Article pubs.acs.org/est
Coordinated EV Adoption: Double-Digit Reductions in Emissions and Fuel Use for $40/Vehicle-Year Dong Gu Choi,†,⊥ Frank Kreikebaum,‡,§ Valerie M. Thomas,†,∥,* and Deepak Divan‡ †
School School § School ∥ School ‡
of of of of
Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States Economics, Georgia Institute of Technology, Atlanta, Georgia 30332, United States Public Policy, Georgia Institute of Technology, Atlanta, Georgia 30332, United States
S Supporting Information *
ABSTRACT: Adoption of electric vehicles (EVs) would affect the costs and sources of electricity and the United States efficiency requirements for conventional vehicles (CVs). We model EV adoption scenarios in each of six regions of the Eastern Interconnection, containing 70% of the United States population. We develop electricity system optimization models at the multidecade, day-ahead, and hour-ahead time scales, incorporating spatial wind energy modeling, endogenous modeling of CV efficiencies, projections for EV efficiencies, and projected CV and EV costs. We find two means to reduce total consumer expenditure (TCE): (i) controlling charge timing and (ii) unlinking the fuel economy regulations for CVs from EVs. Although EVs provide minimal direct GHG reductions, controlled charging provides load flexibility, lowering the cost of renewable electricity. Without EVs, a 33% renewable electricity standard (RES) would cost $193/vehicle-year more than the reference case (10% RES). Combining a 33% RES, EVs with controlled charging and unlinking would reduce combined electric- and vehicle-sector CO2 emissions by 27% and reduce gasoline consumption by 59% for $40/vehicle-year more than the reference case. Coordinating EV adoption with adoption of controlled charging, unlinked fuel economy regulations, and renewable electricity standards would provide low-cost reductions in emissions and fuel usage.
■
INTRODUCTION Numerous renewable electricity technologies are being deployed onto the electric grid. Wind and solar generation are intermittent, and grid integration is technically challenging1 and could be costly.2 Simultaneously, plans for introduction of EVs onto the grid are ramping up,3 and new United States vehicle fuel efficiency standards specify light-duty vehicle (LDV) requirements through model year 2025.4,5 Researchers have studied the emissions impacts of EVs relative to CVs during the use phase, impacts of controlled EV charging, and optimum generation capacity necessary to accommodate EVs.5−17 However, none combine examination of the impact of EVs given the structure of fuel efficiency standards or compare total consumer expenditure under EV and CV scenarios. Here, we address the aggregate implications of EV adoption, fuel economy policy, EV charging methods, and renewable electricity standards (RESs). We simultaneously consider all factors that have been explored singly or in limited combinations in the prior literature. The study area spans the eastern and midwestern portions of the United States, defined by the Eastern Interconnection (Figure 1). Methods. Our modeling approach, described in the Supporting Information, employs a set of coupled models to optimize electricity system generation capacity over multiple © 2013 American Chemical Society
Figure 1. The Eastern Interconnection comprises six regions, each of which are modeled separately in this study: Florida Reliability Coordinating Council (FRCC), Southeast Electric Reliability Council (SERC), Reliability First Corporation (RFC), Southwest Power Pool (SPP), and United States portions of Midwest Reliability Organization (MRO) and Northeast Power Coordinating Council (NPCC).
