power
changes, the fuel consumption should be decreased by
approximately the same amount as the vehicle work.
As shown in Table 1, the fuel economy improvements
ranged from 6.5% to 9.0%, which indicates a reduction
in the aerodynamic drag coefficient from 8% to 10%
for trailing the leading box truck at safe distances. It
should be noted that this analysis is only valid at the
vehicle speed tested.
Table 1. Values of the mean, the standard deviation, and the coefficient of variation for the steady-state results and the percentage
differences with respect to the single truck.
Trailing distance Parameter Vehicle speed,
ECU
(m/s)
Fuel economy Air velocity Trailing distance
(range finder)
(m)
Fuel scale
(km/l)
ECU
(km/l)
Freestream
(m/s)
Front
(m/s)
Single truck
x28.4 4.61 4.80 26.4 22.4 —
s 0.1 0.08 0.07 0.2 0.3
CV 0.4% 1.7% 1.4% 0.8% 1.2% —
3.1L
x28.5 4.98 5.16 26.0 20.5 33.9
s 0.3 0.11 0.09 0.2 0.2 0.6
CV 1.1% 2.2% 1.7% 0.7% 1.2% 1.8%
Difference 0.3%
a
8.0% 7.4% –1.5% –8.5% —
4.7L
x28.4 5.02 5.20 26.2 21.0 50.1
s 0.1 0.07 0.06 0.3 0.3 0.7
CV 0.3% 1.3% 1.1% 1.0% 1.5% 1.4%
Difference –0.1%
a
9.0% 8.2% –0.9%
a
–6.5% —
6.3L
x28.4 4.96 5.11 26.4 21.6 69.2
s 0.2 0.04 0.02 0.1 0.1 2.2
CV 0.5% 0.7% 0.3% 0.4% 0.7% 3.2%
Difference 0.1%
a
7.7% 6.5% –0.1%
a
–3.8% —
ECU: engine control unit; CV: coefficient of variation.
a
No significant difference at a 95% confidence level.
Figure 9. Rate of energy from the road loads for the first
single-truck test.
Aero: aerodynamic drag coefficient.
Figure 10. Influence of the reduction in the drag coefficient on
the road load power for the route and the vehicle speed tested
at the curb vehicle weight and the gross vehicle weight.
6Proc IMechE Part D: J Automobile Engineering
Conclusions
A 10.7 m box truck was tested at highway speeds,
using cruise control, when trailing another box truck
at distances of 3.1L,4.7L,and6.3Lrespectively
(where Lis the length of the vehicle or truck). The
trailing distance was controlled continuously to
within 65 m by the leading vehicle driver. In addi-
tion, the trailing box truck was tested without a lead-
ing truck to establish a baseline. The fuel economy,
the vehicle speed, the trailing distance, and the air
velocity in front of the trailing truck were collected
and analyzed for each test.
The results showed significant differences in the air
velocity directly in front of the box truck when trailing
the other box truck. The air velocity in front of the
trailing vehicle was significantly reduced by 8.5%,
6.5%, and 3.8% at trailing distances of 3.1L, 4.7L, and
6.3Lrespectively. The fuel economy showed significant
increases of 7.4–8.0%, 8.2–9.0%, and 6.5–7.7% when
trailing the other box truck at distances of 3.1L, 4.7L,
and 6.3Lrespectively. This indicated a reduction in the
aerodynamic drag of 8–10%,
The significant reduction in the flow velocity in front
of the trailing truck showed that the leading truck influ-
enced the flow field of the trailing truck at all the trail-
ing distances tested and therefore influenced the change
in the fuel economy. The trailing truck distance of 6.3L
is of particular interest since this is considered a safe
distance by the USFA
26
for the trailing truck to follow
the leading truck at 28 m/s.
Acknowledgements
The authors would like to acknowledge the University
of North Florida Foundation Board for their support
and Ryder Systems Inc. for the use of two box trucks.
