How things work: OMEGA modeling case study based on the 2018 Toyota Camry

Page created by Pamela Sullivan
 
CONTINUE READING
WORKING PAPER 2018-03

How things work: OMEGA modeling case
study based on the 2018 Toyota Camry
Author: John German
Date: February 27, 2018
Keywords: CO2 emissions, fuel consumption, EPA modeling, OMEGA, Novation

The 2018 Toyota Camry incorpo-                     commissioned by the largest vehicle                     improve their analysis. The com-
rates eight technology upgrades that               manufacturer association and per-                       m i t te e re co g n i ze s t h a t s u c h
are specifically modelled in the U.S.              formed by Novation Analytics, are                       m e t h o d s a r e ex p e n s i ve b u t
Environmental Protection Agency’s                  inaccurate and underpredict future                      believes that the added cost is
Optimization Model for Reducing                    technology benefits.                                    well justified because it produces
Emissions of Greenhouse Gases from                                                                         more reliable assessments
Automobiles (OMEGA) and Lumped                                                                             (Recommendation 8.3).1
Parameter Model (LPM). This large                  Background
number of simultaneous technology                  Modeling efficiency technology is                  As noted by the National Academy
upgrades makes the Camry an excel-                 complicated and difficult as there are             of Sciences committee, the agencies
lent case study of how accurately the              synergies and overlaps between tech-               have developed sophisticated
EPA’s models project future technol-               nologies. The efficiency impact of                 co m p u te r s i m u l a t i o n m o d e l s to
ogy benefits and handle synergies                  any technology varies depending on                 address technology synergies. The
between technologies.                              what other technologies are bundled                EPA has developed its own full-
                                                   with it, the order in which the model              vehicle computer simulation model,
This paper carries out that case study.            adds technologies to a vehicle, and                ALPHA, to estimate the effects of
It defines and adjusts for all differ-             the characteristics of the vehicle. The            individual technologies and tech-
ences between the 2015 and 2018                    importance of accurately modeling                  nology packages. The ALPHA data
Camry and all differences between                  these synergies has been widely                    and outputs are used to calibrate
the EPA’s technology assumptions                   acknowledged and promoted. For                     the EPA’s Lumped Parameter Model
and technology on the 2018 Camry.                  example, the 2015 National Academy                 (LPM), which is used to account for
The results show that the actual CO                of Sciences study on CAFE stated:                  synergies between technologies.
                                      2
reductions of 18.6% achieved in the                                                                   Finally, the EPA’s OMEGA model is
2018 Camry exceed the 17.7% reduc-                      Further, the committee notes                  used to assign the technology results
tions predicted by EPA models. This                     that the use of full vehicle simu-            and synergies to 29 vehicle classes
confirms that the OMEGA and LPM                         lation modeling in combination                and to assess the cost of compliance
models accurately predict both new                      with lumped parameter modeling
technology benefits and synergies                       and teardown studies contributed
                                                                                                      1   Committee on the Assessment of
between technologies. If anything,                      substantially to the value of the                 Technologies for Improving Fuel Economy
they are somewhat conservative.                         Agencies’ estimates of fuel con-                  of Light-Duty Vehicles, Phase 2, Cost,
It also suggests that other studies                     sumption and costs, and it recom-                 effectiveness and deployment of fuel
                                                                                                          economy technologies for light-duty vehicles
contradicting EPA model outcomes,                       mends they continue to increase                   (2015). https://www.nap.edu/read/21744/
including specifically a study                          the use of these methods to                       chapter/1#vii.

Acknowledgements: The author would like to thank Dan Meszler of Meszler Engineering Services for his assistance with the OMEGA model.

© INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION, 2018                                                                           WWW.THEICCT.ORG
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

with standards. The LPM is built into                     • LPM-informed OMEGA tech-                    2018 Toyota Camry. Compared with
the various OMEGA modeling runs.2                           nology packages for the Prior               the baseline 2.5L 2015 Camry, Toyota
                                                            Determination continue to fail              added for the 2018 model eight new
                                                            reasonable tests for plausibil-             or improved technologies used in
The issue                                                   ity. Notwithstanding EPA com-               EPA modeling. This makes the 2018
B a s e d o n a st u d y by N ova t i o n                   ments to the contrary, the Alliance         Camry an excellent test of the OMEGA
Analytics,3 the Alliance of Automobile                      believes that the plausibility limits       model’s ability to account for syner-
Manufacturers (AAM) asserted in                             chosen remain appropriate.                  gies between technologies. This report
comments on the Reconsideration of                                                                      assesses engine, transmission, and
                                                          • Only 18% of the Prior
EPA’s Final Determination that there                                                                    road load improvements to the 2018
                                                            Determination’s technology pack-
was systematic bias in the LPM:4                                                                        Camry and compares CO2 projections
                                                            ages are traceable to the ALPHA
                                                            full vehicle simulations through the        by the OMEGA model using the same
The AAM highlights the following                                                                        technologies with actual CO2 reduc-
                                                            LPM. The underlying data for other
issues identified by the study for EPA’s                                                                tions in the 2018 Camry. If the vehicle
                                                            LPM-based assessments is unclear.
attention for reconsideration of the                                                                    meets or exceeds the OMEGA projec-
Prior Determination.                                      • The LPM-based technology                    tions—seven years early—that would
                                                            package simulations appear to               strongly support the accuracy of the
    • The addition of power-to-weight                       have a high-efficiency, low-CO
                                                                                           2            EPA modeling.
      and vehicle road load does little to                  bias, on average as compared to
      improve the accuracy of the LPM.                      the underlying ALPHA data.
                                                                                                        Model inputs
2    The EPA recently announced that as part of       I n a m e e t i n g w i t h t h e AA M o n
                                                                                                        The first step was to map the tech-
     its efforts to improve modeling, the functions   November 8, 2017, the EPA directly
     performed by the LPM will be integrated                                                            nologies on the 2015 and 2018 Camry
     into the ALPHA model. The EPA is making
                                                      addressed the LPM modeling issues
                                                                                                        to the EPA’s technology codes. Table 2
     the maps in ALPHA more realistic and             raised by the AAM. 5 However, this
                                                                                                        lists the technologies on each vehicle
     replacing old “development” maps using           report takes a different approach.
     benchmarking of recent engines. These                                                              and the associated technology codes.
     improvements are not considered here, as
                                                      Rather than argue about the merits
                                                                                                        Technology changes coded into the
     the improved model has not been released.        of individual pieces of the modeling