decades, using linear mixed-integer optimization models taking into account projections of overall demand and a range of Received: Revised: Accepted: Published: 10703
April 17, 2013 July 18, 2013 July 22, 2013 July 22, 2013 dx.doi.org/10.1021/es4016926 | Environ. Sci. Technol. 2013, 47, 10703−10707
Environmental Science & Technology
Article
scenarios for vehicle charging demand.18 We define EVs as including both battery-electric vehicles (BEVs) and plug-in hybrid vehicles (PHEVs). We use data from the 2009 National Household Travel Survey 19 to model the charging demand for cars, vans, and sport utility vehicles. We validate capacity expansion decisions through simulation of the day-ahead unit commitment and hour-ahead economic dispatch processes using data for hourly wind availability in each region and with sensitivity testing for lowest wind and highest demand weeks for each of the six electric system regions.20 We evaluate the required fuel economy of conventional vehicles endogenously, assuming either linked EV and CV fuel economies as in the current policy, or unlinked fuel economies. Under the current policy, there is an inverse relationship between EV adoption levels and CV fuel economy required to comply with the fuel economy standard. We consider two ways of managing the charging of EVs: (i) uncontrolled charging where the driver plugs in the vehicle after completion of the last trip of the day and charging commences immediately and (ii) controlled charging. Under controlled charging, the driver plugs in the vehicle after completion of the last trip of the day, and charging is scheduled to minimize the cost of electricity and ensure that the vehicle is fully charged at the beginning of the first journey of the next day. Uncontrolled charging is the current norm, and we are unaware of any policies requiring controlled charging in the future. We consider three scenarios for EV adoption: (i) that EV sales reach 10% of LDV fleet sales by MY 2025, which is the U.S. EPA and NHTSA projection,21 (ii) that EV sales reach 20% market share by 2025, and (iii) that EV sales reach 100% market share by 2025. At these market share levels, EVs comprise 8.4%, 16.8%, and 80.7% of the LDV fleet, respectively, in 2030. We assume a gasoline price of $4/gallon (2009 $). We consider two RES levels: the current state-level standards with a weighted average of 10% and a 33% level. Only California and Hawaii have legislated binding RES levels of 33% or higher.22 We consider biomass, wind, solar resources, and municipal waste resources within the Eastern Interconnection eligible to meet RESs. We develop region-level renewable resource potentials, assume biomass and solar resources added in a region supply demand in that region, and assume all interregional transfer of wind energy requires construction of dedicated transmission. We then apply the models and assumptions to analyze implications for the 36 states of the Eastern Interconnection region from 2010 to 2030. Electricity and Gasoline Demand. Adoption of EVs at the 10%, 20% and 100% levels results in a 1.7%, 3.3%, or 15.8% increase in overall electric demand in 2030 relative to the reference case with no EVs. Based on the current Corporate Average Fuel Efficiency (CAFE) regulations that link EV and CV fuel economies, the reduction in gasoline demand for the 10%, 20% and 100% EV scenarios in 2030 is 2.9%, 4.4%, and 44.9% respectively relative to the reference case with no EVs, reflecting a combination of the reduced gasoline demand due to EV adoption and the countervailing lower fuel efficiency of the rest of the vehicle fleet under the linked United States fuel efficiency rules. With unlinked EV and CV fuel economies, the 2030 gasoline demand is reduced by 6.2%, 12.4%, and 59.3% for the 10%, 20%, and 100% scenarios respectively compared to the reference case with no EVs. Note that, as shown in Figure 2, even in the reference case with no EVs, the demand for gasoline is projected to decrease, due primarily to the fuel efficiency standards. Also, unlinking fuel economies results in more
Figure 2. Effects of EV adoption and linking fuel efficiency standards on annual gasoline demand, showing the reference case of no EVs and 10%, 20%, and 100% EV adoption scenarios. Linked and unlinked cases shown with solid and dashed lines, respectively.
gasoline savings at the 10% EV adoption level than linking at the 20% level. Electricity Cost. Control of EV charging could provide system reliability and cost benefits by reducing peak load, using lower cost resources, reducing curtailment of intermittent renewable generation, and reducing generator cycling costs. We have evaluated the changes in the electricity generation system for the reference case (no EVs and a 10% RES), a high RES case (no EVs and a 33% RES), and 12 cases comprising of combinations of two RES levels (10% and 33%), two EV charging methods (controlled and uncontrolled), and three EV adoption levels (10%, 20%, and 100%). Figure 3 shows how the
Figure 3. Changes in electrical generating capacity in the Eastern Interconnection in 2030 due to RES adoption or EV charging demand. All values are relative to the 2030 reference case; for example the first bar shows that with a 33% RES and no EVs, more wind and gas generation are required than the reference case. Regional figures are shown in the Supporting Information.
