ACS Publications. Most Trusted. Most Cited. Most Read
Private versus Shared, Automated Electric Vehicles for U.S. Personal Mobility: Energy Use, Greenhouse Gas Emissions, Grid Integration, and Cost Impacts
More

OR SEARCH CITATIONS

My Activity
Publications

Figure 1Loading Img
Download Hi-Res ImageDownload to MS-PowerPointCite This:Environ. Sci. Technol. 2021, 55, 5, 3229-3239
  • Open Access
Energy and Climate

Private versus Shared, Automated Electric Vehicles for U.S. Personal Mobility: Energy Use, Greenhouse Gas Emissions, Grid Integration, and Cost Impacts
Click to copy article linkArticle link copied!

Open PDFSupporting Information (1)

Environmental Science & Technology

Cite this: Environ. Sci. Technol. 2021, 55, 5, 3229–3239
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.est.0c06655
Published February 10, 2021

Copyright © 2021 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Abstract

Click to copy section linkSection link copied!
High Resolution Image
Download MS PowerPoint Slide

Transportation is the fastest-growing source of greenhouse gas (GHG) emissions and energy consumption globally. While the convergence of shared mobility, vehicle automation, and electrification has the potential to drastically reduce transportation impacts, it requires careful integration with rapidly evolving electricity systems. Here, we examine these interactions using a U.S.-wide simulation framework encompassing private electric vehicles (EVs), shared automated EVs (SAEVs), charging infrastructure, controlled EV charging, and a grid economic dispatch model to simulate personal mobility exclusively using EVs. We find that private EVs with uncontrolled charging would reduce GHG emissions by 46% compared to gasoline vehicles. Private EVs with fleetwide controlled charging would achieve a 49% reduction in emissions from baseline and reduce peak charging demand by 53% from the uncontrolled scenario. We also find that an SAEV fleet 9% the size of today’s active vehicle fleet can satisfy trip demand with only 2.6 million chargers (0.2 per EV). Such an SAEV fleet would achieve a 70% reduction in GHG emissions at 41% of the lifecycle cost as a private EV fleet with controlled charging. The emissions and cost advantage of SAEVs is primarily due to reduced vehicle manufacturing compared with private EVs.

This publication is licensed under

CC-BY-NC-ND 4.0 .
  • cc licence
  • by licence
  • nc licence
  • nd licence
Copyright © 2021 The Authors. Published by American Chemical Society

Subjects

what are subjects

Introduction

Click to copy section linkSection link copied!
The transportation sector represents the fastest-growing segment of the world’s GHG emissions, accounting for 23% of global energy-related carbon dioxide emissions in 2014, (1) with car sales set to more than double by 2050. (2) In the U.S., transportation has become the single greatest source of emissions, accounting for almost one-third of carbon dioxide emissions nationwide. (3) EVs have grown to more than 5 million (M) vehicles worldwide in 2019, (4) driven in part by rapidly falling battery costs, (5) but represent a tiny fraction of personal mobility. Prior research has proven the capability of EVs to meet the travel needs of the majority of drivers in the U.S. (6,7) Coupled with clean electricity, EVs have the potential to dramatically reduce the GHG intensity of private transportation. (8,9) However, adoption of EVs has been limited by factors including a sparse charging network, slow charging, short vehicle range, and higher capital costs compared to other types of vehicles. (10−13)
The global proliferation of ride-hailing, micromobility, and other shared mobility services alongside the prospect of widespread vehicle automation (14) has the potential to transform the transportation sector. Indeed, automated vehicles are already serving U.S. passengers without a backup human driver. (15) There is growing consensus that, without sharing rides (i.e., more than one party or passenger per vehicle), vehicle automation could lead to increases in vehicle miles traveled, congestion, energy consumption, and emissions. (16−19) However, synergy between sharing, automation, and electrification (17,20) could result in deep reductions in GHG emissions. (14)
Centrally managed SAEVs could offer services similar to those provided by current ride-hailing companies, but at much lower cost and GHG intensity. (21) Because each SAEV only needs enough battery range for the trip requested and charging can be split over many short periods between trips, shared mobility could reduce requirements for battery range. This shift could overcome barriers of slow charging speeds and high capital costs while reducing GHG emissions. (14,22,23)
Public charging infrastructure is critical to accelerating EV adoption. (24−26) However, the business case for private sector investment in chargers is weak in the context of personally owned EVs. (27) EV charging also introduces significant new loads to electricity systems already challenged both by peak electricity demand and increasing levels of intermittent renewable generation. (28,29)
Many studies have assessed the benefits to electricity grids of controlled EV charging, (30−33) which can provide benefits of decarbonizing transportation and lowering costs of renewable integration and energy storage. (34−37) Preliminary investigations of the potential for SAEV-grid integration highlighted this potential. (38,39) However, only a handful of studies simulate EV-grid interactions at a national (or multinational) level in Australia (40) or western Europe. (41−45)
The current study expands upon prior work (46) using a hybrid modeling framework (Grid-integrated Electric Mobility or GEM) that combines individual vehicle trips, parametrized agent-based model outputs, a cost model for vehicles and charging behavior, and a national-level electricity production cost model. GEM co-optimizes the allocation of SAEV vehicles and charging infrastructure along with charge scheduling and economic dispatch of grid generators to find the minimal cost combination of vehicles, chargers, and operations to satisfy a given demand for trips. The result includes optimized EV fleets and associated charging infrastructure spanning both urban and rural areas, which serve the mobility needs of the entire U.S. population under various assumptions about automation, sharing, and charging strategies. We use GEM to explore how these parameters affect hourly electric load patterns, peak power demand, EV battery capacity, fleet size requirements, charger power levels and quantity, total cost of fleet ownership, renewables curtailment, and GHG emissions, in a future personal transportation system composed of any combination of privately owned EVs and centrally managed SAEVs.
Our goal is both to compare private and shared fleet electrification, as well as the interaction between these fleets and a future grid with higher penetrations of nondispatchable energy sources. The interactions between power generation on the grid and charging behavior of private and shared vehicles are complex. There are substantial cost savings that can be achieved by managing the charging patterns of EV fleets, but it would not be practical for these to come at the cost of decreased mobility. Therefore, our model must be capable of quantifying the temporal flexibility of charging the fleets in a manner that still satisfies mobility as well as resolve how this flexibility will align with low cost generation on the grid.
Furthermore, there is flexibility that can be added to the system by overbuilding, for example, by manufacturing more vehicles or increasing their range or adding more chargers or higher power charging. There are trade-offs inherent in the decision to build more flexibility into the system, they require upfront investment but yield operational benefits by accessing lower cost energy. Our methodological approach is to model all of these factors endogenously to capture these trade-offs in both the system design and operation.

Methods

Click to copy section linkSection link copied!
We leverage previous work (46) where the authors developed an optimization model that examines SAEVs in the context of temporally varying electricity prices. In this study, we have extended the model to jointly optimize mobility (both SAEVs and private EVs), charging scheduling, and the power sector for the entire U.S. by coupling our previous model to the Grid Operation Optimized Dispatch (GOOD) U.S.-wide electricity model. (47) This combined Grid-integrated Electric Mobility (GEM) model treats the size of the SAEV fleet and the amount of charging infrastructure as continuous decision variables (relaxing the problem from mixed-integer to quadratic), allowing for heterogeneous vehicle ranges and charger levels. The model minimizes operational costs by choosing the timing of fleet recharge while requiring that mobility demand be served, energy is conserved, and generation assets on the grid are dispatched in merit order. The future electricity capacity grid mix is an exogenous input into GEM and is based on NREL’s Renewable Energy Futures study. (48) SAEV fleet planning costs are simultaneously minimized by amortizing the cost of the fleet and charging infrastructure to a daily time period.
In addition to extending the optimization model, we curated a set of empirically derived inputs and assumptions for the model application (details in subsequent sections and the Supporting Information). Several of our assumptions were also developed through detailed, agent-based simulation modeling using the Routing and Infrastructure for Shared Electric vehicles (RISE) model (23) and from simulations completed by the National Renewable Energy Laboratory using EVI-Pro. (49)
In Figure 1, we illustrate the source of all major model inputs and assumptions including intermediate modeling and analysis used in their derivation. Each model input is described in further detail below, beginning with the specification of the optimization model.

Figure 1

Figure 1. Sources of data (blue), data processing (dark red), models (light red), intermediate data (gray), and model outputs (yellow) in the overall modeling and processing workflow.

High Resolution Image
Download MS PowerPoint Slide
The GEM model is publicly available as open-source software under a permissive license. (50) Readers are encouraged to review the code-base, run their own simulation studies, and provide feedback or make contributions to the ongoing project. A snapshot of the exact code used to produce the results in this manuscript is also available. (51)
The dimensions of the model include time t, mobility region r, grid region i, vehicle battery size b, charger level l, trip distance d, and electricity generator g. The model is a quadratically constrained program and can be efficiently solved with a second-order cone programming solver.

Objective Function

The objective function minimizes the amortized daily cost of fleet and infrastructure capital, and fleet and electricity grid operations.
(1)
where Ctrd is the demand charge or capacity cost to use the grid and Ctrm is vehicle maintenance cost in hour t and mobility region r, Crc is the amortized daily charging infrastructure capital cost, Crv is the amortized daily fleet capital cost, n is the number of days in the simulation time horizon, Gg is the electricity produced by generator g, Cgg is the cost of producing a unit of energy by generator g, Ti,t,i′ is the electricity transmitted from grid region i to grid region i′, and Ci,t,i′t is the marginal cost of transmission.
The system minimizes the objective function across a large number of decision variables: the energy to charge and consumed by vehicles at different points in time, Pbt̂lr and Ebdt̂r, respectively; demand charge costs, Ctrd; vehicle maintenance costs, Ctrm; maximum power demand, Prmax; size of the SAEV fleet and capital cost of the fleet, Vbr* and Crv; the number of vehicles charging, moving and idle: Vbdtlc, Vbdtrm, and Vbtri, respectively; the number of chargers and capital cost of the infrastructure: Nlr and CrC; demand allocation by battery type, Dbdtr; temporal power profile of private vehicles, Ptrprivate; generation by power plants, Ggt; and transmission of power between regions, Titi′
The optimization is subject to a number of constraints that enforce realistic operations, as described in the Supporting Information.

NHTS Data

We applied the model at a national level based on estimates of hourly demand for private vehicle trips derived from the 2017 National Household Transportation Survey (NHTS). (52) NHTS respondents log trip distance, timing, and vehicle type for all household members on a specified day. The responses are weighted according to demographics to yield typical daily mobility profiles, by distance and trip start time. We computed these profiles by day type (weekday vs weekend), geographic region, and season, across the entire U.S.
To produce our trip-demand model inputs, we partitioned the country into 13 broad geographic regions, made up of the nine U.S. Census Divisions: New England (NE), Mid-Atlantic (MAT), South Atlantic (SAT), East-North-Central (ENC), West-North-Central (WNC), East-South-Central (ESC), West-South-Central (WSC), Mountain (MTN), and Pacific (PAC). In addition, the four largest states (California, Florida, New York, and Texas) are separated into their own individual regions. We refer to these regions as the Census-Division-Large-State (CDLS) partition. In addition, we subdivided the trips into “urban” and “rural” subregions (see Supporting Information for details). This yields a total of 26 regional data sets: 13 CDLS regions, each with urban and rural subregions.
Because mobility demand varies throughout the year, both in terms of the number of trips and their distribution by distance, it was important to capture seasonal variation in trip demand in addition to the hourly and regional distributions. We thus divided the data into four three-month blocks: December–February (winter), March–May (spring), June–August (summer), and September–November (autumn). The final output of this analysis was a set of hourly mobility-demand profiles, over the course of a typical day, subdivided by trip distance, day type, geographic region, and season.