In addition, the authors would like to acknowledge the
help of Samantha Delgado, Nick Squillace, and Skylar
Squillace during data collection.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest
with respect to the research, authorship, and/or publi-
cation of this article.
Funding
The author(s) received no financial support for the
research, authorship, and/or publication of this article.
References
1. Committee on Assessment of Technologies and
Approaches for Reducing the Fuel Consumption of
Medium- and Heavy-Duty Vehicles, Phase Two, Board
on Energy and Environmental Systems. Reducing the
fuel consumption and greenhouse gas emissions of
medium- and heavy-duty vehicles, phase two: first report.
Report, Division on Engineering and Physical Sciences,
Transportation Research Board, National Research
Council of the National Academies, Washington, DC,
USA, 2014.
2. Patten P, McAuliffe B, Mayda W and Tanguay B.
Review of aerodynamic drag reduction devices for heavy
trucks and buses. Technical Report CSTT-HVC-TR-205,
Centre for Surface Transportation Technology, National
Research Council Canada, Ottawa, Ontario, Canada, 11
May 2012.
3. Garry K. A review of commercial vehicle aerodynamic
drag reduction techniques. Proc IMechE Part D: J Auto-
mobile Engineering 1985; 199(3): 215–220.
4. Schaut N and Sengupta R. Aerodynamic optimization of
trailer add-on devices fully- and partially-skirted trailer
configurations. SAE paper 2015-01-2885, 2015.
5. Hoffmann F, Schmidt H, Nayeri C and Paschereit O.
Drag reduction using base flaps combined with vortex
generators and fluidic oscillators on a bluff body. SAE
paper 2015-01-2890, 2015.
6. Hirst T, Li C, Yang Y et al. Bluff body drag reduction
using passive flow control of jet boat tail. SAE paper
2015-01-2891, 2015.
7. Surcel M and Shetty M. The impact of design, position
and combination of aerodynamic devices on drag and
fuel consumption. SAE paper 2015-01-2894, 2015.
8. Salati L, Cheli F and Schito P. Heavy truck drag reduc-
tion obtained from devices installed on the trailer. SAE
paper 2015-01-2898, 2015.
9. Croll R, Gutierrez W, Hassan B et al. Experimental
investigation of the ground transportation systems
(GTS) project for heavy vehicle drag reduction. SAE
paper 960907, 1996.
10. Gutierrez W, Hassan B, Croll R and Rutledge W. Aero-
dynamics overview of the ground transportation systems
(GTS) project for heavy vehicle drag Reduction. SAE
paper 960906, 1996.
11. Rajamani G. CFD analysis of air flow interactions in
vehicle platoons. MS Thesis, School of Aerospace,
Mechanical and Manufacturing Engineering, Royal
Melbourne Institute of Technology, Melbourne, Vic-
toria, Australia, 2006.
12. Pagliarella R, Watkins S and Tempia A. Aerodynamic
performance of vehicles in platoons: the influence of
backlight angles. SAE paper 2007-01-1547, 2007.
13. Schito P and Braghin F. Numerical and experimental
investigation on vehicles in platoon. SAE paper 2012-01-
0175, 2012.
14. Ellis M, Gargoloff J and Sengupta R. Aerodynamic drag
and engine cooling effects on class 8 trucks in platooning
configurations. SAE paper 2015-01-2896, 2015.
15. Mihelic R, Smith J and Ellis M. Aerodynamic compari-
son of tractor–trailer platooning and A-train configura-
tion. SAE paper 2015-01-2897, 2015.
16. Watkins S and Vino G. The effect of vehicle spacing on
the aerodynamics of a representative car shape. JWind
Engng Ind Aerodynamics 2008: 96: 1232–1239.
17. Zabat M, Frascaroli S and Browand F. Drag measure-
ments on 2, 3 and 4 car platoons. SAE paper 940421, 1994.