electricity generating capacity of the United States Eastern Interconnection would change for five of these cases. Results are similar, albeit of smaller magnitude, for the 10% and 20% levels. The only exception is the case with 10% EV adoption and the 10% RES; for this case, the controlled case has more coal capacity than the reference case or uncontrolled case. The second bar of Figure 3, representing uncontrolled charging under the 10% RES, shows that capacity would increase relative 10704
dx.doi.org/10.1021/es4016926 | Environ. Sci. Technol. 2013, 47, 10703−10707
Environmental Science & Technology
Article
Figure 4. Total GHG emissions from electricity generation and vehicle gasoline consumption, in the United States Eastern Interconnection in 2030, for 0%, 20%, and 100% EV adoption scenarios. The orange increment, in the 20% EV and 100% EV cases, is from linking fuel efficiency standards.
to the third bar for controlled charging. This is largely because uncontrolled EV charging increases peak electrical demand. A sensitivity study, described in the Supporting Information, shows that the general cost results are independent of natural gas prices over a range of 3.04−9.26 $/MMBtu (2009 $). Greenhouse Gas Emissions (GHG). Figure 4 shows total GHG emissions from electricity generation, LDV fleet gasoline consumption, and LDV fleet manufacturing for the 0%, 20%, and 100% EV cases, with and without linking fuel efficiency standards. To provide conservative results, we used the highest EV manufacturing emissions of the literature we reviewed.7 With no EVs, total GHG emissions are projected to be 12% lower in 2030 than in 2010 due to the fuel efficiency standards, a projected partial shift from coal to natural gas, and the current RES. With linking at the 10% RES level, total GHG emissions are about the same with EVs (bars 3, 4, 7, and 8) as without (bar 1). With unlinking, all cases with EVs (bars 3−10 without the orange increment) have lower GHG emissions than the reference case (bar 1). For all cases except 100% EVs with 10% RES, the controlled case has marginally higher emissions than uncontrolled (bars 4, 6, 8, and 10 vs 3, 5, 7, and 9). The case with 100% EVs, CAFE unlinking, controlled charging, and a 33% RES (bar 10), results in 27% less emissions than the reference case (bar 1). Change in Total Consumer Expenditure. For each case, we calculate the total consumer expenditure (TCE) by vehicle type for vehicles purchased in 2030 (compact car, midsized car, small SUV, large SUV). TCE includes the upfront vehicle cost, charging facility costs, cost of transportation gasoline or electricity, and per-vehicle share of total expenditure for nonEV electricity. The estimated vehicle cost for 2030 CVs ranges from about $17,000 for compact cars to $25,500 for midsized SUVs when there are no EVs, with costs about $300 and $1000 less for the EV 20% and EV 100% scenarios, respectively, in the linked policy scenario due to the correspondingly lower requirements for CV energy efficiency with higher EV market shares. The estimated vehicle cost for 2030 EVs ranges from about $22,000 for compact cars to $34,000 for midsized SUVs (2009 $). The last item is to evaluate the burden on the nontransit sector of any changes in electricity costs. The
detailed calculations for TCE are shown in the Supporting Information. We calculate the average-CV TCE and average-EV TCE based on the current distribution of vehicle types. Finally, for each of the TCEs, we subtract the TCE of the reference case from the calculated TCE to calculate the difference in TCE, ΔTCE. A negative ΔTCE means that the vehicle owner has lower costs under the given scenario than the reference case. Figure 5 shows that unlinked fuel economies and controlled charging, whether deployed independently or together, save
Figure 5. Change in total consumer expenditure for EVs and CVs purchased in 2030 over the vehicle lifetime for 20% and 100% EV adoption scenarios. A negative value means the case has a lower TCE than the reference case.