Correction Factors Using RISE

Some of the parameters in the optimization model were determined using a spatially explicit, agent-based simulation tool. The parameters included deadheading ratios (the ratio of empty vehicle miles traveled to miles traveled with passengers), as well as oversizing requirements due to spatial mismatch (i.e., more chargers are needed in a region than the maximum coincident demand for charging; likewise extra battery capacity is required to access charging stations and for contingencies). To do this, we coupled trip data obtained from StreetLight Data with an agent-based fleet simulation framework called Routing and Infrastructure for Shared Electric vehicles (RISE) model, originally described in Bauer et al. (23) StreetLight Data is a company that aggregates data from cell phones and GPS devices to produce transportation metrics such as travel times and volume of travel. The resulting trip data sets spanned nine combined statistical areas (CSAs) throughout the U.S. covering distinct census divisions and population densities. For full details of the preprocessing methodology, see the Supporting Information.
Simulations were conducted for each city with 20k, 40k, 100k, 200k, 400k, and 800k trips, and with both 15 kW and 50 kW charging power. Locations of chargers were determined by k-means clustering of trip origins and destinations, which was determined to work as effectively as the siting algorithm described in Bauer et al. (23)
The model proceeds chronologically over 1 day of data, repeating until the fleet’s aggregate battery capacity at the end of the day is within 5% of that at the beginning of the day. In each minute, trips are assigned to the nearest vehicle, and idle vehicles are routed to charge or rebalanced in anticipation of future demand (see reference for details). Travel times and distances between each taxi and trip or charging point are imputed by drawing random values from the corresponding distribution obtained from StreetLight Data. To ensure a reasonable relationship between time, distance, and speed for each trip, distances are resorted in order to best match the relationship between draws for duration and speed. If a trip can only be served by a vehicle with insufficient battery capacity, the vehicle’s range is increased by 50-mi increments until capacity is adequate. If no vehicle can serve a trip within a 10 min wait time, a new vehicle is added to the fleet. Thus, both battery range and fleet size increase organically over the course of the simulation, providing estimates of the minimum values required to serve demand. Vehicles serving trips that end outside the CSA are removed from the fleet; trips starting outside the CSA are served by new vehicles that are then added to the fleet.
For each simulation, we recorded the empty distance and time traveled to each passenger pick-up and charging event to determine the average empty-to-occupied trip ratios for each city, often referred to as the “deadheading” ratio. As shown in Figure 3 in the Supporting Information, we found that deadheading ratios increase roughly with the square root of area per trip. Given that vehicle speeds are different when empty and when serving trips, we estimated separate ratios for time spent empty and energy consumed while empty. These results were used to correlate key system deadheading ratios and oversizing requirements to travel demand spatial density, which were then extrapolated to each region in GEM. The extrapolated values for the six parameters are listed in a table in the Supporting Information.

Private EV Fleet

Electric load from private EVs and its associated flexibility is an exogenous input to GEM. We prepared these load profiles by analysis of simulated individual charging sessions from the EVI-Pro infrastructure planning model. This tool, developed by the California Energy Commission and the National Renewable Energy Laboratory, is based on travel diaries from the California Household Travel Survey and can simulate the mobility and charging behavior of vehicles over a variety of electric vehicle types, charging infrastructure levels, and assumptions about traveler preferences (e.g., preferences for home charging vs workplace charging). Each run of EVI-Pro produced daily charging sessions covering approximately 50 000 travelers. Using a sampling tool developed by the Schatz Energy Research Center at Humboldt State University, we constructed regional fleets of private EVs that match the size of the simulated private fleets in GEM, as well as the average daily mileage of these fleets consistent with the NHTS data.
The individual charging sessions contained plug-in, charging completion, and plug-out times. Using the virtual battery approach, (53) we aggregated these sessions into constraints that represented the maximum and minimum power and energy that must be delivered to the private fleet in each hour of the day. The constraints were constructed so that, if satisfied, it was possible to meet the energy demands of every individual charging session by the plug-out time. Finally, we parametrized the creation of the constraints to assess the impact of partial flexibility in the private charging demand. In other words, if the flexibility parameter was set to 0%, then uncontrolled charging was assumed and the lower cumulative energy bound (eq 22 in Supporting Information) was equal to the upper cumulative energy bound (eq 23 in Supporting Information). If flexibility was set to 100%, then we assumed all charging sessions could be delayed to the maximum possible moment while still delivering the original energy to the fleet. Any value between 0% and 100% was a direct scaling of the degree of flexibility between the two extremes; quantitatively, this meant the lower cumulative energy bound is proportionally closer to the upper bound than the 100% scenario.

Electricity Grid Modeling

The electricity grid model in GEM is developed based on an economic dispatch model known as the Grid Operation Optimized Dispatch (GOOD) model. (47) The GOOD model simulates the operation of power generators across the country by dispatching specific generation assets to meet demand load, the order of which is determined by a “bidding” process, where the lowest cost generation on the margin is the next to participate in providing power. The model also includes transmission between regions on a substate basis, which is then aggregated up to the CDLS regions used in GEM. Within a region, generators are aggregated by their fuel type, variable costs, and CO2 emissions rates.
The data regarding the generators to model the electricity grid are combined from the EPA NEEDS (v5.15) and eGrid 2016 databases. These contain a large number of attributes for every generator, including capacity, fuel costs, operations and maintenance (O&M) costs, emissions rates, and heat rates. These attributes allow us to characterize the operations of each generator in the dispatch model. We employ representative hourly profiles across a full year for wind and solar resources disaggregated by the substate regions and aggregated by the weighted capacity in each region. These data are not averaged solar and wind profiles, which would smooth out intermittency. The transmission between each region has an associated capacity (not always bidirectional) that includes wheeling costs in certain pairwise regions. The dispatch model has been validated against generation and emissions from the power sector in 2018. (47)
While the mobility side of GEM provides an endogenous “input” into the electricity grid side of the optimization via the electricity demand associated with vehicle charging, this is in addition to an exogenously specified baseline demand from all other non-EV loads. This baseline is obtained from EPA NEEDS (v5.15) and contains electricity load demand on an hourly basis throughout the year in all substate regions across the U.S. Once again, these loads are aggregated to the CDLS regions used in GEM. The economic dispatch portion of GEM simultaneously fulfills charging demand from the baseline load and from the EV charging load; it is therefore able to influence the mobility-system decision making process under the assumption that the costs associated with charging would be real-time costs. To the extent that the system can reduce costs by promoting charging at certain times of the day, GEM considers this decision among the suite of cost trade-offs contained in the optimization formulation.

Key Assumptions

In Table 1, we list key assumptions used for the Base scenario of the optimization model.
Table 1. Key Modeling Assumptions Used to Define the Base Scenario
inputsymbolvalues
charger types and powerγlL010 = 10 kW
L020 = 20 kW
L050 = 50 kW
L100 = 100 kW
L250 = 250 kW
charger capital costϕlcL010 = $5k
L020 = $11k
L050 = $31k
L100 = $77k
L250 = $305k
charger lifetimeLc10 years
charger distribution factorδl1.0 for all types
demand charge priceβr$7.7/kW/month
annual discount rater0.05
sharing factorσd1.5
vehicle capital costϕv$30 000 (includes cost of automation)
vehicle daily fixed O&Mϕomv$1.64
vehicle per-mile O&Mβv$0.09
battery capital costϕb$150/kWh
vehicle/battery lifetimeLv2 years
Lb
battery capacityBb75 mi range = 19.7 kWh,
150 mi range = 41.1 kWh
225 mi range = 64.4 kWh
300 mi range = 89.4 kWh
400 mi range = 124.0 kWh
conversion efficiencyηb75 mi range = 310 Wh/mi
150 mi range = 324 Wh/mi
225 mi range = 338 Wh/mi
300 mi range = 351 Wh/mi
400 mi range = 353 Wh/mi
speed by distance binsνdtr0–2 mi = 18 mph
2–5 mi = 22 mph
5–10 mi = 32 mph
10–20 mi = 38 mph
20–30 mi = 40 mph
30–50 mi = 45 mph
50–100 mi = 48 mph
100–300 mi = 48 mph

Results

Click to copy section linkSection link copied!
Our results reveal the tremendous impact that vehicle electrification can have on the transportation system. Private EVs with uncontrolled charging would reduce GHG emissions by 46% from a baseline of 1134 million metric tons CO2-equivalent (MtCO2eq) per year associated with a fleet of personally owned gasoline-fueled vehicles (54) (including vehicle manufacturing emissions). Private EVs with fleetwide controlled charging would achieve a 49% reduction in emissions from baseline and reduce peak charging demand by 53% from the uncontrolled scenario, achieving a 38% reduction in total cost of ownership from a gasoline baseline.
If SAEVs serve all mobility currently served by light-duty vehicles, GHG emissions would be reduced by 70% from a gasoline baseline. Moreover, the total cost of ownership ($1.570 trillion/year or $4800 per capita) would decrease 74%, and peak electricity demand from an SAEV fleet would be 47% lower than the peak from a private EV fleet with uncontrolled charging. In other words, over 80 GW of generation capacity (or demand-side flexibility)—greater than half of current standalone U.S. natural gas combustion turbine capacity (55)—would not be needed to support full electrification.
We present our results as functions of two key parameters that are observed to have the largest impacts: the fraction of trips served by SAEVs, denoted by S (with the remaining fraction served by private EVs) and the fraction of private EVs that engage in controlled charging, denoted by C. Note that simulations assumed all SAEV charging was controlled based on the presumption that all aspects of automated fleet operations will likely be centrally managed.

Time Series of EV Charging and Vehicle State

Figure 2 depicts a reference scenario,where half of nationwide light-duty mobility is served by SAEVs (S = 50%) and the other half is served by privately owned EVs with no controlled charging (C = 0%). Panel a shows charging load versus time over 12 representative days consisting of 1 weekend day and two weekdays from each of four months spaced throughout the year to reflect seasonal variability. In this scenario, we see large differences between the temporal patterns of SAEV and private EV charging. While private EV charging load peaks sharply in the evening, with a much smaller secondary peak in late morning (on weekdays) and the lowest demand in early morning, SAEV charging loads are spread over many more hours of the day, with a broad peak during daytime hours and almost no charging during the private EV evening peak. To satisfy demand, private EV charging is composed of 20% level 1 (1.5 kW) and 80% level 2 (6.7 kW), whereas the cost-minimizing SAEV charging profile is fulfilled by 4% 10 kW, 38% 20 kW, and 56% 50 kW charging, with less than 3% fulfilled by faster levels (100–250 kW).

Figure 2

Figure 2. Reference scenario for S = 50% and C = 0% (half of all trips are satisfied by SAEVs, and half by private EVs with uncontrolled charging), showing (a) charging load by power level, (b) EV state (charging, moving, or idle) by battery range for SAEVs, and (c) EV state (charging, moving, or idle) for private EVs. The scenario was run over 12 days: three consecutive days—Sunday, Monday, and Tuesday—in each season.

High Resolution Image
Download MS PowerPoint Slide
Panel b shows vehicle activity (charging, moving, or idle) disaggregated by battery range for SAEVs, and panel c shows aggregate vehicle status for private EVs. Note that we have not included trips of distances greater than 300 mi. in this study, as such trips are exceedingly rare, making up <0.1% of all trips. Our model also neglects complications from trips that are longer in distance than the highest range vehicle (225 mi.). Trips longer than 225 miles make up 0.17% of all trips. In reality some number of longer range vehicles would be necessary, or vehicles would need to recharge en-route to serve all trips.
Panels b and c indicate that there are over ten times more private vehicles than SAEVs; overall, SAEVs are moving 38% and charging 17% of the time, whereas private EVs are moving 4% and charging 40% of the time. The private EVs follow a typical uncontrolled load shape, with a very high peak in the evening associated with plug-in events happening after the evening commute. There is an analogous smaller peak in the midmorning on weekdays as drivers arrive to work and use workplace charging. We see that very few SAEVs are idle between early morning and early evening, with the highest fraction of moving vehicles after daytime charging ends around 4 p.m. SAEV movement drops to nearly zero after midnight, followed by a resumption of charging activity. However, at most 27% of total SAEVs (1.7 M out of 6.2 M) are charging at any given time, indicating there is sufficient charging capacity to allow the majority of vehicles to be available. The two peak moving periods on weekdays correspond to the morning and evening commutes. During the evening peak period, the SAEV fleet is nearly fully utilized, aside from a small fraction of vehicles that are charging (more SAEVs are charging during the morning than the evening commute peak).
The temporal charging load patterns change dramatically as S and C change. Figure 3 shows 3 days of charging load from each of a set of model scenarios, where S and C each range from 0% to 100%. The highest peak loads are found in the upper left panel (a) due to exclusively uncontrolled private EV charging, with a peak of 162 GW, slightly more than the current total U.S. capacity for standalone natural gas combustion turbines. (55) As either S or C increases, peak loads diminish, approaching a minimum level of 71 GW in panels l and p, where load is nearly evenly spread throughout the day. With no shared vehicles (panels a–d), charging load profiles are smoothed as C increases from 0% to 100%, leading to an increase in the load factor from EVs (the maximum power divided by the average power), which steadily increases from 41% to 85%. Over the course of a year, electricity demand from vehicle charging is ∼540 TWh or nearly 80% of total renewable generation including hydropower and accounts for 13.5% of total projected electricity demand including EVs. In panel l, the 71 GW peak constitutes a little over 11% of the peak demand in the U.S. (∼620 GW), though these peaks are not coincident. In panel q, when S = 100%, the peak load is 45% lower than that of the panel a peak and is spread over many more hours, peaking approximately at midday with a sharp minimum in late afternoon coinciding with a peak in vehicle travel. Of note, scenario S = 100% does not result in the smoothest load, a result that is explained further in the discussion of Figure 4c.

Figure 3

Figure 3. Three days (1 weekend day, followed by two weekdays) of simulated EV charging load as functions of SAEV trip fraction (S) and private EV controlled charging fraction (C), disaggregated by charger level, for both private EVs and SAEVs. Each panel is labeled by the value for S and C used in the simulation. Panels a–d: S = 0%, C = 0–100%. Panels e–h: S = 25%, C = 0–100%. Panels i–l: S = 50%, C = 0–100%. Panels m–p: S = 75%, C = 0–100%. Panel q: S = 100% (value of C is irrelevant as there are no private EVs).

