Subscriber access provided by Northern Illinois University
Article
Life Cycle Assessment of Vehicle Lightweighting: A Physics-based Model to Estimate Use-Phase Fuel Consumption of Electrified Vehicles Hyung Chul Kim, and Timothy J. Wallington Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b02059 • Publication Date (Web): 17 Aug 2016 Downloaded from http://pubs.acs.org on August 21, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 26
Environmental Science & Technology
1
Life Cycle Assessment of Vehicle Lightweighting:
2
A Physics-based Model to Estimate Use-Phase Fuel
3
Consumption of Electrified Vehicles
4
Hyung Chul Kim* and Timothy J. Wallington
5
Materials and Manufacturing R&A Department, Ford Motor Company, Dearborn, Michigan
6
48121-2053, United States
7
*Corresponding author; email:
[email protected]; phone: 313-323-9745
8 9
ABSTRACT
10
Assessing the life-cycle benefits of vehicle lightweighting requires a quantitative description of
11
mass-induced fuel consumption (MIF) and fuel reduction values (FRVs). We have extended our
12
physics-based model of MIF and FRVs for internal combustion engine vehicles (ICEVs) to
13
electrified vehicles (EVs) including hybrid electric vehicles (HEVs), plug-in hybrid electric
14
vehicles (PHEVs), and battery electric vehicles (BEVs). We illustrate the utility of the model by
15
calculating MIFs and FRVs for 37 EVs and 13 ICEVs. BEVs have much smaller MIF and
16
FRVs, both in the range 0.04-0.07 Le/(100 km 100 kg), than those for ICEVs which are in the
17
ranges 0.19-0.32 and 0.16-0.22 L/(100 km 100 kg), respectively. The MIF and FRVs for HEVs 1 ACS Paragon Plus Environment
Environmental Science & Technology
18
and PHEVs mostly lie between those for ICEVs and BEVs. Powertrain resizing increases the
19
FRVs for ICEVs, HEVs and PHEVs.
20
greenhouse gas emissions than lightweighting ICEVs, however the benefits differ substantially
21
for different vehicle models. The physics-based approach outlined here enables model specific
22
assessments for ICEVs, HEVs, PHEVs, and BEVs required to determine the optimal strategy for
23
maximizing the lifecycle benefits of lightweighting the light-duty vehicle fleet.
24
Lightweighting EVs is less effective in reducing
1. INTRODUCTION
25
Lightweighting by material substitution is a key strategy to reduce fuel consumption and
26
greenhouse gas (GHG) emissions during vehicle operation. Life cycle based analyses are
27
required to assess the net benefits of lightweighting since production of lightweight materials
28
(e.g., aluminum, magnesium, and carbon composites) is generally more energy intensive than for
29
conventional materials such as steel and steel alloys. The conclusions from life cycle
30
assessments (LCAs) of the benefits of vehicle lightweighting are often inconsistent due to
31
incongruous modeling methods and parameters employed. LCA practitioners face a challenge
32
of estimating the fuel consumption during the operation of lightweighted vehicles; the most
33
energy consuming stage of the vehicle life cycle1-2. In our previous studies2-3, we proposed a
34
method of estimating Mass-induced Fuel Consumption (MIF) and Fuel Reduction Values
35
(FRVs) for specific models of internal combustion engine vehicles (ICEVs) based on fuel
36
economy and dynamometer test data available in the U.S. Environmental Protection Agency
37
(EPA) database. MIF in ICEVs is defined as the sum of fuel consumption assigned to mass-
38
dependent work and the mass share of engine friction losses corresponding to the baseline fuel
39
consumption while FRV is defined as the rate of change of vehicle fuel consumption upon
40
lightweighting3. Based on the U.S. Federal Test Procedure (FTP) combined (55-city/452 ACS Paragon Plus Environment
Page 2 of 26
Page 3 of 26
Environmental Science & Technology
41
highway) cycle, we estimated MIFs and FRVs of 0.22−0.43 L/(100 km 100 kg) and 0.15-0.26
42
L/(100 km 100 kg) for 2013 Model Year ICEVs. If the powertrain is resized with lightweighting
43
for equivalent performance, FRVs increase to 0.27−0.53 L/(100 km 100 kg).