18. Zabat M, Stabile N, Frascaroli S and Browand F. Drag
forces experienced by 2, 3 and 4-vehicle platoons at close
spacings. SAE paper 950632, 1995.
19. Michaelian M and Browand F. Quantifying platoon fuel
savings: 1999 field experiments. SAE paper 2001-01-1268,
2001.
Nuszkowski et al. 7
20. Bonnet C and Fritz H. Fuel consumption reduction in a
platoon: experimental results with two electronically
coupled trucks at close spacing. SAE paper 2000-01-
3056, 2000.
21. Alam A, Gattami A and Johansson K. An experimental
study on the fuel reduction potential of heavy duty vehi-
cle platooning. In: 2010 13th international IEEE confer-
ence on intelligent transportation systems, Funchal,
Portugal, 19–22 September 2010, pp. 306–311. New
York: IEEE.
22. Fernandes P and Nunes U. Platooning with IVC-enabled
autonomous vehicles: strategies to mitigate communica-
tion delays, improve safety and traffic flow. IEEE Trans
Intell Transpn Systems 2012; 13(1): 91–106.
23. Franke U, Bottiger F, Zomotor Z and Seeberger D.
Truck platooning in mixed traffic. In: 1995 IEEE intelli-
gent vehicles symposium, Detroit, Michigan, USA, 25–26
September 1995, pp. 1–6. New York: IEEE.
24. Gehring O and Fritz H. Practical results of a longitudinal
control concept for truck platooning with vehicle to
vehicle communication. In: 1997 international IEEE
conference on intelligent transportation systems, Boston,
Massachusetts, USA, 9–12 November 1997, pp. 117–122.
New York: IEEE.
25. Liang K, Ma
˚rtensson J and Johansson K. When is it fuel
efficient for a heavy duty vehicle to catch up with a pla-
toon? In: 7th IFAC symposium on advances in automotive
control, Tokyo, Japan, 4–7 September 2013, pp. 738–743.
Laxenburg: IFAC.
26. Klein L and Lane S. Emergency vehicle driver training.
Document FA 110, Federal Emergency Management
Agency, United States Fire Administration, Emmitsburg,
Maryland, USA, 1996.
27. Guzzella L and Sciarretta A. Vehicle propulsion systems:
introduction to modeling and optimization. 2nd edition.
Berlin: Springer, 2007, pp. 13–37.
28. GPS Visualizer. GPS Visualizer: do-it-yourself mapping,
http://www.gpsvisualizer.com (2003– 2016, accessed 18
December 2015).
29. White R and Korst H. The determination of vehicle drag
contributions from coast-down tests. SAE paper 720099,
1972.
Steady-state results for each test run.
Trailing distance Test Vehicle speed,
ECU
(m/s)
Fuel economy Air velocity Trailing distance
(range finder)
(m)
Fuel scale
(km/l)
ECU
(km/l)
Freestream
(m/s)
Front
(m/s)
Single truck 1 28.3 4.55 4.81 26.3 22.2
2 28.3 4.73 4.90 26.3 22.2 —
3 28.5 4.57 4.76 26.5 22.5 —
4 28.6 4.62 4.85 26.2 22.2 —
5 28.3 4.66 4.80 26.6 22.7 —
6 28.3 4.52 4.70 26.6 22.8 —
x28.4 4.61 4.80 26.4 22.4 —
s0.121 0.078 0.070 0.203 0.276 —
CV 0.4% 1.7% 1.4% 0.8% 1.2% —
3.1L1 28.1 4.87 5.11 25.9 20.5 34.0
2 28.5 4.92 5.11 26.0 20.6 33.2
3 28.9 5.00 5.10 26.1 20.5 33.2
4 28.6 4.94 5.16 26.2 20.8 34.6
5 28.3 5.16 5.31 25.8 20.1 34.2
x28.5 4.98 5.16 26.0 20.5 33.9
s0.310 0.112 0.088 0.175 0.240 0.623
CV 1.1% 2.2% 1.7% 0.7% 1.2% 1.8%
4.7L1 28.3 4.98 5.13 26.4 21.3 51.0
2 28.3 5.00 5.20 25.8 20.6 50.1
3 28.3 4.99 5.17 26.4 21.2 49.6
4 28.5 5.00 5.20 26.2 21.0 49.3
5 28.3 5.14 5.29 26.0 20.7 50.6
x28.4 5.02 5.20 26.2 21.0 50.1