High Resolution Image
Download MS PowerPoint Slide

Figure 4

Figure 4. (a) Fleet size, (b) numbers of chargers, (c) peak power demand, (d) total cost of ownership, and (e) consequential GHG emissions vs fraction of SAEV trips (S) with C = 0%.

High Resolution Image
Download MS PowerPoint Slide

Annual Average Results

Figure 4 shows results for key outputs averaged over time and geography, displayed across the full range of S, for C = 0% (uncontrolled private EV charging). This slice focuses on the impact of SAEVs because they show the greatest variation in outcomes. In the Supporting Information, we present results analogous to Figure 4 but for a private EV fleet (with S = 0%), varying C across its range to show how controlled EV charging influences system outcomes.
Figure 4a shows the fleet size, which decreases by an order of magnitude from ∼145 M vehicles in the S = 0% case (these 145 M vehicles are “active” vehicles used on a typical weekday, and represent ∼56% of the current stock of U.S. light-duty vehicles (54)) to ∼12 M vehicles in the S = 100% case. This occurs because the utilization per SAEV is about 12 times higher than a private EV, due both to increased time spent moving, the number of passengers per trip, and faster recharging times. On average across the scenarios, the cost-minimizing SAEV fleet is composed of 27% vehicles with a range of 75 miles, 69% with 150 miles, and 4% with 225 miles. Private EV ranges were a scenario assumption, not a result of the optimization, so their distribution reflects a projection that average range will increase steadily but modestly over time.
Figure 4b shows the number of chargers needed. As with fleet size, there are far more chargers when S = 0% (195M) than S = 100% (2.6M), reflecting much higher utilization among SAEV chargers. These chargers consist of roughly half lower power levels (≤20 kW) and half 50 kW DC fast chargers, with about 1% of the network consisting of faster 100 and 250 kW chargers.
Peak power, shown in Figure 4c, also decreases substantially as S increases: Peak demand is 161 GW at S = 0% and almost half this (∼89 GW) when S = 100%. The dramatic increase in SAEV contribution to peak power between S = 50% and 75% is due to the fact that when S = 50% the SAEV loads can still “valley fill” within the private EV load (see Figure 3, panel i vs m), whereas when S = 75% the SAEV load becomes dominant throughout the day. The increase in peak demand from S = 75% to 100% is driven by an increase in cost minimization opportunities through charge scheduling available to a full SAEV fleet that are not possible when some private EVs are still engaging in uncontrolled charging.
Similar to peak power, the total cost of ownership for the EV fleet shown in Figure 4d decreases by 56% as S increases from 0% to 100%. Across this range, both energy and charger costs are minor components of total cost, which is dominated by vehicle purchase cost. However, because of the higher utilization and smaller average battery sizes of SAEV fleets, total vehicle cost for S = 100% is roughly half that of the S = 0% scenario.
Figure 4e shows the consequential GHG emissions of charging the EV fleet. Emissions fall from ∼600 MtCO2eq in the S = 0% case to ∼340 MtCO2eq in the S = 100% case. By comparison, emissions from a gasoline vehicle fleet including manufacturing is 1134 MtCO2eq. The optimization does not directly minimize emissions (they are not in the objective function) but an overall positive correlation between cost and emissions leads to a reduction in fossil GHG emissions arising from greater use of controlled charging. In addition SAEVs consume less total energy and the vehicle manufacturing emissions are lower because of the smaller (even though more rapidly overturning) SAEV fleet. The result is a 43% reduction in GHG emissions between the S = 0% and 100% cases or a 70% reduction in emissions from a gasoline-fueled baseline.

Urban versus Rural Fleets

To build intuition for how key model results are interrelated and to examine how infrastructure and fleet requirements might differ between regions, we compare simulation outputs between urban and rural geographies. Figure 5 shows national-level SAEV results disaggregated by urban versus a rural region across the range S = 25–100% for two parameters of interest: chargers per vehicle (disaggregated by charging speed) and vehicles per trip (disaggregated by battery size). These two parameters were chosen because they showed the largest differences between urban and rural regions and trends versus S.

Figure 5

Figure 5. (a) Optimal fleet size (in vehicles per trip) and (b) charging infrastructure requirements (in chargers per vehicle) disaggregated by urban and rural regions for C = 0% and with S ranging from 25% to 100%.

High Resolution Image
Download MS PowerPoint Slide
We found that rural regions required a higher fraction of chargers per vehicle than urban regions but that both fractions decreased as S approached 100%, suggesting higher utilization in a larger network. Moreover, rural regions required more 10–20 kW, as well as 100 kW chargers than urban regions, which required more 50 kW chargers. These differences became more pronounced as S progressed from 25% to 100%. The 100 kW chargers accounted for between 10% and 14% of energy consumption among rural vehicles, which in aggregate accounted for about 2% of total vehicle energy consumption and miles traveled.
The total ratio of vehicles per trip is slightly lower in rural versus urban regions, indicating higher utilization, a consequence of longer-range vehicles and more, faster chargers which combine to reduce vehicle time spent out of service. The vehicles per trip ratio is largely invariant as S increases in both regions. By contrast, the distributions of battery capacities are quite distinct: there are many more longer-range (225-mi.) vehicles in rural regions as compared to urban, which have many more 75-mi. range vehicles and virtually no 225-mi. range vehicles; the fraction of medium range (150-mi.) vehicles are mostly equivalent in both regions. Interestingly, as S increases from 25% to 100%, rural regions shift away from the small number of 75-mi. vehicles toward 150-mi. vehicles, with the fraction of 225-mi. vehicles remaining almost the same. A similar, though harder to discern, trend appeared in urban regions. The increased trip densities at higher values of S favor longer range vehicles (which are more expensive and have lower energy efficiency), which enable the lower charger densities discussed in panel a.

Geographic Differences in Electricity Mix and GHG Emissions

There was very little difference in most key variables by geographic region, with the exception of electric generation mix and GHG emissions. Because of much more coal use in certain regions and much more renewable generation in others, we expected large differences in both of these outputs.

Additional Sensitivity Runs

The results presented above represent two of seven parameters that we systematically varied. In the Supporting Information, we present the results of varying other important parameters: the sharing factor, vehicle capital cost, battery capital cost, battery lifetime, and vehicle energy conversion efficiency.

Discussion

Click to copy section linkSection link copied!
Our results paint a very different picture of the future transportation system from the one we see today. Private EVs with 100% controlled charging could reduce GHG emissions by 53% compared to gasoline vehicles and reduce the total cost of ownership by 38%. In addition, they are able to reduce peak demand from charging by 53% from uncontrolled EVs, which is 6 percentage point greater reduction than a system composed entirely of SAEVs.
When all trips taken in light-duty vehicles are provided by SAEVs, personal mobility can be fulfilled with 12 M vehicles, 9% of the active vehicles required when they are privately owned, and with only 2.6 M chargers (0.2 per EV). From an economic standpoint, sharing and automation provide tremendous benefits, enabling a 74% reduction in total fleet cost compared to gasoline vehicles today. The operation of SAEVs and controlled charging of individual vehicles are able to reduce peak charging load by 47% from what would otherwise be present when S = 0% and C = 0%. This can significantly improve both the efficiency and emissions rate of fossil generation (by shifting to lower GHG-emitting fossil generators) while simultaneously better utilizing solar and wind resources (due to flexibility in charging times), resulting in a reduction in solar curtailment by about one-third.
One of the driving forces behind our results is that vehicle autonomy enables significantly higher utilization of any given vehicle. Private EVs remain unused most of the time because most people’s mobility needs can be satisfied in just a small fraction of the day. For SAEVs, vehicles can fulfill multiple individuals’ travel needs throughout the day, remaining highly utilized for many hours, yet with ample time available to relocate to new passengers or to recharge. Shared ride-hailing trips, therefore, permit many fewer vehicles to serve the same number of total trips and enable more efficient pooling, further reducing fleet size, numbers of chargers, energy use, cost, and GHG emissions.
Users of a fleet of SAEVs make no major sacrifices in their mobility needs. All trips are still served, though the particular attributes of trips might change. For example, on-demand mobility entails a wait time, pooled rides could take longer than driving oneself; but these increases could be offset by eliminating the time spent searching for and accessing parking. The net effect of these differences could be positive or negative depending on the individual circumstances of each trip, but every trip is still completed in the same hour of each of day.
While rural regions require longer-range battery capacities than urban regions, we find that only 4% of the total fleet must have a range of 225 miles to satisfy all trips. By comparison, urban regions are able to serve all trips with a mixture of 75- and 150-mi. range vehicles, and virtually no longer-range vehicles. It is possible that the actual fleet would need some longer-range vehicles to serve demand during emergency situations and holiday travel, but in most cases these peaks in demand likely could be satisfied by shuffling vehicles between regions.
While our results are specific to the U.S., they may be generalizable to other regions with comparable levels of private vehicle utilization, average driving distances, and grids that are transitioning to renewable electricity, such as Western Europe and, increasingly, China. More densely populated regions will likely see higher levels of vehicle occupancy and lower deadheading ratios, resulting in even greater impacts. We believe that this study provides strong evidence of the value of SAEV fleets to decarbonize personal transportation without compromising individual mobility needs. While a fully private EV fleet with uncontrolled charging would reduce national GHG emissions by 47% compared to a gasoline-fueled baseline, an additional 3 percentage points could be reduced by employing controlled charging and a further additional 20 percentage points could be reduced by serving all mobility with SAEVs for an overall 70% reduction from the gasoline baseline. If the cost of fossil generation were higher or additional low-GHG electricity sources (e.g., from renewables) became available, this reduction could approach 100%.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.0c06655.

  • Model specification, NHTS data, correction factors using RISE, EVI-Pro, additional metrics, geographic analysis, sensitivity runs, and references (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!
  • Corresponding Author
  • Authors
    • Alan T. Jenn - University of California, DavisCalifornia 95616, United States;  Lawrence Berkeley National Laboratory, Berkeley, California 94720-8099, United States;  Orcidhttp://orcid.org/0000-0003-4232-0697
    • Jeffery B. Greenblatt - Lawrence Berkeley National Laboratory, Berkeley, California 94720-8099, United States;  Emerging Futures, Inc., Portland, Oregon 97201, United States
    • Gordon S. Bauer - The International Council on Clean Transportation, https://theicct.org/;  Orcidhttp://orcid.org/0000-0002-1756-5919
    • Brian F. Gerke - Lawrence Berkeley National Laboratory, Berkeley, California 94720-8099, United States
  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

Modeled output of private EV charging from Eric Wood, Matteo Muratori (National Renewable Energy Laboratory). Private EV sampling tool by Jerome Carman, Peter Alstone (Humboldt State University). StreetLight Data provided GPS input data for the RISE model. This article and the work described were sponsored by the U.S. Department of Energy (DOE) Vehicle Technologies Office (VTO) under the Vehicle Technologies Analysis Program. The following DOE Office of Energy Efficiency and Renewable Energy (EERE) managers played important roles in establishing the project concept, advancing implementation, and providing ongoing guidance: Rachael Nealer, Jake Ward, Katherine McMahon, Kelly Fleming, and Heather Croteau. The authors also acknowledge Tom Wenzel of Lawrence Berkeley National Laboratory for critical feedback on this manuscript. This work was funded by the U.S. Department of Energy Vehicle Technologies Office under Lawrence Berkeley National Laboratory Agreement No. 32048. Lawrence Berkeley National Laboratory is supported by the Office of Science of the United States Department of Energy and operated under Contract Grant No. DE-AC02-05CH11231.

References

Click to copy section linkSection link copied!

This article references 55 other publications.