                                                                                                        model are highlighted in red, to make
     However, a recent EPA presentation said:         procedure, this report conducts a
     “The revised fleet compliance modeling                                                             it easy to identify the changes from
     methodology would be substantially similar
                                                      case study comparing LPM/OMEGA
                                                                                                        2015 to 2018. Here is a brief descrip-
     to the previous methodology, but with a set      modeling of CO 2 reductions for a
                                                                                                        tion of each technology:
     of auto-calibrated response surface equations    specific vehicle with the actual CO2
     taking the place of the LPM.” (EPA Office of
     Transportation and Air Quality, Powertrain
                                                      reduction achieved adopting a wide                Oil and engine friction. The 2015
     efficiency in EPA’s technical assessment to-     range of technologies.                            Camry uses 0W-20 oil, and the 2018
     date, and plans for ongoing updates Meeting
                                                                                                        Camry uses 0W-16 oil. The EPA codes
     with AAM and Global Automakers [2017,            Previously, a meaningful comparison
     September 21].) The EPA’s future modeling                                                          the base 2015 Camry with EFR1, or
     revision thus should not significantly affect
                                                      could not be conducted because a
                                                                                                        engine friction reduction 1, and LUB,
     the results reported here.                       valid test of the EPA’s modeling
                                                                                                        for use of 0W-20 engine oil viscosity.
3    The AAM contracted with Novation Analytics       assumptions requires the simultane-
     for a review of the LPM’s calibration from the                                                     While we have not been able to find
                                                      ous adoption of multiple technolo-
     ALPHA model. That report, Evaluation of the                                                        quantification of the fuel economy
     Environmental Protection Agency’s Lumped         gies that the EPA assumed would be
                                                                                                        benefits of 0W-16, it has lower friction
     Parameter Model Informed Projections from        widespread in 2025, and a suitable
     the Proposed Determination, was provided                                                           at high temperatures, and there are
                                                      case-study vehicle did not exist. Now
     to the EPA as Attachment 3 to the AAM’s                                                            suggestions that the fuel economy
     comments on the Reconsideration of EPA’s         one does after the redesign of the
     Final Determination.
                                                                                                        benefit could be significant.6 Also, EPA
4    AAM comments re: Request for Comment on                                                            modeling of ATK2 engines, or Atkinson
     Reconsideration of the Final Determination       5    EPA Memorandum from Kevin Bolon to           cycle engines for non-hybrid vehicles,
     of the Mid-Term Evaluation of Greenhouse              Docket EPA-HQ-OAR-2015-0827, Stakeholder
                                                                                                        assumes that they have the EFR2, or
     Gas Emissions Standards for Model Year                Meeting with Novation Analytics and EPA
     2022–2025 Light-Duty Vehicles; Request for            summary presentation on Technical Response
     Comment on Model Year 2021 Greenhouse                 to Assertions of ‘ALPHA-to-OMEGA Bias’.
     Gas Emissions Standards (EPA–HQ–OAR–                  EPA Office of Transportation and Air
     2015–0827; FRL–9966–62–OAR). Submitted                Quality, (December 18, 2017). https://www.   6   Tammy Neal, “The skinny on OW-16 oil,”
     to Docket ID No. EPA–HQ–OAR–2015–0827                 regulations.gov/document?D=EPA-HQ-               National Oil & Lube News, June 30, 2017.
     by AAM on October 5, 2017.                            OAR-2015-0827-10995                              https://noln.net/2017/06/30/skinny-ow-16-oil/.

