higher potential waste heat recovery rates are being researched (Ricardo, 2012).
This committee’s approach to estimating future vehicle fuel economy differs from most projections of future ICE efficiency, which have generally assessed the benefits of specific technologies that can be incorporated in vehicle designs (see Appendix F). Such assessments work well for estimates out 15 to 20 years, but their usefulness for 2050 suffers from two major problems. One is that it is impossible to know what specific technologies will be used in 2050. The traditional approaches taken to assess efficiency, such as PSAT and ADVISOR, depend on having representative engine maps, which do not exist for the engines of 2050. The second is that as vehicles approach the boundaries of ICE efficiency, the synergies, positive and negative, between different technologies become more and more important; that is, when several new technologies are combined, the total effect may be greater or less than the sum of the individual contributions.
The three-step approach used here avoids these problems. First, for ICE and HEV technologies, sophisticated computer simulations conducted by Ricardo were used to establish powertrain efficiencies and losses for the baseline and 2030 midrange cases.4 These simulations fully accounted for synergies between technologies. Second, the efficiencies and losses of the different powertrain components and categories were determined. Using these categories to extrapolate efficiencies and losses allowed the committee to properly assess synergies through 2050. Third, the estimates of future efficiencies and losses were simultaneously combined with modeling of the energy required to propel the vehicle as loads, such as weight, aerodynamics, and rolling resistance, were reduced. This approach ensures that synergies are properly assessed and that the modeled efficiency results do not violate basic principles.
The committee estimated conventional powertrain improvements using the results of sophisticated simulation modeling conducted by Ricardo (2011). This modeling was used by the U.S. Environmental Protection Agency (EPA) to help set the proposed 2025 light-duty vehicle CO2 standards. Ricardo conducted simulations on six different vehicles, three cars and three light trucks, which examined drivetrain efficiency (not load reduction) in the 2020-2025 timeframe. The simulations were based on both existing cutting-edge technologies and analyses of technologies at advanced stages of development.
EPA post-analyzed Ricardo’s simulation runs and apportioned the losses and efficiencies to six categories—engine thermal efficiency, friction, pumping losses, transmission efficiency, torque converter losses, and accessory losses. The committee used these results as representative of potential new-vehicle fleet average values in 2025 for the optimistic case and in 2030 for the midrange case. The 2050 mid-level and 2050 optimistic vehicles were constructed by assuming that the rates of improvement in key drivetrain efficiencies and vehicle loads would continue, although at a slower rate, based on the availability of numerous developing technologies and limited by the magnitude of the remaining opportunities for improvement.
Baseline inputs for 2010 ICEVs were developed by the committee from energy audit data that corresponded with specific baseline fuel economy. The model calculates changes in mpg based on changes in input assumptions over EPA’s test cycles. Additional details of the model are in Appendix F. The results were averaged to one car and one truck for analysis in the scenarios, but the analysis for all six vehicles is in Appendix F.
Starting with the results for ICEVs, the energy audit model was then applied to the other types of vehicles considered in this report for each analysis year and for the midrange and optimistic scenarios. PHEVs were assumed to have fuel economy identical to their corresponding BEVs5 while in charge-depleting mode (that is, when energy is supplied by the battery) and to HEVs in charge-sustaining mode (when energy is supplied by gasoline or diesel). Natural gas vehicles were assumed to have the same efficiency as other gasoline fueled vehicles.
Care was taken to use consistent assumptions across the different technologies. For example, the same vehicle load reduction assumptions (weight, aero, rolling resistance) were applied to all of the drivetrain technology packages.
4The committee accepts the Ricardo results. However, it should be noted that they are based in part on input data that has not been peer reviewed because it is proprietary.
5The BEVs evaluated have a 100 mile range. BEVs with longer range would have substantially heavier battery packs (and supporting structures), adversely affecting vehicle efficiency. PHEVs might have higher electric efficiency than long-range BEVs.