  1. 1
    World Energy Outlook 2016 , 2016. https://www.iea.org/reports/world-energy-outlook-2016.
    Google Scholar
    There is no corresponding record for this reference.
  2. 2
    Hao, H.; Geng, Y.; Sarkis, J. Carbon footprint of global passenger cars: Scenarios through 2050. Energy 2016, 101, 121– 131,  DOI: 10.1016/j.energy.2016.01.089
    Google Scholar
    There is no corresponding record for this reference.
  3. 3
    U.S. Environmental Protection Agency. Green Vehicle Guide: Fast Facts on Transportation Greenhouse Gas Emissions , 2019. https://www.epa.gov/greenvehicles/fast-facts-transportation-greenhouse-gas-emissions.
    Google Scholar
    There is no corresponding record for this reference.
  4. 4
    IEA. Global EV Outlook 2019: Scaling-up the transition to electric mobility , 2019. https://webstore.iea.org/global-ev-outlook-2019.
    Google Scholar
    There is no corresponding record for this reference.
  5. 5
    Bloomberg New Energy Finance. Electric Vehicle Outlook , 2019. https://about.bnef.com/electric-vehicle-outlook/.
    Google Scholar
    There is no corresponding record for this reference.
  6. 6
    Needell, Z.; McNerney, J.; Chang, M.; Trancik, J. Potential for widespread electrification of personal vehicle travel in the United States. Nature Energy 2016, 1, 16112,  DOI: 10.1038/nenergy.2016.112
    Google Scholar
    There is no corresponding record for this reference.
  7. 7
    Saxena, S.; MacDonald, J.; Moura, S. Charging ahead on the transition to electric vehicles with standard wall outlets. Appl. Energy 2015, 157, 720– 728,  DOI: 10.1016/j.apenergy.2015.05.005
    Google Scholar
    There is no corresponding record for this reference.
  8. 8
    Hawkins, T.; Gausen, O.; Stromman, A. Environmental impacts of hybrid and electric vehicles, a review. Int. J. Life Cycle Assess. 2012, 17, 997– 1014,  DOI: 10.1007/s11367-012-0440-9
    Google Scholar
    8
    Environmental impacts of hybrid and electric vehicles-a review
    Hawkins, Troy R.; Gausen, Ola Moa; Stromman, Anders Hammer
    International Journal of Life Cycle Assessment (2012), 17 (8), 997-1014CODEN: IJLCFF; ISSN:0948-3349. (Springer)
    Purpose A literature review is undertaken to understand how well existing studies of the environmental impacts of hybrid and elec. vehicles (EV) address the full life cycle of these technologies. Results of studies are synthesized to compare the global warming potential (GWP) of different EV and internal combustion engine vehicle (ICEV) options. Other impacts are compared; however, data availability limits the extent to which this could be accomplished. Method We define what should be included in a complete, state-of-the-art environmental assessment of hybrid and elec. vehicles considering components and life cycle stages, emission categories, impact categories, and resource use and compare the content of 51 environmental assessments of hybrid and elec. vehicles to our definition. Impact assessment results assocd. with full life cycle inventories (LCI) are compared for GWP as well as emissions of other pollutants. GWP results by life cycle stage and key parameters are extd. and used to perform a meta-anal. quantifying the impacts of vehicle options. Results Few studies provide a full LCI for EVs together with assessment of multiple impacts. Research has focused on well to wheel studies comparing fossil fuel and electricity use as the use phase has been seen to dominate the life cycle of vehicles. Only very recently have studies begun to better address prodn. impacts. Apart from batteries, very few studies provide transparent LCIs of other key EV drive train components. Ests. of EV energy use in the literature span a wide range, 0.10-0.24 kWh/km. Similarly, battery and vehicle lifetime plays an important role in results, yet lifetime assumptions range between 150,000-300,000 km. CO2 and GWP are the most frequently reported results. Compiled results suggest the GWP of EVs powered by coal electricity falls between small and large conventional vehicles while EVs powered by natural gas or low-carbon energy sources perform better than the most efficient ICEVs. EV results in regions dependant on coal electricity demonstrated a trend toward increased SO x emissions compared to fuel use by ICEVs. Conclusions Moving forward research should focus on providing consensus around a transparent inventory for prodn. of elec. vehicles, appropriate electricity grid mix assumptions, the implications of EV adoption on the existing grid, and means of comparing vehicle on the basis of common driving and charging patterns. Although EVs appear to demonstrate decreases in GWP compared to conventional ICEVs, high efficiency ICEVs and grid-independent hybrid elec. vehicles perform better than EVs using coal-fired electricity.
  9. 9
    Cai, H.; Xu, M. Greenhouse Gas Implications of Fleet Electrification Based on Big Data-Informed Individual Travel Patterns. Environ. Sci. Technol. 2013, 47, 9035– 9043,  DOI: 10.1021/es401008f
    Google Scholar
    9
    Greenhouse Gas Implications of Fleet Electrification Based on Big Data-Informed Individual Travel Patterns
    Cai, Hua; Xu, Ming
    Environmental Science & Technology (2013), 47 (16), 9035-9043CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
    Environmental implications of fleet electrification highly depend on the adoption and utilization of elec. vehicles at the individual level. Past research has been constrained by using aggregated data to assume all vehicles with the same travel pattern as the aggregated av. This neglects the inherent heterogeneity of individual travel behaviors and may lead to unrealistic estn. of environmental impacts of fleet electrification. Using "big data" mining techniques, this research examines real-time vehicle trajectory data for 10,375 taxis in Beijing in one week to characterize the travel patterns of individual taxis. We then evaluate the impact of adopting plug-in hybrid elec. vehicles (PHEV) in the taxi fleet on life cycle greenhouse gas emissions based on the characterized individual travel patterns. The results indicate that 1) the largest gasoline displacement (1.1 million gal per yr) can be achieved by adopting PHEVs with modest elec. range (approx. 80 mi) with current battery cost, limited public charging infrastructure, and no government subsidy; 2) reducing battery cost has the largest impact on increasing the electrification rate of vehicle mileage traveled (VMT), thus increasing gasoline displacement, followed by diversified charging opportunities; 3) government subsidies can be more effective to increase the VMT electrification rate and gasoline displacement if targeted to PHEVs with modest elec. ranges (80 to 120 mi); and 4) while taxi fleet electrification can increase greenhouse gas emissions by up to 115 kiloton CO2-eq per yr with the current grid in Beijing, emission redn. of up to 36.5 kiloton CO2-eq per yr can be achieved if the fuel cycle emission factor of electricity can be reduced to 168.7 g/km. Although the results are based on a specific public fleet, this study demonstrates the benefit of using large-scale individual-based trajectory data (big data) to better understand environmental implications of fleet electrification and inform better decision making.
  10. 10
    Green, E.; Skerlos, S.; Winebrake, J. Increasing electric vehicle policy efficiency and effectiveness by reducing mainstream market bias. Energy Policy 2014, 65, 562– 566,  DOI: 10.1016/j.enpol.2013.10.024
    Google Scholar
    There is no corresponding record for this reference.
  11. 11
    King, C.; Griggs, W.; Wirth, F.; Quinn, K.; Shorten, R. Alleviating a form of electric vehicle range anxiety through on-demand vehicle access. Int. J. Control 2015, 88, 717– 728,  DOI: 10.1080/00207179.2014.971521
    Google Scholar
    There is no corresponding record for this reference.
  12. 12
    Barbour, N.; Menon, N.; Zhang, Y.; Mannering, F. Shared automated vehicles: A statistical analysis of consumer use likelihoods and concerns. Transport Policy 2019, 80, 86– 93,  DOI: 10.1016/j.tranpol.2019.05.013
    Google Scholar
    There is no corresponding record for this reference.
  13. 13
    Merfeld, K.; Wilhelms, M.-P.; Henkel, S.; Kreutzer, K. Carsharing with shared autonomous vehicles: Uncovering drivers, barriers and future developments–A four-stage Delphi study. Technological Forecasting and Social Change 2019, 144, 66– 81,  DOI: 10.1016/j.techfore.2019.03.012
    Google Scholar
    There is no corresponding record for this reference.
  14. 14
    Greenblatt, J.; Saxena, S. Autonomous taxis could greatly reduce greenhouse-gas emissions of US light-duty vehicles. Nature Climate Change 2015, 5 (9), 860– 865,  DOI: 10.1038/nclimate2685
    Google Scholar
    There is no corresponding record for this reference.
  15. 15
    Associated Press. Waymo: Phoenix begins Testing Self-Driving Vans on Public Roads. CBS News , 2018. https://www.cbsnews.com/news/waymo-phoenix-tests-self-driving-vans-on-public-roads/.
    Google Scholar
    There is no corresponding record for this reference.
  16. 16
    Litman, T. Autonomous Vehicle Implementation Predictions; Victoria Transport Policy Institute, 2017.
    Google Scholar
    There is no corresponding record for this reference.
  17. 17
    Fulton, L. M. Three Revolutions in Urban Passenger Travel. Joule 2018, 2, 575– 578,  DOI: 10.1016/j.joule.2018.03.005
    Google Scholar
    There is no corresponding record for this reference.
  18. 18
    Zhang, W.; Guhathakurta, S.; Khalil, E. B. The impact of private autonomous vehicles on vehicle ownership and unoccupied VMT generation. Transportation Research Part C: Emerging Technologies 2018, 90, 156– 165,  DOI: 10.1016/j.trc.2018.03.005
    Google Scholar
    There is no corresponding record for this reference.
  19. 19
    Henao, A.; Marshall, W. E. The impact of ride-hailing on vehicle miles traveled. Transportation 2019, 46, 2173– 2194,  DOI: 10.1007/s11116-018-9923-2
    Google Scholar
    There is no corresponding record for this reference.
  20. 20
    Sperling, D. Three Revolution: Steering Automated, Shared, and Electric Vehicles to a Better Future; Island Press, 2018.
    Google Scholar
    There is no corresponding record for this reference.
  21. 21
    Chen, T. D.; Kockelman, K. M.; Hanna, J. P. Operations of a shared, autonomous, electric vehicle fleet: Implications of vehicle & charging infrastructure decisions. Transportation Research Part A: Policy and Practice 2016, 94, 243– 254,  DOI: 10.1016/j.tra.2016.08.020
    Google Scholar
    There is no corresponding record for this reference.
  22. 22
    Luk, J. M.; Kim, H. C.; De Kleine, R.; Wallington, T. J.; MacLean, H. L. Review of the Fuel Saving, Life Cycle GHG Emission, and Ownership Cost Impacts of Lightweighting Vehicles with Different Powertrains. Environ. Sci. Technol. 2017, 51, 8215– 8228,  DOI: 10.1021/acs.est.7b00909
    Google Scholar
    22
    Review of the Fuel Saving, Life Cycle GHG Emission, and Ownership Cost Impacts of Lightweighting Vehicles with Different Powertrains
    Luk, Jason M.; Kim, Hyung Chul; De Kleine, Robert; Wallington, Timothy J.; MacLean, Heather L.
    Environmental Science & Technology (2017), 51 (15), 8215-8228CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
    A review concerning fuel conservation, life cycle greenhouse gas (GHG) emissions from, and ownership costs assocd. with light-weighting vehicles with alternative power-trains from financial and environmental perspectives, is given. Topics discussed include: introduction; lightwt. materials background; literature overview; fuel conservation from light-weighting vehicles with different power-trains; power-train re-sizing, drive cycle, and vehicle class effects on fuel conservation from light-weighting vehicles with different power-trains; life cycle literature ests. of fuel conservation from light-weighting vehicles with different power-trains; life cycle greenhouse gas (GHG) emissions; fuel cycle, fuel and vehicle cycle, vehicle cycle, and life cycle GHG emissions impact from light-weighting vehicles with different power-trains; life cycle ownership costs, ownership costs, and secondary mass redns. from light-weighting vehicles with different power-trains; vehicle prodn. costs from light-weighting vehicles with different power-trains; discussion; and supporting information (alternative power-train background, literature overview).
  23. 23
    Bauer, G. S.; Greenblatt, J. B.; Gerke, B. F. Cost, Energy, and Environmental Impact of Automated Electric Taxi Fleets in Manhattan. Environ. Sci. Technol. 2018, 52, 4920– 4928,  DOI: 10.1021/acs.est.7b04732
    Google Scholar
    23
    Cost, Energy, and Environmental Impact of Automated Electric Taxi Fleets in Manhattan
    Bauer, Gordon S.; Greenblatt, Jeffery B.; Gerke, Brian F.
    Environmental Science & Technology (2018), 52 (8), 4920-4928CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
    Shared automated elec. vehicles (SAEVs) hold great promise for improving transportation access in urban centers while drastically reducing transportation-related energy consumption and air pollution. Using taxi-trip data from New York City, we develop an agent-based model to predict the battery range and charging infrastructure requirements of a fleet of SAEVs operating on Manhattan Island. We also develop a model to est. the cost and environmental impact of providing service and perform extensive sensitivity anal. to test the robustness of our predictions. We est. that costs will be lowest with a battery range of 50-90 mi, with either 66 chargers per square mile, rated at 11 kW or 44 chargers per square mile, rated at 22 kW. We est. that the cost of service provided by such an SAEV fleet will be $0.29-$0.61 per revenue mile, an order of magnitude lower than the cost of service of present-day Manhattan taxis and $0.05-$0.08/mi lower than that of an automated fleet composed of any currently available hybrid or internal combustion engine vehicle (ICEV). We est. that such an SAEV fleet drawing power from the current NYC power grid would reduce GHG emissions by 73% and energy consumption by 58% compared to an automated fleet of ICEVs.