2    INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION                                                                            WORKING PAPER 2018-03
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

engine friction reduction 2, package.7          Fuel injection. Gasoline Direct Injection          and no aero for 2015 to LRRT2 and
The modeling of the 2018 Camry was              (DI) was added for the 2018 Camry                  Aero1 for 2018.
run with both EFR1+LUB and EFR2                 engine and is coded as such.
coding to evaluate the sensitivity of                                                              Table 1. Road load coefficients for 2015
this assumption, but the EFR2 case is           Atkinson cycle and cooled EGR. Both
                                                                                                   and 2018 Toyota Camry.
considered to be more appropriate.              technologies were added for the 2018
                                                Camry and are coded as such. The                                        4-cylinder – 3,500
Cam phasing. The 2015 Camry uses                Camry engine has a 13.0:1 compres-                                        pounds ETW
variable valve timing on both intake            sion ratio and does not have fast                   Coefficients        2015           2018
and exhaust camshaft (dual VVT-i,               engine warmup. Note that ATK2 in the                      A               27.232         21.006
where “i” means intelligent control).           EPA modeling is based on an engine
                                                                                                          B             0.04319         0.17604
The EPA codes the base 2015 Camry as            with a 14.0:1 compression ratio and
                                                                                                          C            0.019374       0.016028
VVT.8 The 2018 Camry adds an electric           fast engine warmup, with a lower CO2
motor on the intake camshaft for faster         penalty on the cold start portion of the           ETW = Equivalent test weight.
and more precise control, but the EPA           Federal Test Procedure (FTP).                      See Appendix A for discussion of road load
                                                                                                   coefficients.
has no coding to reflect this, and the
benefits are unknown and likely to be           Rolling resistance and aerodynamic
                                                drag. The 2018 Camry has significantly             Performance. While not a technology
small. So the 2018 model is also coded
                                                lower road load coefficients, as shown             change, the EPA’s modeling assumes
as VVT.
                                                in Table 1. The road load is about 15%             that vehicles maintain constant levels
Transmission. The 2015 Camry uses a             lower at 15 mph and 10% lower at 40                of performance when adding technol-
6-speed automatic transmission with             mph. Weighted by the speeds on the                 ogy. Thus, any change in performance
electronic controls, which Toyota calls         FTP and highway cycle, the average                 also needs to be evaluated for trad-
ECT-i. The 2018 Camry also uses ECT-i           road load reduction is about 11%                   eoffs with efficiency. Performance is
and increases the number of gears to            (see Appendix A for details). This is              proportional to the power-to-weight
eight. The 2015 transmission clearly            reflected in the modeling by reducing              ratio of the vehicle, so both weight
fits the TRX11 transmission code defi-          both aerodynamic drag and rolling                  and performance changes need to be
nition, and the 2018 transmission, the          resistance, using low rolling resistance           evaluated. The base 2015 and 2018
TRX21 code.                                     tires (LRRT), by 10%. This is handled by           Camrys were both tested at 3,500
                                                changing the model inputs from LRRT1               equivalent test weight (ETW), and
Steering. Both the 2015 and the 2018
Camry use electric power steering
                                                Table 2. Actual Camry technology, 2015 and 2018, and EPA coding. Changes from the
(EPS). Note that there is an error in the       2015 Camry to the 2018 Camry are highlighted in red.
EPA’s coding for the 2015 Camry, as path
0 for the Camry does not include EPS.                                Actual Camry technology                        EPA coding
To correct for the presence of EPS on                                  2015           2018             path 0          2015             2018
the baseline 2015 Camry, we modeled             Oil                   0W-20          0W-16              LUB             LUB           LUB / —
scenarios with and without EPS.
                                                Engine friction                                         EFR1           EFR1         EFR1 / EFR2
Accessories. The 2018 Camry added               Cam phasing          dual VVT-i    VVT-iE (I)           VVT             VVT             VVT
variable cooling and a variable oil             Transmission      6A w ECT-i       8A w ECT-i          TRX11           TRX11           TRX21
pump, meaning that both are com-                Steering              electric       electric                           EPS             EPS
puter-controlled for increased effi-
                                                                                  var. cooling &
ciency. This fits the EPA definition            Accessories                                                                            IACC1
                                                                                    oil pump
o f I ACC 1 ( i m p rove d a cce ss o r i e s
package 1), so IACC1 is added for the           Fuel injection          SPI         DI + SPI                                             DI
2018 Camry.                                     Atkinson cycle                       13:1 CR                                           ATK2
                                                EGR                                  cooled                                            cEGR
                                                Rolling resistance                    -11.3 %          LRRT1           LRRT1           LRRT2
7   Note that LUB is removed from the model
    inputs when EFR2 is added, as LUB is        Aero                 Cd = 0.28      road load                                          Aero1
    included in the EFR2 package.
                                                                       3,500          3,500
8   Note that the LPM codes dual-VVT as dual    Weight (ETW)
    cam phasing.
                                                                      pounds         pounds

WORKING PAPER 2018-03                                                              INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION                 3
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