money for the owners of CVs and EVs. At 20% EV market share, unlinking saves CV owners $240/vehicle-year and EV owners $57/vehicle-year. The savings of CV owners under unlinking show that lower gasoline expenditures with more efficient CVs outweigh the higher cost of more efficient vehicles. The savings from controlled charging range from $14/ vehicle-year for CV owners (10% RES, 20% EV adoption) to $100/vehicle-year (all 100% EV cases). For the cases with EVs, the average EV has a lower TCE than the average CV for all combinations of EV adoption level, linking policy, charging method, and RES level. 10705
dx.doi.org/10.1021/es4016926 | Environ. Sci. Technol. 2013, 47, 10703−10707
Environmental Science & Technology
■
We calculate the average vehicle ΔTCE of each case based on the results of Figure 5 and the market shares of CVs and EVs in each case. With the current 10% RES, all of the unlinked cases have a negative ΔTCE, which means that the average consumer spends less for electricity and transportation in a system with EVs than without. Given the relative attractiveness of EVs compared to CVs, the lower lifetime cost of EVs results in lower average consumer expenditure as EV adoption increases. The ΔTCE for the case with 0% EVs and a 33% RES is $193/vehicle-year. In contrast, the ΔTCE for the case with 100% EVs, unlinked fuel economies, controlled charging, and a 33% RES is $40/vehicle-year. The ΔTCE of $40/vehicleyear with 100% EVs assumes 90% of EVs are BEVs. Alternatively if PHEVs comprise 90% of EVs, the ΔTCE increases to $111/vehicle-year. Benefits of EVs Depend on Fuel Efficiency Rules and Managed Charging. Our results show that EV adoption can reduce the cost of RES compliance, gasoline consumption, and energy system costs, including the costs of GHG reductions. The potential to achieve these benefits can be strongly affected by linkage of CV fuel efficiency, EV adoption levels, and by how EVs are charged. The linking of CV fuel efficiency standards with EV adoption rates does provide flexibility in meeting the standard and may also support development and adoption of EV technologies. However, due to the energy system linkages considered here, the linkage has negative consequences for gasoline consumption and consumer expenditure. Only in the case of high EV market share and a high RES do EVs make a material contribution to GHG reductions. EV adoption can, however, reduce the cost of achieving GHG reductions through a RES. The controlled charging of EVs can reduce electricity costs and improve the integration of wind energy. The benefit per average vehicle is small at low to moderate EV adoption levels because EVs are a small fraction of the fleet. However, the cost benefit of controllability is significant at all EV adoption levels on a per EV basis. The main cost saving is from reduced electric system capacity requirements rather than switching to generation with lower marginal costs. EV adoption links the electricity system with the transportation system. Understanding these linkages provides a basis for developing energy strategies with consideration of cost, technology, and policy goals.
■
ACKNOWLEDGMENTS This work was supported in part by the Intelligent Power Infrastructure Consortium (F.K. and D.D.) and by the Anderson Interface Chair at the Georgia Tech School of Industrial and Systems Engineering (DG.C. and V.T.). We thank Gyungwon Kim and Marilyn Brown for assistance with Figure 1. We also appreciate the input of the Ford Motor Company and Rye Kennedy.
■
REFERENCES