s0.076 0.067 0.059 0.253 0.308 0.695
CV 0.3% 1.3% 1.1% 1.0% 1.5% 1.4%
6.3L1 28.3 4.93 5.10 26.3 21.5 70.3
2 28.3 5.01 5.10 26.4 21.7 66.7
3 28.3 4.97 5.12 26.3 21.5 70.2
4 28.6 4.92 5.14 26.3 21.4 67.1
5 28.6 4.98 5.11 26.6 21.7 71.9
x28.4 4.96 5.11 26.4 21.6 69.2
s0.153 0.037 0.017 0.117 0.141 2.248
CV 0.5% 0.7% 0.3% 0.4% 0.7% 3.2%
ECU: engine control unit; cV: coefficient of variation.
Appendix 1
8Proc IMechE Part D: J Automobile Engineering
Appendix 2
Notation
aacceleration of the vehicle
Afrontal area of the vehicle
C
d
aerodynamic drag coefficient
_
ERpower due to the road load
F
D
drag force
F
R
force due to the road load
gacceleration due to gravity
Llength of the vehicle
mmass of the vehicle
sstandard deviation of the sample
_
mFmass flow rate of fuel
v
r
velocity of the freestream air
v
v
velocity of the vehicle
xmean of the sample
h
a
losses due to auxiliary devices
h
e
efficiency of the engine
h
g
efficiency of the transmission
uroad grade
mrolling resistance of the tires
rdensity of air
Abbreviations
CFD computational fluid dynamics
CV coefficient of variation
ECU engine control unit
GPS Global Positioning System
LHV lower heating value
UNF University of North Florida
USFA United States Fire Administration
Nuszkowski et al. 9
... Passively blown ducks at the front of the car proved to be efficient in reducing the drag of the car situated in the wake of another car [17]. The distance between the two box trucks previously proved to notably influence their aerodynamic drag force and fuel economy [18], while the Ahmed car aerodynamic characteristics alter as well when following another car [19]. Nevertheless, the aerodynamically particularly sensitive race cars equipped with the rear wing have not been studied in that regard. ...
Article
A computational model was developed to study aerodynamic forces acting on two closed-wheel race cars in the slipstreaming (drafting) arrangement, i.e., when the cars are characterized by the same driving speed and situated one directly behind the other one. A particular focus was on the influence of a distance between the two cars on flow characteristics and aerodynamic loads experienced by the cars. Computational fluid dynamics (CFD) simulations were performed for a steady viscous fluid flow using the realizable k-ε turbulence model and non-equilibrium wall functions. The results indicate some important findings. The leading car experiences the smaller drag force coefficient for all studied distances when compared to the single car case. For larger distances between the cars, the drag force coefficient of the trailing car is generally larger than for a single car due to a complex turbulent wake flow of the leading car (drag bubble) that impinges the trailing car. The drag force coefficient of the trailing car decreases when decreasing a distance from the leading car. The modifications of the lift force coefficient are more pronounced for the trailing car as there is a decrease in the negative lift force (downforce). The downforce coefficient is significantly smaller for the trailing car than for the single car configuration at all studied distances. A slight decrease in the downforce coefficient is present for the leading car as well.