  24. 24
    Franke, T.; Krems, J. F. Understanding charging behaviour of electric vehicle users. Transportation Research Part F: Traffic Psychology and Behaviour 2013, 21, 75– 89,  DOI: 10.1016/j.trf.2013.09.002
    Google Scholar
    There is no corresponding record for this reference.
  25. 25
    Franke, T.; Gunther, M.; Trantow, M.; Krems, J. F. Does this range suit me? Range satisfaction of battery electric vehicle users. Applied Ergonomics 2017, 65, 191– 199,  DOI: 10.1016/j.apergo.2017.06.013
    Google Scholar
    25
    Does this range suit me? Range satisfaction of battery electric vehicle users
    Franke Thomas; Gunther Madlen; Trantow Maria; Krems Josef F
    Applied ergonomics (2017), 65 (), 191-199 ISSN:.
    User satisfaction is a vital design criterion for sustainable systems. The present research aimed to understand factors relating to individually perceived range satisfaction of battery electric vehicle (BEV) users. Data from a large-scale BEV field trial (N = 72) were analyzed. Apart from an initial drop in range satisfaction, increasing practical experience was related to increased range satisfaction. Classical indicators of users' mobility profiles (daily travel distances) were only weakly related to lower range satisfaction (not significant), after controlling for practical experience and preferred coverage of mobility needs. The regularity/predictability of users' mobility patterns, the percentage of journeys not coverable because of range issues, and users' individual comfortable range accounted for variance in range satisfaction. Finally, range satisfaction was related to key indicators of general BEV acceptance (e.g., purchase intentions). These results underline the complex dynamics involved in individual range satisfaction, as well as its central role for BEV acceptance.
  26. 26
    Wolbertus, R.; Kroesen, M.; van den Hoed, R.; Chorus, C. G. Policy effects on charging behaviour of electric vehicle owners and on purchase intentions of prospective owners: Natural and stated choice experiments. Transportation Research Part D: Transport and Environment 2018, 62, 283– 297,  DOI: 10.1016/j.trd.2018.03.012
    Google Scholar
    There is no corresponding record for this reference.
  27. 27
    Melaina, M.; Muratori, M.; McLaren, J.; Schwabe, P. Investing in Alternative Fuel Infrastructure: Insights for California from Stakeholder Interviews: Preprint. National Renewable Energy Laboratory Publications , 2017, NREL/CP-5400-67617. https://www.nrel.gov/docs/fy17osti/67617.pdf
    Google Scholar
    There is no corresponding record for this reference.
  28. 28
    Nelson, J. H.; Wisland, L. M. Achieving 50% Renewable Electricity in California: The Role of Non-Fossil Flexibility in a Cleaner Electricity Grid; Union of Concerned Scientists, 2015.
    Google Scholar
    There is no corresponding record for this reference.
  29. 29
    Alstone, P.; Potter, J.; Piette, M.; Schwartz, P.; Berger, M.; Dunn, L.; Smith, S.; Sohn, M.; Aghajanzadeh, A.; Stensson, S.; Szinai, J.; Walter, T.; McKenzie, L.; Lavin, L.; Schneiderman, B.; Mileva, A.; Cutter, E.; Olson, A.; Bode, J.; Ciccone, A.; Jain, A. Final Report on Phase 2 Results, 2015 California Demand Response Potential Study: Charting California’s Demand Response Future; Lawrence Berkeley National Laboratory, 2016.
    Google Scholar
    There is no corresponding record for this reference.
  30. 30
    Richardson, D. B. Electric vehicles and the electric grid: A review of modeling approaches, Impacts, and renewable energy integration. Renewable Sustainable Energy Rev. 2013, 19, 247– 254,  DOI: 10.1016/j.rser.2012.11.042
    Google Scholar
    There is no corresponding record for this reference.
  31. 31
    Foley, A.; Tyther, B.; Calnan, P.; O Gallachoir, B. Impacts of Electric Vehicle charging under electricity market operations. Appl. Energy 2013, 101, 93– 102,  DOI: 10.1016/j.apenergy.2012.06.052
    Google Scholar
    There is no corresponding record for this reference.
  32. 32
    Sheppard, C.; Szinai, J.; Abhyankar, N.; Gopal, A. Grid Impacts of Electric Vehicles and Managed Charging in California: Linking Agent-Based Electric Vehicle Charging with Power System Dispatch Models; Sustainable Energy Systems Group, Sustainable Energy Department, Energy Analysis and Environmental Impacts Division, International Energy Studies Group, International Energy Analysis Department, 2019.
    Google Scholar
    There is no corresponding record for this reference.
  33. 33
    Szinai, J. K.; Sheppard, C. J. R.; Abhyankar, N.; Gopal, A. R. Reduced grid operating costs and renewable energy curtailment with electric vehicle charge management. Energy Policy 2020, 136, 111051,  DOI: 10.1016/j.enpol.2019.111051
    Google Scholar
    There is no corresponding record for this reference.
  34. 34
    Lopes, J.; Soares, F.; Almeida, P. Integration of electric vehicles in the electric power system. Proc. IEEE 2011, 99, 168– 183,  DOI: 10.1109/JPROC.2010.2066250
    Google Scholar
    There is no corresponding record for this reference.
  35. 35
    Tarroja, B. Assessing the stationary energy storage equivalency of vehicle-to-grid charging battery electric vehicles. Energy 2016, 106, 673– 690,  DOI: 10.1016/j.energy.2016.03.094
    Google Scholar
    There is no corresponding record for this reference.
  36. 36
    Forrest, K. Charging a renewable future: The impact of electric vehicle charging intelligence on energy storage requirements to meet renewable portfolio standards. J. Power Sources 2016, 336, 63– 74,  DOI: 10.1016/j.jpowsour.2016.10.048
    Google Scholar
    36
    Charging a renewable future: The impact of electric vehicle charging intelligence on energy storage requirements to meet renewable portfolio standards
    Forrest, Kate E.; Tarroja, Brian; Zhang, Li; Shaffer, Brendan; Samuelsen, Scott
    Journal of Power Sources (2016), 336 (), 63-74CODEN: JPSODZ; ISSN:0378-7753. (Elsevier B.V.)
    Increased usage of renewable energy resources is key for energy system evolution to address environmental concerns. Capturing variable renewable power requires the use of energy storage to shift generation and load demand. The integration of plug-in elec. vehicles, however, impacts the load demand profile and therefore the capacity of energy storage required to meet renewable utilization targets. This study examines how the intelligence of plug-in elec. vehicle (PEV) integration impacts the required capacity of energy storage systems to meet renewable utilization targets for a large-scale energy system, using California as an example for meeting a 50% and 80% renewable portfolio std. (RPS) in 2030 and 2050. For an 80% RPS in 2050, immediate charging of PEVs requires the installation of an aggregate energy storage system with a power capacity of 60% of the installed renewable capacity and an energy capacity of 2.3% of annual renewable generation. With smart charging of PEVs, required power capacity drops to 16% and required energy capacity drops to 0.6%, and with vehicle-to-grid (V2G) charging, non-vehicle energy storage systems are no longer required. Overall, this study highlights the importance of intelligent PEV charging for minimizing the scale of infrastructure required to meet renewable utilization targets.
  37. 37
    Coignard, J.; Saxena, S.; Greenblatt, J.; Wang, D. Clean Vehicles as an Enabler for a Clean Electricity Grid. Environ. Res. Lett. 2018, 13, 054031,  DOI: 10.1088/1748-9326/aabe97
    Google Scholar
    There is no corresponding record for this reference.
  38. 38
    Iacobucci, R.; McLellan, B.; Tezuka, T. Modeling shared autonomous electric vehicles: Potential for transport and power grid integration. Energy 2018, 158, 148– 163,  DOI: 10.1016/j.energy.2018.06.024
    Google Scholar
    There is no corresponding record for this reference.
  39. 39
    Jones, E. C.; Leibowicz, B. D. Contributions of shared autonomous vehicles to climate change mitigation. Transportation Research Part D: Transport and Environment 2019, 72, 279– 298,  DOI: 10.1016/j.trd.2019.05.005
    Google Scholar
    There is no corresponding record for this reference.
  40. 40
    Li, M.; Lenzen, M.; Wang, D.; Nansai, K. GIS-based modelling of electric-vehicle-grid integration in a 100% renewable electricity grid. Appl. Energy 2020, 262, 114577,  DOI: 10.1016/j.apenergy.2020.114577
    Google Scholar
    There is no corresponding record for this reference.
  41. 41
    Hedegaard, K.; Ravn, H.; Juul, N.; Meibom, P. Effects of electric vehicles on power systems in Northern Europe. Energy 2012, 48, 356– 368,  DOI: 10.1016/j.energy.2012.06.012
    Google Scholar
    There is no corresponding record for this reference.
  42. 42
    Haque, A. N. M. M.; Ibn Saif, A. U. N.; Nguyen, P. H.; Torbaghan, S. S. Exploration of dispatch model integrating wind generators and electric vehicles. Appl. Energy 2016, 183, 1441– 1451,  DOI: 10.1016/j.apenergy.2016.09.078
    Google Scholar
    There is no corresponding record for this reference.
  43. 43
    Hanemann, P.; Behnert, M.; Bruckner, T. Effects of electric vehicle charging strategies on the German power system. Appl. Energy 2017, 203, 608– 622,  DOI: 10.1016/j.apenergy.2017.06.039
    Google Scholar
    There is no corresponding record for this reference.
  44. 44
    Taljegard, M.; Göransson, L.; Odenberger, M.; Johnsson, F. Impacts of electric vehicles on the electricity generation portfolio - A Scandinavian-German case study. Appl. Energy 2019, 235, 1637– 1650,  DOI: 10.1016/j.apenergy.2018.10.133
    Google Scholar
    There is no corresponding record for this reference.
  45. 45
    Limmer, S.; Rodemann, T. Peak load reduction through dynamic pricing for electric vehicle charging. International Journal of Electrical Power & Energy Systems 2019, 113, 117– 128,  DOI: 10.1016/j.ijepes.2019.05.031
    Google Scholar
    There is no corresponding record for this reference.
  46. 46
    Sheppard, C.; Jenn, A.; Bauer, G.; Gerke, B.; Greenblatt, J.; Gopal, A. A joint optimization scheme for the planning and operations of a regional electrified fleets of ride hailing vehicles serving mobility on demand. Transportation Research Record: Journal of the Transportation Research Board 2019, 2673 (6), 579– 597,  DOI: 10.1177/0361198119838270
    Google Scholar
    There is no corresponding record for this reference.
  47. 47
    Jenn, A.; Clark-Sutton, K.; Gallaher, M. P.; Petrusa, J. Environmental impacts of extreme fast charging. Environ. Res. Lett. 2020, 15, 094060,  DOI: 10.1088/1748-9326/ab9870
    Google Scholar
    47
    Environmental impacts of extreme fast charging
    Jenn, Alan; Clark-Sutton, Kyle; Gallaher, Michael; Petrusa, Jeffrey
    Environmental Research Letters (2020), 15 (9), 094060CODEN: ERLNAL; ISSN:1748-9326. (IOP Publishing Ltd.)
    As elec. vehicles and their assocd. charging infrastructure continue to evolve, there is potential to simultaneous alleviate range and recharge concerns with the development of extreme fast chargers (XFC) that can fully charge batteries in PEVs in the span of a few minutes. Recent announcements from EVSE providers and vehicle manufacturers suggest that XFC charging stations, which can recharge a BEV at roughly 20 to 25 mi per min of charging, and XFC-capable BEVs, could be com. available within the next 5 years. Our study investigates the potential emission impacts of widespread use of extreme fast charging (350 kW) for elec. vehicles in 2030. We conduct a novel vehicle charging simulation model by combining empirical charging behavior data across several data sources. These charging demands are then added as exogenous load to the Grid Optimized Operation Dispatch (GOOD) model, which simulates the operation of generators across the Untied States. We fiend that XFC can increase both greenhouse gas emissions and local air pollutants, though the results are sensitive to local contexts and grid compn.
  48. 48
    Hand, M. M.; Baldwin, S.; DeMeo, E.; Reilly, J. M.; Mai, T.; Arent, D.; Porro, G.; Meshek, M.; Sandor, D. Renewable Electricity Futures Study; National Renewable Energy Laboratory, 2012.
    Google Scholar
    There is no corresponding record for this reference.
  49. 49
    Wood, E. W.; Rames, C. L.; Muratori, M.; Srinivasa Raghavan, S.; Young, S. E. Charging Electric Vehicles in Smart Cities: An EVI-Pro Analysis of Columbus; NREL: OH, 2018.
    Google Scholar
    There is no corresponding record for this reference.
  50. 50
    Sheppard, C.; Jenn, A. Grid-integrated Electric Mobility (GEM) Model. GitHub , 2021. https://github.com/LBNL-UCB-STI/gem.
    Google Scholar
    There is no corresponding record for this reference.
  51. 51
    Sheppard, C.; Jenn, A.; Bauer, G.; Greenblatt, J.; Gerke, B. Grid-integrated Electric Mobility (GEM) Model – Manuscript 2021 Branch. GitHub , 2021. https://github.com/LBNL-UCB-STI/gem/tree/manuscript-2021.
    Google Scholar
    There is no corresponding record for this reference.
  52. 52
    U.S. Department of Transportation, Federal Highway Administration 2017 National Household Travel Survey , 2017. https://nhts.ornl.gov.
    Google Scholar
    There is no corresponding record for this reference.
  53. 53
    Xu, Z.; Su, W.; Hu, Z.; Song, Y.; Zhang, H. A Hierarchical Framework for Coordinated Charging of Plug-In Electric Vehicles in China. IEEE Trans. Smart Grid 2016, 7, 428– 438,  DOI: 10.1109/TSG.2014.2387436
    Google Scholar
    There is no corresponding record for this reference.
  54. 54
    U.S. Energy Information Administration. Annual Energy Outlook , 2019. https://www.eia.gov/outlooks/aeo/.
    Google Scholar
    There is no corresponding record for this reference.
  55. 55
    U.S. Energy Information Administration Preliminary Monthly Electric Generator Inventory (based on Form EIA-860M as a supplement to Form EIA-860). https://www.eia.gov/electricity/data/eia860M/ (accessed September 2020).
    Google Scholar
    There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!
Citation Statements
Explore this article's citation statements on scite.ai
powered by  Scite