44
Fuel-mass correlation for advanced technology vehicles is discussed in the literature in terms of
45
fuel-mass elasticity, i.e., proportional fuel economy (consumption) change divided by
46
proportional mass reduction. Wohlecker et al. (2007)4 compared the mass elasticity of fuel
47
consumption for ICEVs, hybrid electric vehicles (HEVs) and fuel cell vehicles (FCVs) using a
48
simulation tool. They found that with powertrain resizing, the fuel-mass elasticity of ICEVs is
49
0.7-0.8 based on New European Driving Cycle (NEDC) compared to 0.5-0.8 and 0.5-0.6 for
50
HEVs and FCVs, respectively. In the absence of powertrain resizing, ICEVs have the lowest
51
elasticity, 0.2-0.3, with HEVs and FCVs having elasticities of ~0.5 and 0.3-0.6, respectively.
52
Pagerit et al.5 compared mass elasticity of fuel consumption for ICEVs, HEVs, FCVs, and hybrid
53
FCVs based on a simulation tool called Powertrain System Analysis Toolkit (PSAT). They
54
estimated that for a 10% mass reduction with powertrain resizing, the fuel consumption
55
reductions for ICEVs, HEVs, FCVs, and hybrid FCVs were 7%, 6.3%, 5%, and 3.3%,
56
respectively. Without powertrain resizing, the fuel reduction rate for ICEVs was estimated to be
57
smaller (4.1-4.7%) than for advanced vehicles (4.7-5.8%). Lewis et al.6 also reported a higher
58
mass elasticity for ICEVs with powertrain resizing rated at 0.68 compared to 0.62 for HEVs and
59
0.65 for plug-in hybrid electric vehicles (PHEVs). In summary, the available literature studies
60
indicate that fuel-mass elasticity is greater for advanced vehicles than for ICEVs without
61
powertrain resizing while with powertrain resizing, ICEVs show a greater mass elasticity upon
62
lightweighting.
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 26
63
As discussed in our previous studies, the lightweighting LCA method based on fuel-mass
64
elasticity or FRV cannot clearly define the baseline fuel consumption making it difficult to assess
65
the implication of lightweighting benefit2-3. Both MIF and FRV estimates are needed in LCA of
66
the benefits of vehicle or component lightweighting. Use of model-specific instead of generic
67
mass-fuel elasticity values would substantially reduce the uncertainties in such LCAs.
68
Simulation tools such as FASTSim and Autonomie are difficult to adapt for specific vehicle
69
models7-8. We present an extension of our previous MIF and FRV model for ICEVs to
70
incorporate electrified vehicles (EVs), i.e., HEVs, PHEVs, and battery electric vehicles (BEVs).
71
EPA fuel economy test data9 and Fuel Economy Guide10 data for 2015 model year vehicles were
72
used in the extended model to compute MIFs and FRVs for EVs with and without powertrain
73
resizing. To illustrate the utility of the model, we used the MIFs and FRVs to determine the life
74
cycle GHG emissions savings for a lightweighted component. Detailed steps for estimating the
75
use-phase energy consumption and GHG emissions are presented for LCA practitioners. We
76
highlight and discuss the critical parameters for assessments of the life cycle benefits of vehicle
77
lightweighting.
78
2.