even their curb weight differed by less            Table 3. Actual CO2 emissions for 2015 and 2018 Camry compared with modeled results
than 10 pounds. However, the 2018                  for specific technology packages.
Camry engine has significantly greater
                                                                                                                          % reduction vs
power than the 2015 Camry, with                                                       gCO2 /mi              Model          2015 Camry
horsepower increasing by 14% from
                                                                                 Actual                reduction
178 to 203 and torque rising by 8%                                                veh.       Model     v no-tech       Actual        Model
from 170 ft-lb to 184 ft-lb. In addition,           OMEGA tech path 0                         237.5         15.2%
the number of transmission gears
                                                    2015 actual                   232.0      234.7          16.2%
increased from six to eight, which also
increases performance by keeping                    2018 actual w/o EFR2          188.9       189.9         32.2%       18.6%         19.1%
the engine at higher rpm after each                 2018 actual                   188.9       187.9         32.9%       18.6%         19.9%
shift and providing higher gear multi-                   Adjustments                                                              Adj. Model      CO2 adj.
plication for launch. While the actual              14:1 to 13:1 CR               188.9                                 18.6%        19.0%         + 1.2%
performance improvement is clearly                  +10% performance              188.9                                 18.6%         17.7%        + 1.5%
more than 10%, for the purposes of
this comparison the performance gain
                                                   part this reflects that the 2015 baseline           to 14:1 is 0.9%.11 Modeling in this paper
is conservatively estimated at 10%.
                                                   CO2 in the modeling is 2.7 g/mi higher              conducted with Autonomie found
This reduces the downward adjust-
                                                   than the actual readings. The percent-              that the fuel-consumption reduction,
ment of the modeled fuel economy to
                                                   age reductions from 2015 to 2018 were               if performance is maintained, is 1.32
reflect the fuel economy/performance
                                                   19.9% for the modeling results and                  times the brake efficiency improve-
tradeoff and thus causes the modeled
                                                   18.6% for the actual 2018 Camry.                    ment. So increasing the compression
fuel economy to be slightly overstated.
                                                                                                       ratio from 13:1 to 14:1 would reduce fuel
                                                   However, this is not the most appro-                consumption by 1.19%. Applying this to
                                                   priate comparison, as the modeling
Analysis                                           results assume a 2025 powertrain with
                                                                                                       the 2018 Camry modeling narrows the
                                                                                                       percent reduction from 19.9% to 19.0%
Table 3 presents actual CO2 emissions              additional optimization and improve-                percent—only 0.4% more than the
for the 2015 and 2018 Camry from                   ments beyond those incorporated in                  actual reduction of the 2018 Camry
EPA test car lists and compares the                the 2018 Camry and assume constant                  compared with the 2015 Camry.
findings with the modeled results for              vehicle performance. As noted above,
the technology packages applied spe-               the 2018 Camry uses a 13.0:1 com-                   OMEGA assumed constant perfor-
cifically to the Camry.9                           pression ratio, while EPA modeling                  mance for all of the modeling results.
                                                   assumes 14.0:1, the Camry does not                  However, the 2018 Camry has sig-
Comparing the “2018 actual” row,                   have the fast engine warmup assumed                 nificantly better performance than
where EFR2 is included in the 2018                 in EPA modeling, and the 2018 Camry                 the 2015 Camry, as noted above. For
Camry technology, the reductions                   has at least 10% better performance                 modeling purposes, the overall per-
achieved by the 2018 Camry do not                  than the 2015 Camry.                                formance gain was conservatively
quite match the reductions projected
                                                                                                       estimated to be 10%. While there is no
by EPA modeling. The actual CO2 emis-              Based on a 2014 paper by Speth et al.,10            consensus on the tradeoff between
sions are very close, just 1.0 g/mi, or            the brake efficiency improvement for                performance and fuel consumption,
0.5%, higher than the modeled result. In           increasing compression ratio from 13:1              it can be derived from the impacts of

                                                                                                       11    This is extrapolated from the modeled
9   The file MS_Control_in2025AB_20161118_         10 Raymond L. Speth, Eric W. Chow, Robert                 results, as the benefits of increasing
    icm_aeoR_ScenarioPackages was used to             Malina, Steven R. H. Barrett, John B. Heywood,         compression ratio decrease as the baseline
    generate the model results. A description of      and William H. Green, “Economic and                    compression ratio increases. For example,
    the OMEGA model and links to download the         Environmental Benefits of Higher-Octane                the paper found that increasing the
    model, including this file, can be found at:      Gasoline,” Environmental Science &                     compression ratio from 10.5:1 to 11.5:1 would
    https://www.epa.gov/regulations-emissions-        Technology, 2014, 48 (12), 6561-6568 DOI:              improve brake efficiency by 1.9%, or more
    vehicles-and-engines/optimization-model-          10.1021/es405557p. https://pubs.acs.org/doi/           than twice the 0.9% benefit of increasing
    reducing-emissions-greenhouse-gases               abs/10.1021/es405557p                                  from 13:1 to 14:1. See Appendix B for details.

4   INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION                                                                              WORKING PAPER 2018-03
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