... However, the platooning of the longest allowed configuration could provide further savings. Regarding safe inter-vehicle distances, Nuszkowski et al. (2017) reported an 8% fuel reduction for the trailing vehicle with the platoon traveling at 100 km/h and a separation distance of 67.4 m. The authors deliberately choose such big distance to meet highway guidelines provided by the United States Fire Administration. ...
Conference Paper
Full-text available
This paper presents a review of theoretical and experimental works related to heavy-duty truck platooning. The platoon formation is characterized by the string of vehicles traveling with small separation distances. Only the first vehicle is driven by a human driver while the followers are assisted by the control system. The main goal of this approach is to reduce fuel consumption and greenhouse gas emissions. This is achieved because the proximity of the vehicles in this configuration provides a more efficient airflow around the set of trucks reducing the overall energy consumption. In addition, the platoon configuration allows the increase of road safety, transport capacity, driver comfort and reduce congestion and personnel cost due to the lack of human intervention in the following vehicles. The academic studies related to heavy-duty truck platooning are reviewed according to three research areas: fuel consumption in truck platoons, maintenance of platoon formation and truck coordination. In this paper, the state of the art in each of these areas is presented, the related publications are categorized according to their focus of contribution and potential future research directions are discussed.
Article
Platooning can reduce air drag when driving at high speed and improve aerodynamic characteristics. It can not only improve fuel economy, driving comfort, and safety of drivers but also reduce emissions and alleviate traffic congestion. At the same time, the energy recovery of vehicle driving is also in line with the general trend of international environmental governance. It has played an active role in reducing emissions and improving the ecological environment. In this paper, the effects of vehicle longitudinal spacing, number of vehicles and crosswind on the aerodynamic performance of longitudinal and side-by-side platoons are reviewed, as well as energy recovery. The study found that the three-vehicle platoon reduces drag by 30% compared to the two-vehicle platoon. Comparing the aerodynamic drag of the three-vehicle platoon with that of the single vehicle, the drag coefficient of the first and last vehicles can be reduced by 20%, and the drag coefficient of the middle vehicle can be reduced by about 50%. As the crosswind speed and crosswind angle increase, the drag coefficient and pressure of each member in the platoon increase. Recycling wind energy through wind turbines and alternators can increase the driving range of electric vehicles by 23%. The study of the aerodynamic characteristics and energy recovery of platooning vehicles can provide direction and theoretical basics for the development and sustainable development of intelligent transportation.
Article
Full-text available
In a team time trial (TTT), the main strategy is based on drafting, where team members alternately take the lead while others ride behind the leading cyclist. TTTs can contain up to 9 riders of the same team. To the best of our knowledge, systematic aerodynamic studies of drafting groups from 2 up to 9 riders have not yet been published. Therefore, this paper presents such an analysis for up to 9 drafting cyclists in a single paceline, with wheel-to-wheel spacings d = 0.05, 0.15, 0.5, 1 and 5 m. A total of 47 Computational Fluid Dynamics (CFD) simulations are performed with the 3D RANS equations, standard k-ε model and scalable wall functions and validated with wind-tunnel measurements. In groups of up to 5 identical riders with d up to 1 m, the last rider has the lowest drag but this is not the case for larger groups. A closely drafting group of 7, 8 or 9 riders has an average drag that is about half that of an isolated rider. However, for much longer theoretical single pacelines, a staggered peloton configuration can yet be about two times more drag efficient.
Article
Full-text available
The flow-field around a “common” European heavy truck, equipped with several different trailer devices, is investigated using steady and unsteady simulations. This work demonstrates how with simple devices added on the trailer it is possible to strongly decrease the aerodynamic drag over 10%, with an increase of overall dimensions below 1% without any change to the load capacity of the trailer. Several devices, installed on the trailer, are tested on a target vehicle and the shape of the “airbag”, the “fin”, the “boat tail” and the “front-rear trailer device” has been optimized to achieve the maximum in drag reduction in front wind. The performance of the optimized devices are tested also in cross wind conditions with the yaw angle varying from 0° to 30°. The truck equipped with the front-rear trailer device is also investigated using time variant simulation with yaw angle of 0°, 5°, 10°. Detached eddy simulation (DES) with one equation Spalart-Allmaras as turbulence model is used to perform the analysis of the unsteady flow; the two equation k-ω SST turbulence model is selected for the steady-state RANS simulations. The unsteady simulations are performed using computational meshes on the order of 40 million of elements while RANS simulation are done with mesh of 10 million of elements. The analysis is performed in the open-source CFD framework OpenFOAM®.