MASS-INDUCED FUEL CONSUMPTION FOR ICEVS
79
The mass-induced fuel consumption and fuel reduction model for ICEVs2 accounts for fuel
80
consumption associated with vehicle loads, useful work, and wasteful losses. The vehicle
81
dynamics which describe the vehicle loads over a test driving cycle, P, are:
82
= + + + +
83
where v, a, M, and α are vehicle speed (m/s), acceleration (m/s2), mass (kg), and power demand
84
for accessories, respectively. The coefficients, A (N), B (N/(m/s)), and C (N/(m/s)2) are rolling,
(1)
4 ACS Paragon Plus Environment
Page 5 of 26
Environmental Science & Technology
85
rotating and aerodynamic resistive coefficients, respectively. They are often called "coast-down"
86
or "target" coefficients estimated from dynamometer test and are available in the EPA fuel
87
economy test database9 for specific vehicle models. Av, Bv2, and avM are the rolling, rotating,
88
and acceleration loads, respectively, and are proportional to vehicle mass while Cv3 and α, the
89
aerodynamic and accessary loads are independent of vehicle mass. Integrating the mass-
90
dependent loads over the test schedule gives the mass-dependent fuel consumption, Fw (L):
91
=
92
Hf, ηi, and ηt, are the heating value of fuel, thermodynamic efficiency of the engine, and
93
transmission efficiency, respectively. Although ηi and ηt change during a driving cycle for
94
simplicity we assume they are constant. Only the positive acceleration load (avM) is considered
95
since a vehicle when idling or decelerating does not require tractive work11. Likewise,
96
integrating the aerodynamic and accessory loads over a test schedule provides the mass-
97
independent fuel consumption, Fx (L):
98
=
99
Adding the loss energy inside and outside of the engine, the total fuel use in liters, F, to move a
+ +
(2)
+
(3)
100
vehicle through a drive cycle can be written as follows:
101
= + + ! + "
102
Here Ff and Fl denote fuel energy losses inside the engine (friction and pumping) and
103
miscellaneous losses in the drivetrain, respectively. The descriptions of the fundamental vehicle
(4)
5 ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 26
104
dynamics and fuel consumptions for ICEVs are given in equations (S1)-(S7) in the Supporting
105
Information (SI).
106
In a vehicle lightweighting LCA, determining the use-phase fuel consumption involves
107
separating the mass related fuel consumption from the balance, i.e., those from aerodynamic
108
resistance and accessory power demand. Mass induced fuel consumption (MIF) is defined to
109
assign the ‘baseline’ fuel consumption for status-quo design or material of the component2-3.
110
# = $
111
FC is the fuel consumption rate in units of L/km measured in the fuel economy test, i.e. = *
112
where = or distance driven.
113
The rate of fuel consumption change over mass change in unit of L/(100 km 100 kg) is used to
114
assess fuel consumption of a lightweighted vehicle or a component and is defined as the fuel
115
reduction value (FRV)12. There are two types of FRVs depending on whether the powertrain is
116
resized for equivalent vehicle performance as follows3.
117
+, = -)
118
+, & = -)
119
where FRV and FRV+ are fuel reduction values without and with powertrain resizing
120
respectively. FF is the sum of Ff and Fl and for practical purposes can be approximated by Ff
121
since Fl is much smaller than Ff,3 i.e., FF≈Ff. Equations (5), (6) and (7) can be used with EPA
122
test data to calculate vehicle model specific MIF and FRV(+) for ICEVs9. The complete
123
modeling derivation and application is discussed elsewhere3.
%$% &$'
$
)(
(5) $
-$
-$
*.
*.
$
= )% =
*.
$% &$ )
$
$
%$ )(
)( = $
$% &$
= $% )( = $
*.
=
$
% &$' &$/
$% &$ $
$
% &$' &$/
6 ACS Paragon Plus Environment
(6) $
)(
(7)
Page 7 of 26
124
Environmental Science & Technology
3. MASS-INDUCED FUEL CONSUMPTION FOR EVS
125
To estimate the MIFs and FRV(+)s of HEVs, PHEVs, and BEVs, the ICEV model was revised to
126
include regenerative braking and the electric powertrain maintaining the model framework in
127
equations (5)-(7). For PHEVs, we evaluate MIFs and FRV(+)s separately for the charge
128
sustaining (CS) and charge depletion (CD) mode. In CS mode of the operation of PHEVs is the
129
same as for HEVs while in CD mode PHEVs resemble BEVs. Depending on the power source
130
in the CD mode, PHEVs can be either ‘non-blended’ if they operate with no power supply from
131
the engine, or ‘blended’ if they operate with supplementary power from the engine.
132
EVs capture part of the kinetic energy via regenerative braking system and store it in a battery to
133
provide additional tractive energy11, 13. In contrast, conventional ICEVs employ frictional
134
braking and the kinetic energy accumulated during acceleration in a driving cycle is dissipated
135
during braking. Since the origin of braking energy is the acceleration load term, avM in equation
136
(2), adding regenerative braking has an effect similar to mass reduction13-14. Thus, equations (2)
137
and (3) can be revised as follows for EVs:
138
1 2 =
139
1 2 =
3
3
+ + 1 − ∅7
+
(8)
(9)
140
where Ø is the ratio of braking to kinetic energy and µ is the regenerative braking efficiency.