weight reduction on fuel economy.12                Comparison with footprint                        the OMEGA model properly accounts
Reducing weight by 10% results in                                                                   for synergies between technologies
                                                   targets
about a 5.2% decrease in fuel con-                                                                  and does not overpredict technology
sumption with a 10% increase in                    The 2018 2.5L Camry meets its 2022               benefits. In fact, its projections are most
acceleration. Reducing weight by 10%               footprint target without any consid-             likely conservative.
while maintaining constant levels of               eration of air conditioning or off-cycle
performance, for example by down-                  credits, without weight reduction, and           Another unavoidable conclusion is that
sizing the engine, results in about a              without any kind of hybridization, not           the Novation critique of EPA modeling
6.6% decrease in fuel consumption.                 even stop/start.13 Just maximizing off-          is demonstrably wrong. There is clearly
Thus, the impact of increasing per-                cycle credits of 10 grams of CO2 per             no systematic bias in the LPM, and
formance by 10% results in roughly a               mile (g CO2/mi) and air conditioning             the 2018 Camry can easily meet its
1.5% increase in fuel consumption or               efficiency credits of 5 g CO2 /mi would          2025 targets without hybridization
CO2 emissions.                                     allow the Camry to almost meet the               beyond possibly a stop/start system.
                                                   2024 targets. The Camry still has seven          This further suggests that Novation’s
Applying this performance adjustment               years to make modest improvements                analyses are not properly constructed
to the 2018 Camry modeling drops the               and can easily reach its 2025 target             and that the bias exists in Novation’s
modeled reduction from 19.0% to 17.7%              by adding a stop/start system, with              modeling, not the EPA’s.
—significantly less than the actual                modest reductions of weight, aerody-
18.6% reduction of the 2018 Camry                  namic drag, or tire rolling resistance, or
versus the 2015 Camry.                             even petitioning the EPA for additional          APPENDIX A
                                                   off-cycle credits beyond those listed in         Rolling resistance and
Note that this result does not account             the off-cycle menu.                              aerodynamic drag
for calibration improvements from
2018 to 2025. Minor improvements                                                                    Road load coefficients are generated
to technologies and calibrations are               Conclusion                                       using a coast-down test on a track.
routine and will certainly increase the                                                             For a given vehicle weight, the longer
                                                   Before considering differences in perfor-        it takes for the vehicle to lose velocity,
CO2 reduction that can be achieved                 mance, OMEGA modeling of the tech-
with this technology set by 2025.                                                                   the lower the tire rolling resistance
                                                   nology improvements in the 2018 Camry            or aerodynamic drag, or both. The
Also note that a likely conservative               almost exactly matches the actual
adjustment was made for the impact                                                                  vehicle speed versus time curve is
                                                   reductions achieved by the vehicle after         fitted using A, B, and C coefficients,
of the improved performance of the                 adjusting for technology differences,
2018 Camry on fuel economy, plus no                                                                 where the A coefficient is a fixed value
                                                   such as lower compression ratio and              regardless of speed, B is a function of
adjustment was made for the lower                  higher CO2 after cold start. However, this
cold start fuel consumption included                                                                vehicle speed, and C is a function of
                                                   does not take into account the much              the square of vehicle speed.
by OMEGA in their ATK2 technology,                 higher performance of the 2018 Camry.
which is also conservative.                        After properly adjusting for the perfor-         The 2018 Camry had significantly
                                                   mance gains, the 2018 Camry achieved             lower A and C road load coefficients
                                                   a CO2 reduction of 18.6%, significantly          than the 2015 Camry, although the B
12 Details of the calculation of the tradeoffs     greater than the 17.7% predicted by the          coefficient was higher (see Table A-1).
   between fuel economy and performance            adjusted OMEGA modeling. And this
   are in Appendix B. Note that Appendix B
   also provides an example of the 2018 Honda      is without accounting for seven more
                                                                                                    Table A-1. Road load coefficients for 2015
   Accord, which uses a similar turbocharger       years of development and calibration
   system on both the base 1.5L and the
                                                                                                    and 2018 Toyota Camry.
                                                   improvements and the reductions in
   optional 2.0L engines. Compared with the
   1.5L Accord engine, the 2.0L engine increases   cold start fuel consumption modeled by                                 4-cylinder – 3,500
   horsepower by 31% from 192 to 252, and          OMEGA, plus a likely conservative per-                                   pounds ETW
   torque, by 42% from 192 ft-lb to 273. Using     formance adjustment. This confirms that           Coefficients         2015         2018
   the 6-speed manual as the control (as the
   automatics are different for the 1.5L and the                                                           A             27.232       21.006
   2.0L), the 1.5L is rated at 30 mpg and the      13 John German, “Technology Leapfrog:
                                                                                                           B            0.04319       0.17604
   2.0L at 26 mpg, or 13% lower. So, for every        Or, all recent auto technology forecasts
   10% increase in performance on this engine,        underestimate how fast innovation is                 C            0.019374     0.016028
   fuel economy decreased by about 4%—or              happening,” ICCT blog, September 25,
   over 2.5 times the amount calculated from          2017. https://www.theicct.org/blog/staff/     ETW = Equivalent test weight.
   the weight-reduction formulas.                     technology-leapfrogging