(1) Ackermann, T.; Ancell, G.; Borup, L. D.; Eriksen, P. B.; Ernst, B.; Groome, F.; Lange, M.; Mohrlen, C.; Orths, A. G.; O’Sullivan, J.; et al. Where the wind blows. IEEE Power Energy Mag. 2009, 7 (6), 65−75. (2) EnerNex Corporation. Eastern Wind Integration and Transmission Study; Report NREL/SR-550-47078; The National Renewable Energy Laboratory: Golden, CO, 2010. (3) Michalek, J. J.; Chester, M.; Jaramillo, P.; Samaras, C.; Shiau, C.S. N.; Lave, L. B. Valuation of plug-in vehicle life-cycle air emissionsand oil displacement benefits. Proc. Natl. Acad. Sci 2011, 108 (40), 16554−8. (4) U.S. Department of Transportation, U.S. Environmental Protection Agency, National Highway Traffic Safety Administration. 2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards. Fed Regist. 2012, 77 (199), 62623−63200. (5) Shortt, A.; O’Malley, M. Impact of optimal charging of electric vehicles on future generation portfolios. 2009 IEEE PES/IAS Conference on Sustainable Alternative Energy (SAE), 2009. (6) Göransson, L.; Karlsson, S.; Johnsson, F. Integration of plug-in hybrid electric vehicles in a regional wind-thermal power system. Energy Policy 2010, 38 (10), 5482−92. (7) Hawkins, T. R.; Singh, B.; Majeau-Bettez, G.; Strømman, A. H. Comparative environmental life cycle assessment of conventional and electric vehicles. J. Ind. Ecol. 2013, 17 (1), 53−64. (8) Hadley, S. W.; Tsvetkova, A. A. Potential impacts of plug-in hybrid electric vehicles on regional power generation. Electr. J. 2009, 22 (10), 56−68. (9) Duvall, M.; Knipping, E.; Alexander, M.; Tonachel, L.; Clark C. Environmental Assessment of Plug-In Hybrid Electric Vehicles Volume 1: Nationwide Greenhouse Gas Emissions; EPRI Rep. 1015325; Electric Power Research Institute: Palo Alto, CA, 2007. (10) Peterson, S. B.; Whitacre, J. F.; Apt, J. Net air emissions from electric vehicles: the effect of carbon price and charging strategies. Environ. Sci. Technol. 2011, 45 (5), 1792−7. (11) Baptista, P.; Tomás, M.; Silva., C. Plug-in hybrid fuel cell vehicles market penetration scenarios. Int. J. Hydrogen Energy 2010, 35 (18), 10024−30. (12) Notter, D. A.; Gauch, M.; Widmer, R.; Wäger, P.; Stamp, A.; Zah, R.; Althaus, H. J. Contribution of Li-ion batteries to the environmental impact of electric vehicles. Environ. Sci. Technol. 2010, 44 (17), 6550−6. (13) Samaras, C.; Meisterling, K. Life cycle assessment of greenhouse gas emissions from plug-in hybrid vehicles: Implications for policy. Environ. Sci. Technol. 2008, 42 (9), 3170−6. (14) Weiller, C. Plug-in hybrid electric vehicle impacts on hourly electricity demand in the United States. Energy Policy 2011, 39 (6), 3766−78. (15) Sioshansi, R.; Denholm, P. Emissions impacts and benefits of plug-in hybrid electric vehicles and vehicle-to-grid services. Environ. Sci. Technol. 2009, 43 (4), 1199−204. (16) Sioshansi, R.; Miller, J. Plug-in hybrid electric vehicles can be clean and economical in dirty power systems. Energy Policy 2011, 39 (10), 6151−61. (17) Sioshansi, R.; Denholm, P. The value of plug-in hybrid electric vehicles as grid resources. Energy J. 2010, 31 (3), 1−23. (18) Choi, D. G.; Thomas, V. M. An electricity generation planning model incorporating demand response. Energy Policy 2012, 42, 429− 441.
ASSOCIATED CONTENT
S Supporting Information *
Details on data and analysis, additional figures, regional results, and sensitivity studies are provided. This material is available free of charge via the Internet at http://pubs.acs.org.
■
Article
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Present Address ⊥
Dong Gu Choi: Korea Institute of Energy Research, 152 Gajeong-ro, Yuseong-go, Daejeon 305−343, Republic of Korea. Notes
The authors declare no competing financial interest. 10706
dx.doi.org/10.1021/es4016926 | Environ. Sci. Technol. 2013, 47, 10703−10707
Environmental Science & Technology
Article
(19) National Household Travel Survey, 2009. U.S. Department of Transportation, U.S. Federal Highway Administration. http://nhts. ornl.gov. (20) Eastern Wind Integration and Transmission Study Dataset, 2010. U.S. Department of Energy. http://www.nrel.gov/electricity/ transmission/eastern_wind_dataset.html. (21) Final Rulemaking to Establish Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards: Joint Technical Support Document; Report EPA-420-R-10-901; U.S Environmental Protection Agency: Wshington, DC, 2010. (22) Renewable Portfolio Standard Policies, 2013. U.S. Department of Energy. www.dsireusa.org/documents/summarymaps/RPS_map. pdf.
10707
dx.doi.org/10.1021/es4016926 | Environ. Sci. Technol. 2013, 47, 10703−10707