Book
Automobiles are responsible for a substantial part of the world's consumption of primary energy, mostly fossil liquid hydrocarbons. The reduction of the fuel consumption of these vehicles has become a top priority. Many ideas to reach that objective have been presented. In most cases these systems are more complex than the traditional approaches. For such complex systems a heuristic design approach fails. The only way to deal with this situation is to employ model-based methods. This text provides an introduction to the mathematical modeling and subsequent optimization of vehicle propulsion systems and their supervisory control algorithms.
Conference Paper
This paper presents the fuel consumption reduction experienced by two heavy-duty trucks driving at close spacing. The trucks were coupled without any mechanical contact using an "Electronic Tow Bar" developed within the European project PROMOTE-CHAUFFEUR at DaimlerChrysler. The experiments were carried out on a level test track with constant velocities and with spacing between 6 and 16 m. As expected from other studies, both trucks experienced a considerable fuel consumption reduction compared to drives in isolation. At a velocity of 80km/h and a spacing of 10 m the measured fuel consumption reduction was of about 21 percent for the trail truck which had a total weight of 28 t. An extrapolation of the results for other vehicle weights is also presented. For the covering abstract see ITRD E108169.
Article
The increasing importance of reducing greenhouse gas emissions and the ongoing evolution of vehicle-to-vehicle connectivity technologies have generated significant interest in platooning for commercial vehicles, where two or more vehicles travel in same traffic lane in relatively close proximity. This paper examines the effect of platooning on four increasingly aerodynamic tractor-trailer configurations, using a Lattice Boltzmann based CFD solver. Each platoon consisted of three identical tractor-trailer configurations traveling in the same lane at 65mph. Two different vehicle to vehicle gaps were studied, 5m and 9m, in addition to singleton (solitary) vehicles, representing an effectively infinite gap. Aerodynamic drag for the lead, middle, and trailing vehicle in the platooning configurations were compared to the corresponding single vehicle tractor-trailer configuration. Unlike similar investigations on platooning that solely focused on the potential fuel economy gain via aerodynamic drag reduction, this paper also highlights the adverse impact of platooning on the engine cooling performance in trailing vehicles, due to the reduction in ram air cooling caused by wake of the upstream vehicle. Since much of the fuel economy gain from aerodynamics could be at risk if the fan has to be engaged to meet engine cooling demand, this analysis should be an essential part of the cost-benefit consideration of platooning in the context of reducing fuel consumption and thereby, greenhouse gas emissions.
Article
Modern aerodynamic Class 8 freight tractors can improve vehicle freight efficiency and fuel economy versus older traditional style tractors when pulling Canadian style A- or B-Train double trailer long combination vehicles (LCV's) at highway speeds. This paper compares the aerodynamic performance of a current generation aerodynamic tractor with several freight hauling configurations through computational fluid dynamics evaluations using the Lattice-Boltzmann methodology. The configurations investigated include the tractor hauling a standard 53′ trailer, a platooning configuration with a 30′ separation distance, and an A-Train configuration including two 48′ trailers connected with a dolly converter. The study demonstrates CFD's capability of evaluating extremely long vehicle combinations that might be difficult to accomplish in traditional wind tunnels due to size limitations. The paper discusses the relevance of improvements in freight configurations to legislative discussions in the United States and other countries on rules modifications to achieve improved freight efficiency and associated reductions in greenhouse gas emissions.