WORKING PAPER 2018-03                                                                    INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION          5
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

Figure A-1 illustrates the total road
                                                                                                      2018 vs 2015 Camry Road Load
load from the A, B, and C coefficients
                                                                       120                                                                                 24%
at different vehicle speeds.

                                          Road Load (pound-force)

                                                                                                                                                                 Percentage difference between
Figure A-2 shows the speed                                             100                                                                                 20%
distribution on the FTP and highway

                                                                                                                                                                     2018 and 2015 Camry
test cycles. Weighting the 2015 and                                    80                                                                                  16%
2018 2.5L Camry road load (blue and
red lines in Figure A-1) by the FTP                                    60                                                                                  12%
speed distribution (blue line in Figure
A-2) yields an average road load                                       40                                                                                  8%
reduction of 12.8%, and weighting by
the highway speed distribution (red                                    20                                                                                  4%
line in Figure A-2), 10.4%. Finally,
                                                                                                                          2015      2018    % reduction
weighting the FTP and highway                                           0                                                                                  0%
results by 55% city and 45% highway                                          0   5       10    15     20   25   30   35     40   45    50   55   60   65
yields an overall average road load                                                                        Vehicle speed (mph)
reduction of 11.3%.
                                             Figure A-1. Road load comparison of 2018 and 2015 2.5L Camry.

APPENDIX B                                                                                FTP and Highway Test Cycle Speed Distribution
Performance tradeoffs with                                             120                                                                                 24%
fuel economy
                                                                                     2018 Camry %

                                                                                                                                                                 Percentage difference between
                                                                       100                                                                                 20%
The tradeoff between performance                                                       reduction in
                                          # seconds at vehicle speed

                                                                                         road load
and fuel consumption is difficult

                                                                                                                                                                     2018 and 2015 Camry
                                                                       80
to calculate. Theoretically the fuel-                                                                                                                      16%
consumption increase with improv-                                      60
ing performance should diminish in                                                                                                                         12%
the future as engines improve fuel                                     40
                                                                                                                          Highway
consumption at low engine loads. No                                                    FTP                                                                 8%
consensus has been reached on how                                      20
to quantify the tradeoff.                                                                                                                                  4%
                                                                        0
                                                                             0   5       10    15     20   25   30   35     40   45    50   55   60   65
The approach taken here is to derive
the performance versus fuel economy                                    -20                                                                                 0%
                                                                                               FTP # sec        Hwy # sec        % reduction
tradeoff from the impacts of weight
reduction on fuel economy. Reducing          Figure A-2. FTP and highway test cycle speed distributions and related reductions in
weight affects both fuel consump-            road load for 2018 Camry compared with 2015 Camry.
tion and performance. Most studies
                                             only technology variant.14 The results                                    0.467 %, and 0.447% per % change
of the impacts of weight reduction
                                             of this evaluation revealed average                                       in vehicle mass for vehicles executing
have found that a 10% weight reduc-
                                             CO 2 emission changes of 0.456%,                                          the NEDC, WLTP low road-load, and
tion directly results in about a 4.5%
                                                                                                                       WLTP high road-load cycles, respec-
decrease in fuel consumption. To                                                                                       tively. While not listed in this report, the
                                             14 Dan Meszler, John German, Peter Mock,
generalize the CO 2 effects of mass             and Anup Bandivadekar, CO2 reduction                                   same methodology applied to the US
reduction, as part of an ICCT study on          technologies for the European car and                                  cycles revealed average CO2 emission
European post-2020 CO2 standards,               van fleet: a 2025-2030 assessment
                                                methodology and summary of compliance                                  changes of 0.596% and 0.426% per
ICCT performed a detailed analysis of           costs for potential EU CO2 standards,                                  % change in vehicle mass for vehicles
26 technology packages developed                (ICCT: Washington, DC, November, 2016).                                executing the FTP and highway test
                                                https://www.theicct.org/sites/default/
by FEV (spanning all modeled vehicle            files/publications/EU-Cost-Curves_ICCT_                                cycle, respectively. Weighting the FTP
classes) where mass reduction was the           nov2016.pdf                                                            and highway results by 55%/45% yields

