An integrated transportation-power system model for a decarbonizing world

ABSTRACT

Rising demand for electricity from electric vehicles (EVs) will require new paradigms to guarantee reliable and low-cost electricity. This study couples an agent-based travel demand simulator and an electricity grid model to assess the economic costs of supplying power to meet EVs' added demand across the Chicago region. Results suggest that shifting from personal EVs to a fleet of shared, fully-automated all-electric vehicles (SAEVs) could lower per-mile emissions, congestion, and embodied vehicle and charging infrastructure emissions. The results should compel policymakers to shift the cost of providing power onto commercial customers, like electric ride-hail fleets, through price-indexed electricity prices, which can shift charging to off-peak periods or away from resource-scarce hours.

KEYWORDS:

• Integrated modeling
• electrified mobility services
• electrified transportation systems
• infrastructure systems
• agent-based modeling
• transportation simulations
• electric vehicles
• shared autonomous electric vehicles

Acknowledgements

This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under grant number DGE-1610403. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

The work done in this paper was sponsored by the U.S. Department of Energy (DOE) Vehicle Technologies Office (VTO) under the Systems and Modeling for Accelerated Research in Transportation (SMART) Mobility Laboratory Consortium, an initiative of the Energy Efficient Mobility Systems (EEMS) Program. The U.S. Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.

Credit authorship statement

Matthew D. Dean: Conceptualization, Methodology, Software, Formal Analysis, Writing – original draft, Visualization. Krishna Murthy Gurumurthy: Conceptualization, Methodology. Zhi Zhou: Conceptualization, Software. Omer Verbas: Software. Taner Cokyasar: Software. Kara M. Kockelman: Supervision. All authors reviewed and approved the final version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Within the Census Bureau’s defined Chicago–Naperville–Elgin, IL–IN–WI metropolitan statistical area (MSA), 9.2% of households used taxi services at least a few times a month versus 2.8% using this mode a few times a week.

2 On the other hand, studies using agent-based transportation models have relied on exogenously defined electricity prices to estimate charging demand, even though large fleets of SAEVs may influence production costs (Dean et al. 2022; Iacobucci, McLellan, and Tezuka 2018).

3 The turnover rate for a U.S. light-duty vehicle (LDV) is 20 years, which means that high EV sales in a given year will have a delayed response in LDV fleet share. One U.S.-based study estimated that EV sales rising to 89% in 2035 come with a total LDV fleet share of 31% (Woody, Keoleian, and Vaishnav 2023).

4 A metropolitan area like Chicago, with a sprawling suburban area and dense urban core, can benefit from an optimization-based control strategy for SAEVs to rebalance vehicles and ensure equitable short pick-up times. However, the spatial-temporal distribution of trips and sprawling suburban regions creates a challenge in reducing deadheading between paid rides and support trips to charging stations and maintenance depots. Prior work has compared multi-objective optimization-based control dispatch strategies to heuristic-based control strategies (Dean et al. 2022) and found benefits in integrating charging decisions with other idle-vehicle dispatch decisions, like repositioning and vehicle maintenance and cleaning (Dean et al. 2023).

5 As of April 18, 2023, for non-summer months. The characteristic summer day in this study used the following approved rates for summer days: $33.50/MWh for off-peak hours, $80.10/MWh for super-peak hours, and $42.50/MWh for peak hours.

6 This model uses respondent attributes (e.g., age, education, driver's license, household income, and the presence of at least one household worker) and trip-to-trip attributes (e.g., the trip origin's population and job density and the estimated detour delay due to sharing) to decide whether or not to allow sharing.

7 Fig. 4 plots ‘baseline demand,’ which is non-transportation demand minus generation from distributed rooftop solar (seasonally adjusted), since the model does not consider distributed energy resources.

8 According to the 2020 Residential Energy Consumption Survey (RECS) and the 2019 American Community Survey, around 95% of Illinois households use electric air conditioning, while less than 20% have an electric heating system (U.S. Energy Information Administration 2022c).

9 About half of the state’s population lives within the Greater Chicago metro area. Assuming that 20% personal EV adoption increases peak demand by 1-2%, a 100% EV scenario might increase peak demand by 10%-20%.

10 Assuming a 33% increase in energy consumption during the winter (see methods). See Fig. A.8 for a boxplot of daily energy consumption per SAEV by season.

11 This is equal to the average daily electricity consumption of 1,900 U.S. residential customers, according to 2021 data from the U.S. Energy Information Administration (EIA).

Additional information

Funding

This work was supported by National Science Foundation: [Grant Number DGE-1610403]; Vehicle Technologies Office.