This article is cited by 24 publications.

  1. Matthew Bruchon, Zihao Lance Chen, Jeremy Michalek. Cleaning up while Changing Gears: The Role of Battery Design, Fossil Fuel Power Plants, and Vehicle Policy for Reducing Emissions in the Transition to Electric Vehicles. Environmental Science & Technology 2024, 58 (8) , 3787-3799. https://doi.org/10.1021/acs.est.3c07098
  2. Yuliang Zhang, Yifan Chen, Yichen Wang, Chengcheng Yu, Xiqun (Michael) Chen, Der-Horng Lee. Understanding spatial differences in CO2 emissions between passenger and freight transportation. Transportation Research Part D: Transport and Environment 2025, 145 , 104800. https://doi.org/10.1016/j.trd.2025.104800
  3. Dimitrios Rimpas, Dimitrios E. Barkas, Vasilios A. Orfanos, Ioannis Christakis. Decarbonizing the Transportation Sector: A Review on the Role of Electric Vehicles Towards the European Green Deal for the New Emission Standards. Air 2025, 3 (2) , 10. https://doi.org/10.3390/air3020010
  4. Soongil Kwon, Yoon-Seong Chang. A Study on CO2 Emission Reduction Using Operating Internal Combustion Engine Vehicles (ICEVs) and Electric Vehicles (EVs) for Rental Vehicles, Focusing on South Korea. Energies 2025, 18 (11) , 2997. https://doi.org/10.3390/en18112997
  5. Mohd Aiman Khan, Wilco Burghout, Oded Cats, Erik Jenelius, Matej Cebecauer. Charge-on-the-move solutions for future mobility: A review of current and future prospects. Transportation Research Interdisciplinary Perspectives 2025, 29 , 101323. https://doi.org/10.1016/j.trip.2025.101323
  6. Madalsa Singh, Tugce Yuksel, Jeremy J. Michalek, Inês M. L. Azevedo. Ensuring greenhouse gas reductions from electric vehicles compared to hybrid gasoline vehicles requires a cleaner U.S. electricity grid. Scientific Reports 2024, 14 (1) https://doi.org/10.1038/s41598-024-51697-1
  7. Matthew D. Dean, Krishna Murthy Gurumurthy, Zhi Zhou, Omer Verbas, Taner Cokyasar, Kara Kockelman. An integrated transportation-power system model for a decarbonizing world. Transportation Planning and Technology 2024, 47 (8) , 1277-1313. https://doi.org/10.1080/03081060.2024.2348721
  8. Xu Hao, Deyu Zhou, Ruiheng Zhong, Shunxi Li, Xianming Meng, Bo Liu. Electrification pathways for light-duty logistics vehicles based on perceived cost of ownership in Northern China. Carbon Footprints 2024, 3 (3) https://doi.org/10.20517/cf.2024.24
  9. Lichen Luo, Joana Portugal-Pereira, Kiyoshi Takami, Giancarlos Parady. Unintended environmental impacts of private automated vehicles: Insights from Gunma Prefecture, Japan. Transportation Research Part D: Transport and Environment 2024, 134 , 104298. https://doi.org/10.1016/j.trd.2024.104298
  10. Ada Garus, Andromachi Mourtzouchou, Jaime Suarez, Georgios Fontaras, Biagio Ciuffo. Exploring Sustainable Urban Transportation: Insights from Shared Mobility Services and Their Environmental Impact. Smart Cities 2024, 7 (3) , 1199-1220. https://doi.org/10.3390/smartcities7030051
  11. Shaopeng Zhong, Ao Liu, Yu Jiang, Simon Hu, Feng Xiao, Hai-Jun Huang, Yan Song. Energy and environmental impacts of shared autonomous vehicles under different pricing strategies. npj Urban Sustainability 2023, 3 (1) https://doi.org/10.1038/s42949-023-00092-2
  12. Yujie Sheng, Yanxi Lin, Hongtai Zeng, Yang Yu, Qinglai Guo, Shiwei Xie. Emission-concerned coordinated dispatching of electrified autonomous mobility-on-demand system and power system incorporating heterogeneous spatiotemporal scales. Sustainable Cities and Society 2023, 98 , 104755. https://doi.org/10.1016/j.scs.2023.104755
  13. Chandan Bhardwaj, Jonn Axsen, Curran Crawford. Simulating long-term emissions from private automated vehicles under climate policies. Transportation Research Part D: Transport and Environment 2023, 118 , 103665. https://doi.org/10.1016/j.trd.2023.103665
  14. Wanshi Hong, Alan Jenn, Bin Wang. Electrified autonomous freight benefit analysis on fleet, infrastructure and grid leveraging Grid-Electrified Mobility (GEM) model. Applied Energy 2023, 335 , 120760. https://doi.org/10.1016/j.apenergy.2023.120760
  15. Ona Van den bergh, Simon Weekx, Cedric De Cauwer, Lieselot Vanhaverbeke. Locating Charging Infrastructure for Shared Autonomous Electric Vehicles and for Vehicle-to-Grid Strategy: A Systematic Review and Research Agenda from an Energy and Mobility Perspective. World Electric Vehicle Journal 2023, 14 (3) , 56. https://doi.org/10.3390/wevj14030056
  16. Vishnu Vijayakumar, Alan Jenn, Joan Ogden. Modeling future hydrogen supply chains in the western United States under uncertainties: an optimization-based approach focusing on California as a hydrogen hub. Sustainable Energy & Fuels 2023, 7 (5) , 1223-1244. https://doi.org/10.1039/D3SE00043E
  17. Daniel Rasbash, Kevin Joseph Dillman, Jukka Heinonen, Eyjólfur Ingi Ásgeirsson. A National and Regional Greenhouse Gas Breakeven Assessment of EVs across North America. Sustainability 2023, 15 (3) , 2181. https://doi.org/10.3390/su15032181
  18. Steven J. Davis, Rebecca S. Dodder, David D. Turner, Ines M. L. Azevedo, Morgan Bazilian, John Bistline, Sanya Carley, Christopher T. M. Clack, Joseph E. Fargione, Emily Grubert, Jason Hill, Adrienne L. Hollis, Alan Jenn, Ryan A. Jones, Eric Masanet, Erin N. Mayfield, Matteo Muratori, Wei Peng, Brittany C. Sellers, , , , , , . Chapter 32 : Mitigation. Fifth National Climate Assessment. 2023https://doi.org/10.7930/NCA5.2023.CH32
  19. Yanning Li, Xinwei Li, Alan Jenn. Evaluating the emission benefits of shared autonomous electric vehicle fleets: A case study in California. Applied Energy 2022, 323 , 119638. https://doi.org/10.1016/j.apenergy.2022.119638
  20. Matthew D. Dean, Kara M. Kockelman. Are Electric Vehicle Targets Enough? The Decarbonization Benefits of Managed Charging and Second-Life Battery Uses. Transportation Research Record: Journal of the Transportation Research Board 2022, 2676 (8) , 24-43. https://doi.org/10.1177/03611981221082572
  21. Matthew Moniot, Brennan Borlaug, Yanbo Ge, Eric Wood, Jason Zimbler. Electrifying New York City Ride-Hailing fleets: An examination of the need for public fast charging. iScience 2022, 25 (4) , 104171. https://doi.org/10.1016/j.isci.2022.104171
  22. Riccardo Curtale, Feixiong Liao, Ella Rebalski. Transitional behavioral intention to use autonomous electric car-sharing services: Evidence from four European countries. Transportation Research Part C: Emerging Technologies 2022, 135 , 103516. https://doi.org/10.1016/j.trc.2021.103516
  23. Felix Creutzig, Leila Niamir, Xuemei Bai, Max Callaghan, Jonathan Cullen, Julio Díaz-José, Maria Figueroa, Arnulf Grubler, William F. Lamb, Adrian Leip, Eric Masanet, Érika Mata, Linus Mattauch, Jan C. Minx, Sebastian Mirasgedis, Yacob Mulugetta, Sudarmanto Budi Nugroho, Minal Pathak, Patricia Perkins, Joyashree Roy, Stephane de la Rue du Can, Yamina Saheb, Shreya Some, Linda Steg, Julia Steinberger, Diana Ürge-Vorsatz. Demand-side solutions to climate change mitigation consistent with high levels of well-being. Nature Climate Change 2022, 12 (1) , 36-46. https://doi.org/10.1038/s41558-021-01219-y
  24. Marco Raugei, Mashael Kamran, Allan Hutchinson. Environmental implications of the ongoing electrification of the UK light duty vehicle fleet. Resources, Conservation and Recycling 2021, 174 , 105818. https://doi.org/10.1016/j.resconrec.2021.105818
Download PDF
Go to Environmental Science & Technology
Get e-Alerts
Get e-Alerts

Environmental Science & Technology

Cite this: Environ. Sci. Technol. 2021, 55, 5, 3229–3239
Click to copy citationCitation copied!
https://doi.org/10.1021/acs.est.0c06655
Published February 10, 2021

Copyright © 2021 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Article Views

4769

Altmetric

-

Citations

24
Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 1

    Figure 1. Sources of data (blue), data processing (dark red), models (light red), intermediate data (gray), and model outputs (yellow) in the overall modeling and processing workflow.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 2

    Figure 2. Reference scenario for S = 50% and C = 0% (half of all trips are satisfied by SAEVs, and half by private EVs with uncontrolled charging), showing (a) charging load by power level, (b) EV state (charging, moving, or idle) by battery range for SAEVs, and (c) EV state (charging, moving, or idle) for private EVs. The scenario was run over 12 days: three consecutive days—Sunday, Monday, and Tuesday—in each season.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 3

    Figure 3. Three days (1 weekend day, followed by two weekdays) of simulated EV charging load as functions of SAEV trip fraction (S) and private EV controlled charging fraction (C), disaggregated by charger level, for both private EVs and SAEVs. Each panel is labeled by the value for S and C used in the simulation. Panels a–d: S = 0%, C = 0–100%. Panels e–h: S = 25%, C = 0–100%. Panels i–l: S = 50%, C = 0–100%. Panels m–p: S = 75%, C = 0–100%. Panel q: S = 100% (value of C is irrelevant as there are no private EVs).

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 4

    Figure 4. (a) Fleet size, (b) numbers of chargers, (c) peak power demand, (d) total cost of ownership, and (e) consequential GHG emissions vs fraction of SAEV trips (S) with C = 0%.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 5

    Figure 5. (a) Optimal fleet size (in vehicles per trip) and (b) charging infrastructure requirements (in chargers per vehicle) disaggregated by urban and rural regions for C = 0% and with S ranging from 25% to 100%.

    High Resolution Image
    Download MS PowerPoint Slide

  • References

    This article references 55 other publications.