6   INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION                                                                                          WORKING PAPER 2018-03
HOW THINGS WORK: OMEGA MODELING CASE STUDY BASED ON THE 2018 TOYOTA CAMRY

an average of 0.520% per % change in              weight reduction. Thus, the impact on    192 ft-lb for the 1.5L. Using the 6-speed
vehicle mass.                                     fuel economy of increasing perfor-       manual as the control, because the
                                                  mance by 10% is roughly the fuel-        automatics are different for the 1.5L
However, 10% weight reduction also                economy impact without correcting        and the 2.0L, the 1.5L is rated at 30
improves performance by 10%. Modeling             for performance, or 1 – 5.2% = 94.8%,    mpg and the 2.0L at 26 mpg, or 13%
of the impact of weight reduction while           divided by the fuel-economy impact       lower. So, for every 10% increase in
maintaining constant levels of perfor-            after correcting for performance, or     horsepower, fuel economy decreased
mance—for example, by downsizing                  1 - 6.6% = 93.4%. This yields about a    by 4.3%, and for every 10% increase
the engine—is difficult, as it is affected                                                 in torque, fuel economy dropped by
                                                  1.5% increase in fuel consumption or
by powertrain characteristics, vehicle                                                     3.2%. Overall, for a 10% gain in perfor-
                                                  CO 2 emissions with a 10% improve-
attributes, and how the performance                                                        mance, fuel economy fell by roughly
                                                  ment in performance.
is adjusted. A more reliable method                                                        4%. This is over 2.5 times the amount
is to evaluate the change in the total            The 2018 Honda Accord provides a         calculated from the weight-reduction
energy required to drive a vehicle over           quick check on the accuracy of this      formulas, indicating that the perfor-
a specific drive cycle. ICCT developed            calculation. The vehicle offers a good   mance adjustment used for the 2018
such a method as part of the 2016                 comparison, using a similar 4-cylinder   Camry is most likely conservative.
ICCT study on European post-2020                  turbocharger system on both the base
CO2 standards. It was later applied to                                                     Table B-1. 2018 Honda Accord performance
                                                  1.5L and the optional 2.0L engines,
the US FTP and highway test cycles to                                                      and fuel economy comparison.
                                                  and each engine can be paired with
generate lightweighing benefit inputs             a manual 6-speed transmission. This
to the OMEGA model.15 Because the                                                            2018 Honda Accord w/ M6 transmission
                                                  minimizes the confounding factors of
relationship between the various road-            different engine types, number of cyl-
                                                                                                               HP     ft-lb   MPG
load parameters (i.e., mass, rolling              inders, and transmissions.               1.5L                192     192     30
resistance, and aerodynamic drag)                                                          2.0L                252    273      26
varies across vehicles, estimates were            As shown in Table B-1, the 2.0L engine   % change           31.3%   42.2%   -13.3%
made for six different types of vehicles.         is rated at 252 horsepower, 31% more     MPG decrease
The mid-size car estimate was a 6.6%              than the 1.5L’s 192. The 273 ft-lb of    per 10% HP/ft-lb   -4.3%   -3.2%
decrease in fuel consumption for a 10%            torque for the 2.0L is 42% higher than   increase

15 Nic Lutsey, Dan Meszler, Aaron Isenstadt,
   John German, and Josh Miller, Efficiency
   technology and cost assessment for 2025-
   2030 light-duty vehicles, (ICCT: Washington,
   DC, 22 March, 2017). https://www.theicct.
   org/publications/US-2030-technology-cost-
   assessment

WORKING PAPER 2018-03                                                          INTERNATIONAL COUNCIL ON CLEAN TRANSPORTATION          7
You can also read