141
Here, electrical energy is represented as fuel volume equivalence, Le, based on a conversion
142
factor of 8.9 kilowatt-hours of electricity per liter of gasoline15. Note that instead of
143
thermodynamic efficiency (ηi,) energy conversion efficiency (ηc) is used for electric powertrains.
144
We note that since all the losses in the electric motor are accounted for in FEw and FEx, 7 ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 26
145
independent losses such as the engine friction Ff are considered zero in BEVs,3, i.e., FF≈Ff=0.
146
Thus, from equations (5)-(7), the following approximation holds for BEVs and for the CD mode
147
of non-blended PHEVs.
148
# = +, = +, &
149
HEVs and PHEVs do have engine friction losses from the engine work. Detailed energy
150
equations for EVs are available in equations (S8)-(S16) in the SI.
151
(10)
3.1 Regenerative Braking
152
Physical braking is needed when the vehicle is under a negative load during a driving schedule.
153
The amount of braking energy, eb, can be calculated by integrating the negative load over the
154
driving schedule11, 13-14.
155
89 = ;
< =>
.: = ;
+ + +
< =>
(11)
156
Here, Ptr is tractive load, i.e., vehicle load excluding accessory load (=P-α). The amount of
157
braking energy, eb, can be estimated from the second-by-second FTP dynamometer test
158
schedules and coast-down coefficients9, 16. Then, the ratio of braking energy to total kinetic
159
energy over the driving cycle, Ø, in equation (8) can be written as follows:
160
∅=−
2?
ABC @ )*.
(12)
161
Ø is the upper limit for the fraction of kinetic energy that can be collected through regenerative
162
braking during the driving cycle. The detailed steps to determine eb and Ø based on the FTP-city
163
cycle is illustrated in section S2 in the SI. We estimate Ø values of 0.64-0.79 for the FTP-75 city
164
driving cycle (identical to the Urban Dynamometer Driving Schedule [UDDS] plus the first 505 8 ACS Paragon Plus Environment
Page 9 of 26
Environmental Science & Technology
165
seconds of an additional UDDS) and 0.27-0.46 for the Highway Fuel Economy Test (HWFET)
166
driving cycle for the EVs analyzed in this study. The larger value for the city cycle reflects more
167
frequent braking and deceleration than in the highway cycle. To facilitate modeling an average
168
value of 0.74 and 0.41 may be used for the city and highway cycle, respectively. Tables S3 and
169
S4 in the SI list vehicle model specific Ø values and provide uncertainty analysis associated with
170
using the average value.
171
The regenerative braking efficiency, µ, in the literature is a compounded efficiency of the
172
generator, motor, and energy storage (battery or capacitor) systems. With varying rotational
173
speed and torque during the drive cycle, developing a highly efficient system is challenging13.
174
Limited data are available concerning the efficiency of regenerative braking. Weiss et al.
175
(2000)17 used a regenerative efficiency of 0.85 corresponding to the generator only while An and
176
Santini (2004)11 assumed 0.7 for a front wheel drive vehicle. We use a regenerative braking
177
efficiency of 0.8 including a battery charging loss of 5%. The modeling uncertainty in MIF,
178
FRV, and FRV+ associated with the use of regenerative braking efficiencies over the range 0.7-
179
0.85 (rather than 0.8) is small as shown in Table S5 in the SI.
180 181
182
3.2 Energy Conversion Efficiency The energy conversion efficiency in equations (8) and (9) can be written in a simplified form.
3
D
= + E
FD
(13)
183
where ηe and ηi are the motor efficiency including battery loss and the thermodynamic engine
184
efficiency, respectively, and θ is the fraction of tractive work performed by the electric motor.
185
The energy conversion efficiency, ηc, corresponds to motor efficiency, ηe, for BEVs and non-
9 ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 26
186
blended PHEVs in the CD mode, i.e., θ=1. For HEVs, PHEVs in the CS mode, and blended
187
PHEVs in the CD mode, it is a compounded electric and thermodynamic efficiency of motor,
188
generator, battery and engine, i.e., 0