    1. 1
      World Energy Outlook 2016 , 2016. https://www.iea.org/reports/world-energy-outlook-2016.
      There is no corresponding record for this reference.
    2. 2
      Hao, H.; Geng, Y.; Sarkis, J. Carbon footprint of global passenger cars: Scenarios through 2050. Energy 2016, 101, 121– 131,  DOI: 10.1016/j.energy.2016.01.089
      There is no corresponding record for this reference.
    3. 3
      U.S. Environmental Protection Agency. Green Vehicle Guide: Fast Facts on Transportation Greenhouse Gas Emissions , 2019. https://www.epa.gov/greenvehicles/fast-facts-transportation-greenhouse-gas-emissions.
      There is no corresponding record for this reference.
    4. 4
      IEA. Global EV Outlook 2019: Scaling-up the transition to electric mobility , 2019. https://webstore.iea.org/global-ev-outlook-2019.
      There is no corresponding record for this reference.
    5. 5
      Bloomberg New Energy Finance. Electric Vehicle Outlook , 2019. https://about.bnef.com/electric-vehicle-outlook/.
      There is no corresponding record for this reference.
    6. 6
      Needell, Z.; McNerney, J.; Chang, M.; Trancik, J. Potential for widespread electrification of personal vehicle travel in the United States. Nature Energy 2016, 1, 16112,  DOI: 10.1038/nenergy.2016.112
      There is no corresponding record for this reference.
    7. 7
      Saxena, S.; MacDonald, J.; Moura, S. Charging ahead on the transition to electric vehicles with standard wall outlets. Appl. Energy 2015, 157, 720– 728,  DOI: 10.1016/j.apenergy.2015.05.005
      There is no corresponding record for this reference.
    8. 8
      Hawkins, T.; Gausen, O.; Stromman, A. Environmental impacts of hybrid and electric vehicles, a review. Int. J. Life Cycle Assess. 2012, 17, 997– 1014,  DOI: 10.1007/s11367-012-0440-9
      8
      Environmental impacts of hybrid and electric vehicles-a review
      Hawkins, Troy R.; Gausen, Ola Moa; Stromman, Anders Hammer
      International Journal of Life Cycle Assessment (2012), 17 (8), 997-1014CODEN: IJLCFF; ISSN:0948-3349. (Springer)
      Purpose A literature review is undertaken to understand how well existing studies of the environmental impacts of hybrid and elec. vehicles (EV) address the full life cycle of these technologies. Results of studies are synthesized to compare the global warming potential (GWP) of different EV and internal combustion engine vehicle (ICEV) options. Other impacts are compared; however, data availability limits the extent to which this could be accomplished. Method We define what should be included in a complete, state-of-the-art environmental assessment of hybrid and elec. vehicles considering components and life cycle stages, emission categories, impact categories, and resource use and compare the content of 51 environmental assessments of hybrid and elec. vehicles to our definition. Impact assessment results assocd. with full life cycle inventories (LCI) are compared for GWP as well as emissions of other pollutants. GWP results by life cycle stage and key parameters are extd. and used to perform a meta-anal. quantifying the impacts of vehicle options. Results Few studies provide a full LCI for EVs together with assessment of multiple impacts. Research has focused on well to wheel studies comparing fossil fuel and electricity use as the use phase has been seen to dominate the life cycle of vehicles. Only very recently have studies begun to better address prodn. impacts. Apart from batteries, very few studies provide transparent LCIs of other key EV drive train components. Ests. of EV energy use in the literature span a wide range, 0.10-0.24 kWh/km. Similarly, battery and vehicle lifetime plays an important role in results, yet lifetime assumptions range between 150,000-300,000 km. CO2 and GWP are the most frequently reported results. Compiled results suggest the GWP of EVs powered by coal electricity falls between small and large conventional vehicles while EVs powered by natural gas or low-carbon energy sources perform better than the most efficient ICEVs. EV results in regions dependant on coal electricity demonstrated a trend toward increased SO x emissions compared to fuel use by ICEVs. Conclusions Moving forward research should focus on providing consensus around a transparent inventory for prodn. of elec. vehicles, appropriate electricity grid mix assumptions, the implications of EV adoption on the existing grid, and means of comparing vehicle on the basis of common driving and charging patterns. Although EVs appear to demonstrate decreases in GWP compared to conventional ICEVs, high efficiency ICEVs and grid-independent hybrid elec. vehicles perform better than EVs using coal-fired electricity.
    9. 9
      Cai, H.; Xu, M. Greenhouse Gas Implications of Fleet Electrification Based on Big Data-Informed Individual Travel Patterns. Environ. Sci. Technol. 2013, 47, 9035– 9043,  DOI: 10.1021/es401008f
      9
      Greenhouse Gas Implications of Fleet Electrification Based on Big Data-Informed Individual Travel Patterns
      Cai, Hua; Xu, Ming
      Environmental Science & Technology (2013), 47 (16), 9035-9043CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
      Environmental implications of fleet electrification highly depend on the adoption and utilization of elec. vehicles at the individual level. Past research has been constrained by using aggregated data to assume all vehicles with the same travel pattern as the aggregated av. This neglects the inherent heterogeneity of individual travel behaviors and may lead to unrealistic estn. of environmental impacts of fleet electrification. Using "big data" mining techniques, this research examines real-time vehicle trajectory data for 10,375 taxis in Beijing in one week to characterize the travel patterns of individual taxis. We then evaluate the impact of adopting plug-in hybrid elec. vehicles (PHEV) in the taxi fleet on life cycle greenhouse gas emissions based on the characterized individual travel patterns. The results indicate that 1) the largest gasoline displacement (1.1 million gal per yr) can be achieved by adopting PHEVs with modest elec. range (approx. 80 mi) with current battery cost, limited public charging infrastructure, and no government subsidy; 2) reducing battery cost has the largest impact on increasing the electrification rate of vehicle mileage traveled (VMT), thus increasing gasoline displacement, followed by diversified charging opportunities; 3) government subsidies can be more effective to increase the VMT electrification rate and gasoline displacement if targeted to PHEVs with modest elec. ranges (80 to 120 mi); and 4) while taxi fleet electrification can increase greenhouse gas emissions by up to 115 kiloton CO2-eq per yr with the current grid in Beijing, emission redn. of up to 36.5 kiloton CO2-eq per yr can be achieved if the fuel cycle emission factor of electricity can be reduced to 168.7 g/km. Although the results are based on a specific public fleet, this study demonstrates the benefit of using large-scale individual-based trajectory data (big data) to better understand environmental implications of fleet electrification and inform better decision making.
    10. 10
      Green, E.; Skerlos, S.; Winebrake, J. Increasing electric vehicle policy efficiency and effectiveness by reducing mainstream market bias. Energy Policy 2014, 65, 562– 566,  DOI: 10.1016/j.enpol.2013.10.024
      There is no corresponding record for this reference.
    11. 11
      King, C.; Griggs, W.; Wirth, F.; Quinn, K.; Shorten, R. Alleviating a form of electric vehicle range anxiety through on-demand vehicle access. Int. J. Control 2015, 88, 717– 728,  DOI: 10.1080/00207179.2014.971521
      There is no corresponding record for this reference.
    12. 12
      Barbour, N.; Menon, N.; Zhang, Y.; Mannering, F. Shared automated vehicles: A statistical analysis of consumer use likelihoods and concerns. Transport Policy 2019, 80, 86– 93,  DOI: 10.1016/j.tranpol.2019.05.013
      There is no corresponding record for this reference.
    13. 13
      Merfeld, K.; Wilhelms, M.-P.; Henkel, S.; Kreutzer, K. Carsharing with shared autonomous vehicles: Uncovering drivers, barriers and future developments–A four-stage Delphi study. Technological Forecasting and Social Change 2019, 144, 66– 81,  DOI: 10.1016/j.techfore.2019.03.012
      There is no corresponding record for this reference.
    14. 14
      Greenblatt, J.; Saxena, S. Autonomous taxis could greatly reduce greenhouse-gas emissions of US light-duty vehicles. Nature Climate Change 2015, 5 (9), 860– 865,  DOI: 10.1038/nclimate2685
      There is no corresponding record for this reference.
    15. 15
      Associated Press. Waymo: Phoenix begins Testing Self-Driving Vans on Public Roads. CBS News , 2018. https://www.cbsnews.com/news/waymo-phoenix-tests-self-driving-vans-on-public-roads/.
      There is no corresponding record for this reference.
    16. 16
      Litman, T. Autonomous Vehicle Implementation Predictions; Victoria Transport Policy Institute, 2017.
      There is no corresponding record for this reference.
    17. 17
      Fulton, L. M. Three Revolutions in Urban Passenger Travel. Joule 2018, 2, 575– 578,  DOI: 10.1016/j.joule.2018.03.005
      There is no corresponding record for this reference.
    18. 18
      Zhang, W.; Guhathakurta, S.; Khalil, E. B. The impact of private autonomous vehicles on vehicle ownership and unoccupied VMT generation. Transportation Research Part C: Emerging Technologies 2018, 90, 156– 165,  DOI: 10.1016/j.trc.2018.03.005
      There is no corresponding record for this reference.
    19. 19
      Henao, A.; Marshall, W. E. The impact of ride-hailing on vehicle miles traveled. Transportation 2019, 46, 2173– 2194,  DOI: 10.1007/s11116-018-9923-2
      There is no corresponding record for this reference.
    20. 20
      Sperling, D. Three Revolution: Steering Automated, Shared, and Electric Vehicles to a Better Future; Island Press, 2018.
      There is no corresponding record for this reference.
    21. 21
      Chen, T. D.; Kockelman, K. M.; Hanna, J. P. Operations of a shared, autonomous, electric vehicle fleet: Implications of vehicle & charging infrastructure decisions. Transportation Research Part A: Policy and Practice 2016, 94, 243– 254,  DOI: 10.1016/j.tra.2016.08.020
      There is no corresponding record for this reference.
    22. 22
      Luk, J. M.; Kim, H. C.; De Kleine, R.; Wallington, T. J.; MacLean, H. L. Review of the Fuel Saving, Life Cycle GHG Emission, and Ownership Cost Impacts of Lightweighting Vehicles with Different Powertrains. Environ. Sci. Technol. 2017, 51, 8215– 8228,  DOI: 10.1021/acs.est.7b00909
      22
      Review of the Fuel Saving, Life Cycle GHG Emission, and Ownership Cost Impacts of Lightweighting Vehicles with Different Powertrains
      Luk, Jason M.; Kim, Hyung Chul; De Kleine, Robert; Wallington, Timothy J.; MacLean, Heather L.
      Environmental Science & Technology (2017), 51 (15), 8215-8228CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
      A review concerning fuel conservation, life cycle greenhouse gas (GHG) emissions from, and ownership costs assocd. with light-weighting vehicles with alternative power-trains from financial and environmental perspectives, is given. Topics discussed include: introduction; lightwt. materials background; literature overview; fuel conservation from light-weighting vehicles with different power-trains; power-train re-sizing, drive cycle, and vehicle class effects on fuel conservation from light-weighting vehicles with different power-trains; life cycle literature ests. of fuel conservation from light-weighting vehicles with different power-trains; life cycle greenhouse gas (GHG) emissions; fuel cycle, fuel and vehicle cycle, vehicle cycle, and life cycle GHG emissions impact from light-weighting vehicles with different power-trains; life cycle ownership costs, ownership costs, and secondary mass redns. from light-weighting vehicles with different power-trains; vehicle prodn. costs from light-weighting vehicles with different power-trains; discussion; and supporting information (alternative power-train background, literature overview).
    23. 23
      Bauer, G. S.; Greenblatt, J. B.; Gerke, B. F. Cost, Energy, and Environmental Impact of Automated Electric Taxi Fleets in Manhattan. Environ. Sci. Technol. 2018, 52, 4920– 4928,  DOI: 10.1021/acs.est.7b04732
      23
      Cost, Energy, and Environmental Impact of Automated Electric Taxi Fleets in Manhattan
      Bauer, Gordon S.; Greenblatt, Jeffery B.; Gerke, Brian F.
      Environmental Science & Technology (2018), 52 (8), 4920-4928CODEN: ESTHAG; ISSN:0013-936X. (American Chemical Society)
      Shared automated elec. vehicles (SAEVs) hold great promise for improving transportation access in urban centers while drastically reducing transportation-related energy consumption and air pollution. Using taxi-trip data from New York City, we develop an agent-based model to predict the battery range and charging infrastructure requirements of a fleet of SAEVs operating on Manhattan Island. We also develop a model to est. the cost and environmental impact of providing service and perform extensive sensitivity anal. to test the robustness of our predictions. We est. that costs will be lowest with a battery range of 50-90 mi, with either 66 chargers per square mile, rated at 11 kW or 44 chargers per square mile, rated at 22 kW. We est. that the cost of service provided by such an SAEV fleet will be $0.29-$0.61 per revenue mile, an order of magnitude lower than the cost of service of present-day Manhattan taxis and $0.05-$0.08/mi lower than that of an automated fleet composed of any currently available hybrid or internal combustion engine vehicle (ICEV). We est. that such an SAEV fleet drawing power from the current NYC power grid would reduce GHG emissions by 73% and energy consumption by 58% compared to an automated fleet of ICEVs.
    24. 24
      Franke, T.; Krems, J. F. Understanding charging behaviour of electric vehicle users. Transportation Research Part F: Traffic Psychology and Behaviour 2013, 21, 75– 89,  DOI: 10.1016/j.trf.2013.09.002
      There is no corresponding record for this reference.
    25. 25
      Franke, T.; Gunther, M.; Trantow, M.; Krems, J. F. Does this range suit me? Range satisfaction of battery electric vehicle users. Applied Ergonomics 2017, 65, 191– 199,  DOI: 10.1016/j.apergo.2017.06.013
      25
      Does this range suit me? Range satisfaction of battery electric vehicle users
      Franke Thomas; Gunther Madlen; Trantow Maria; Krems Josef F
      Applied ergonomics (2017), 65 (), 191-199 ISSN:.
      User satisfaction is a vital design criterion for sustainable systems. The present research aimed to understand factors relating to individually perceived range satisfaction of battery electric vehicle (BEV) users. Data from a large-scale BEV field trial (N = 72) were analyzed. Apart from an initial drop in range satisfaction, increasing practical experience was related to increased range satisfaction. Classical indicators of users' mobility profiles (daily travel distances) were only weakly related to lower range satisfaction (not significant), after controlling for practical experience and preferred coverage of mobility needs. The regularity/predictability of users' mobility patterns, the percentage of journeys not coverable because of range issues, and users' individual comfortable range accounted for variance in range satisfaction. Finally, range satisfaction was related to key indicators of general BEV acceptance (e.g., purchase intentions). These results underline the complex dynamics involved in individual range satisfaction, as well as its central role for BEV acceptance.
    26. 26
      Wolbertus, R.; Kroesen, M.; van den Hoed, R.; Chorus, C. G. Policy effects on charging behaviour of electric vehicle owners and on purchase intentions of prospective owners: Natural and stated choice experiments. Transportation Research Part D: Transport and Environment 2018, 62, 283– 297,  DOI: 10.1016/j.trd.2018.03.012
      There is no corresponding record for this reference.
    27. 27
      Melaina, M.; Muratori, M.; McLaren, J.; Schwabe, P. Investing in Alternative Fuel Infrastructure: Insights for California from Stakeholder Interviews: Preprint. National Renewable Energy Laboratory Publications , 2017, NREL/CP-5400-67617. https://www.nrel.gov/docs/fy17osti/67617.pdf
      There is no corresponding record for this reference.
    28. 28
      Nelson, J. H.; Wisland, L. M. Achieving 50% Renewable Electricity in California: The Role of Non-Fossil Flexibility in a Cleaner Electricity Grid; Union of Concerned Scientists, 2015.
      There is no corresponding record for this reference.
    29. 29
      Alstone, P.; Potter, J.; Piette, M.; Schwartz, P.; Berger, M.; Dunn, L.; Smith, S.; Sohn, M.; Aghajanzadeh, A.; Stensson, S.; Szinai, J.; Walter, T.; McKenzie, L.; Lavin, L.; Schneiderman, B.; Mileva, A.; Cutter, E.; Olson, A.; Bode, J.; Ciccone, A.; Jain, A. Final Report on Phase 2 Results, 2015 California Demand Response Potential Study: Charting California’s Demand Response Future; Lawrence Berkeley National Laboratory, 2016.
      There is no corresponding record for this reference.
    30. 30
      Richardson, D. B. Electric vehicles and the electric grid: A review of modeling approaches, Impacts, and renewable energy integration. Renewable Sustainable Energy Rev. 2013, 19, 247– 254,  DOI: 10.1016/j.rser.2012.11.042
      There is no corresponding record for this reference.
    31. 31
      Foley, A.; Tyther, B.; Calnan, P.; O Gallachoir, B. Impacts of Electric Vehicle charging under electricity market operations. Appl. Energy 2013, 101, 93– 102,  DOI: 10.1016/j.apenergy.2012.06.052
      There is no corresponding record for this reference.
    32. 32
      Sheppard, C.; Szinai, J.; Abhyankar, N.; Gopal, A. Grid Impacts of Electric Vehicles and Managed Charging in California: Linking Agent-Based Electric Vehicle Charging with Power System Dispatch Models; Sustainable Energy Systems Group, Sustainable Energy Department, Energy Analysis and Environmental Impacts Division, International Energy Studies Group, International Energy Analysis Department, 2019.
      There is no corresponding record for this reference.
    33. 33
      Szinai, J. K.; Sheppard, C. J. R.; Abhyankar, N.; Gopal, A. R. Reduced grid operating costs and renewable energy curtailment with electric vehicle charge management. Energy Policy 2020, 136, 111051,  DOI: 10.1016/j.enpol.2019.111051
      There is no corresponding record for this reference.
    34. 34
      Lopes, J.; Soares, F.; Almeida, P. Integration of electric vehicles in the electric power system. Proc. IEEE 2011, 99, 168– 183,  DOI: 10.1109/JPROC.2010.2066250
      There is no corresponding record for this reference.
    35. 35
      Tarroja, B. Assessing the stationary energy storage equivalency of vehicle-to-grid charging battery electric vehicles. Energy 2016, 106, 673– 690,  DOI: 10.1016/j.energy.2016.03.094
      There is no corresponding record for this reference.
    36. 36
      Forrest, K. Charging a renewable future: The impact of electric vehicle charging intelligence on energy storage requirements to meet renewable portfolio standards. J. Power Sources 2016, 336, 63– 74,  DOI: 10.1016/j.jpowsour.2016.10.048
      36
      Charging a renewable future: The impact of electric vehicle charging intelligence on energy storage requirements to meet renewable portfolio standards
      Forrest, Kate E.; Tarroja, Brian; Zhang, Li; Shaffer, Brendan; Samuelsen, Scott
      Journal of Power Sources (2016), 336 (), 63-74CODEN: JPSODZ; ISSN:0378-7753. (Elsevier B.V.)
      Increased usage of renewable energy resources is key for energy system evolution to address environmental concerns. Capturing variable renewable power requires the use of energy storage to shift generation and load demand. The integration of plug-in elec. vehicles, however, impacts the load demand profile and therefore the capacity of energy storage required to meet renewable utilization targets. This study examines how the intelligence of plug-in elec. vehicle (PEV) integration impacts the required capacity of energy storage systems to meet renewable utilization targets for a large-scale energy system, using California as an example for meeting a 50% and 80% renewable portfolio std. (RPS) in 2030 and 2050. For an 80% RPS in 2050, immediate charging of PEVs requires the installation of an aggregate energy storage system with a power capacity of 60% of the installed renewable capacity and an energy capacity of 2.3% of annual renewable generation. With smart charging of PEVs, required power capacity drops to 16% and required energy capacity drops to 0.6%, and with vehicle-to-grid (V2G) charging, non-vehicle energy storage systems are no longer required. Overall, this study highlights the importance of intelligent PEV charging for minimizing the scale of infrastructure required to meet renewable utilization targets.
    37. 37
      Coignard, J.; Saxena, S.; Greenblatt, J.; Wang, D. Clean Vehicles as an Enabler for a Clean Electricity Grid. Environ. Res. Lett. 2018, 13, 054031,  DOI: 10.1088/1748-9326/aabe97
      There is no corresponding record for this reference.
    38. 38
      Iacobucci, R.; McLellan, B.; Tezuka, T. Modeling shared autonomous electric vehicles: Potential for transport and power grid integration. Energy 2018, 158, 148– 163,  DOI: 10.1016/j.energy.2018.06.024
      There is no corresponding record for this reference.
    39. 39
      Jones, E. C.; Leibowicz, B. D. Contributions of shared autonomous vehicles to climate change mitigation. Transportation Research Part D: Transport and Environment 2019, 72, 279– 298,  DOI: 10.1016/j.trd.2019.05.005
      There is no corresponding record for this reference.
    40. 40
      Li, M.; Lenzen, M.; Wang, D.; Nansai, K. GIS-based modelling of electric-vehicle-grid integration in a 100% renewable electricity grid. Appl. Energy 2020, 262, 114577,  DOI: 10.1016/j.apenergy.2020.114577
      There is no corresponding record for this reference.
    41. 41
      Hedegaard, K.; Ravn, H.; Juul, N.; Meibom, P. Effects of electric vehicles on power systems in Northern Europe. Energy 2012, 48, 356– 368,  DOI: 10.1016/j.energy.2012.06.012
      There is no corresponding record for this reference.
    42. 42
      Haque, A. N. M. M.; Ibn Saif, A. U. N.; Nguyen, P. H.; Torbaghan, S. S. Exploration of dispatch model integrating wind generators and electric vehicles. Appl. Energy 2016, 183, 1441– 1451,  DOI: 10.1016/j.apenergy.2016.09.078
      There is no corresponding record for this reference.
    43. 43
      Hanemann, P.; Behnert, M.; Bruckner, T. Effects of electric vehicle charging strategies on the German power system. Appl. Energy 2017, 203, 608– 622,  DOI: 10.1016/j.apenergy.2017.06.039
      There is no corresponding record for this reference.
    44. 44
      Taljegard, M.; Göransson, L.; Odenberger, M.; Johnsson, F. Impacts of electric vehicles on the electricity generation portfolio - A Scandinavian-German case study. Appl. Energy 2019, 235, 1637– 1650,  DOI: 10.1016/j.apenergy.2018.10.133
      There is no corresponding record for this reference.
    45. 45
      Limmer, S.; Rodemann, T. Peak load reduction through dynamic pricing for electric vehicle charging. International Journal of Electrical Power & Energy Systems 2019, 113, 117– 128,  DOI: 10.1016/j.ijepes.2019.05.031
      There is no corresponding record for this reference.
    46. 46
      Sheppard, C.; Jenn, A.; Bauer, G.; Gerke, B.; Greenblatt, J.; Gopal, A. A joint optimization scheme for the planning and operations of a regional electrified fleets of ride hailing vehicles serving mobility on demand. Transportation Research Record: Journal of the Transportation Research Board 2019, 2673 (6), 579– 597,  DOI: 10.1177/0361198119838270
      There is no corresponding record for this reference.
    47. 47
      Jenn, A.; Clark-Sutton, K.; Gallaher, M. P.; Petrusa, J. Environmental impacts of extreme fast charging. Environ. Res. Lett. 2020, 15, 094060,  DOI: 10.1088/1748-9326/ab9870
      47
      Environmental impacts of extreme fast charging
      Jenn, Alan; Clark-Sutton, Kyle; Gallaher, Michael; Petrusa, Jeffrey
      Environmental Research Letters (2020), 15 (9), 094060CODEN: ERLNAL; ISSN:1748-9326. (IOP Publishing Ltd.)
      As elec. vehicles and their assocd. charging infrastructure continue to evolve, there is potential to simultaneous alleviate range and recharge concerns with the development of extreme fast chargers (XFC) that can fully charge batteries in PEVs in the span of a few minutes. Recent announcements from EVSE providers and vehicle manufacturers suggest that XFC charging stations, which can recharge a BEV at roughly 20 to 25 mi per min of charging, and XFC-capable BEVs, could be com. available within the next 5 years. Our study investigates the potential emission impacts of widespread use of extreme fast charging (350 kW) for elec. vehicles in 2030. We conduct a novel vehicle charging simulation model by combining empirical charging behavior data across several data sources. These charging demands are then added as exogenous load to the Grid Optimized Operation Dispatch (GOOD) model, which simulates the operation of generators across the Untied States. We fiend that XFC can increase both greenhouse gas emissions and local air pollutants, though the results are sensitive to local contexts and grid compn.
    48. 48
      Hand, M. M.; Baldwin, S.; DeMeo, E.; Reilly, J. M.; Mai, T.; Arent, D.; Porro, G.; Meshek, M.; Sandor, D. Renewable Electricity Futures Study; National Renewable Energy Laboratory, 2012.
      There is no corresponding record for this reference.
    49. 49
      Wood, E. W.; Rames, C. L.; Muratori, M.; Srinivasa Raghavan, S.; Young, S. E. Charging Electric Vehicles in Smart Cities: An EVI-Pro Analysis of Columbus; NREL: OH, 2018.
      There is no corresponding record for this reference.
    50. 50
      Sheppard, C.; Jenn, A. Grid-integrated Electric Mobility (GEM) Model. GitHub , 2021. https://github.com/LBNL-UCB-STI/gem.
      There is no corresponding record for this reference.
    51. 51
      Sheppard, C.; Jenn, A.; Bauer, G.; Greenblatt, J.; Gerke, B. Grid-integrated Electric Mobility (GEM) Model – Manuscript 2021 Branch. GitHub , 2021. https://github.com/LBNL-UCB-STI/gem/tree/manuscript-2021.
      There is no corresponding record for this reference.
    52. 52
      U.S. Department of Transportation, Federal Highway Administration 2017 National Household Travel Survey , 2017. https://nhts.ornl.gov.
      There is no corresponding record for this reference.
    53. 53
      Xu, Z.; Su, W.; Hu, Z.; Song, Y.; Zhang, H. A Hierarchical Framework for Coordinated Charging of Plug-In Electric Vehicles in China. IEEE Trans. Smart Grid 2016, 7, 428– 438,  DOI: 10.1109/TSG.2014.2387436
      There is no corresponding record for this reference.
    54. 54
      U.S. Energy Information Administration. Annual Energy Outlook , 2019. https://www.eia.gov/outlooks/aeo/.
      There is no corresponding record for this reference.
    55. 55
      U.S. Energy Information Administration Preliminary Monthly Electric Generator Inventory (based on Form EIA-860M as a supplement to Form EIA-860). https://www.eia.gov/electricity/data/eia860M/ (accessed September 2020).
      There is no corresponding record for this reference.

  • Supporting Information

    Supporting Information

    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.0c06655.

    • Model specification, NHTS data, correction factors using RISE, EVI-Pro, additional metrics, geographic analysis, sensitivity runs, and references (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.