PreprintPDF Available

The Impact of Transitioning to Shared Electric Vehicles on Grid Congestion and Management

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

The transition towards a sharing economy and the increasing electrification of the transport sector are occurring simultaneously. Consequently, we can expect more car sharing schemes using electric vehicles (EVs) to emerge in coming years. Numerous studies looked into the grid impact of EV charging and its potential to provide ancillary services, but these studies only considered regular EVs. This study compares the charging patterns of regular and shared EVs and creates insight in the grid impact and potential to provide ancillary services with future adoption of shared EVs. Four scenarios for the adoption of shared EVs are proposed, and a method to generate a set of future charging transactions based on historical charging data is presented. The analysis is performed using charging data from an EV sharing company in the Netherlands. Results indicate that charging demand peaks and grid congestion levels decrease substantially with higher adoption of shared EVs. Adoption of shared EVs increases the potential of EVs to provide ancillary services, due to a higher charging flexibility of shared EVs.
Content may be subject to copyright.
The Impact of Transitioning to Shared Electric
Vehicles on Grid Congestion and Management
Nico Brinkel
Copernicus Institute of Sustainable
Development, Utrecht University
n.b.g.brinkel@uu.nl
Tarek AlSkaif
Information Technology Group
Wageningen University and Research
tarek.alskaif@wur.nl
Wilfried van Sark
Copernicus Institute of Sustainable
Development, Utrecht University
w.g.j.h.m.vansark@uu.nl
Abstract—The transition towards a sharing economy and the
increasing electrification of the transport sector are occurring
simultaneously. Consequently, we can expect more car sharing
schemes using electric vehicles (EVs) to emerge in coming years.
Numerous studies looked into the grid impact of EV charging
and its potential to provide ancillary services, but these studies
only considered regular EVs. This study compares the charging
patterns of regular and shared EVs and creates insight in the grid
impact and potential to provide ancillary services with future
adoption of shared EVs. Four scenarios for the adoption of
shared EVs are proposed, and a method to generate a set of
future charging transactions based on historical charging data is
presented. The analysis is performed using charging data from
an EV sharing company in the Netherlands. Results indicate
that charging demand peaks and grid congestion levels decrease
substantially with higher adoption of shared EVs. Adoption of
shared EVs increases the potential of EVs to provide ancillary
services, due to a higher charging flexibility of shared EVs.
Index Terms—sharing economy, electric vehicles, car sharing,
grid congestion, ancillary services
NOMENCLATURE
Indices and sets
j∈ J Set of EV categories (shared EVs and regular EVs)
n∈ N Set of charging transactions
Symbols
Tch,n Actual charging time of transaction n.
Tcon,n Connection time to charging station of transaction n.
Tflex,n Flexibility in charging demand of transaction n.
darr,n,j Arrival day of the week of charging transaction nof
category j.
Eann,j Predetermined annual charging demand of all EV
transactions of category jin an LV grid.
Ereq,n,j Charging demand of transaction nof category j.
harr,n,j Arrival hour of charging transaction nof category j.
Pav,n Average charging power of transaction n.
Pmax,n Average charging power of transaction n.
Spl,j Set of historical charging transactions of category j
of which the charging power is logged.
SjSet of historical charging transactions of category j.
I. INTRODUCTION
Stimulated by among others battery cost reductions and gov-
ernmental incentives to decarbonize road transport, the share
This study was supported by the European Regional Development Fund
‘EFRO Kansen voor West II’ through the project ‘Smart Solar Charging’.
of electric vehicles (EVs) in the passenger car fleet is growing
rapidly [1], [2]. High adoption of EVs can cause power quality
and congestion problems in the low-voltage (LV) grid [3],
since the extra load of EVs was not considered when designing
the grid. At the same time, the battery capacity of EVs can be
used to mitigate power quality and grid congestion problems
due to the high flexibility in EV charging demand. For this
reason, a large number of studies have estimated the future
impact of EV charging on grid congestion (i.a., [4], [5]) and
have assessed the potential of EVs to mitigate power quality
and congestion problems (i.a., [6]–[8]).
In parallel, a paradigm shift towards the sharing economy
is emerging in recent years [9]. The application of the sharing
economy to the transport sector could result in high market
penetration of car sharing schemes [10]. Car sharing is an
alternative to car ownership and provides users short-term
access to a set of shared cars managed by a third-party orga-
nization [11]. The number of car sharing members worldwide
has increased with a factor of 40 between 2006 and 2018
[12]. Different studies expect that the market share of car
sharing will continue to increase [13]–[15]. Further adoption
of car sharing in combination with a market uptake of EVs
could result in an increasing charging demand of shared EVs.
This increased role of shared EVs could affect the results of
previous studies on the future grid impact of EV charging and
the potential of EVs to provide ancillary services, since shared
EVs are driven by more people, have different arrival times
and are used for different functions.
The first aim of this study is to create insights in the impact
of a shift towards shared EVs on the electricity grid. This study
uses historical charging data from an EV sharing company
in the Netherlands to compare the charging characteristics
of shared and regular EVs, and proposes a novel method
to generate future sets of EV charging transactions. These
transaction sets are used to compare EV charging patterns
in different adoption scenarios of shared EVs. Second, this
study compares the flexibility in charging demand of shared
and regular EVs to determine how the potential of EVs in pro-
viding ancillary services changes with a shift towards shared
EVs. The results of this analysis provide distribution system
operators (DSOs) with insights in future grid congestion levels
with high adoption of shared EVs, and also indicate whether
shared EVs are a suitable technology for the provision of
ancillary services. In addition, the results of this study show
policymakers whether a shift towards shared EVs should be
stimulated from a grid perspective.
This study is outlined as follows: Section II introduces
the considered car sharing scheme and the investigated case
study grid. Section III provides a comparison of the key
charging characteristics of shared and regular EVs. Section
IV presents a method to simulate future charging transactions
and outlines different future adoption scenarios of shared EVs.
The aggregated EV charging behavior in a LV grid, flexibility
in charging demand and grid impact are presented for all
considered scenarios in Section V. Lastly, a discussion and
conclusion are presented in respectively Section VI and VII.
II. CA SE S TU DY SP EC IFI CATI ON S
We Drive Solar is a car sharing company offering station-
based car sharing services using EVs. As of February 2020, it
has a portfolio of three Tesla Model 3’s (battery capacity 50
kWh) and 72 Renault ZOE’s (battery capacity 44-52 kWh).
Users need a subscription to drive a We Drive Solar EV, and
also pay per driven kilometer.
EV charging data is obtained between 8 January 2019 and
10 March 2020 from 230 charging stations with two charging
sockets in residential areas in the Netherlands, of which the
large majority are located in the city of Utrecht. Each of
these charging stations log the arrival time, departure time,
car ID and charging volume of each charging transaction. 24
stations also log the charging power over time, which is used
to determine the average charging power, maximum charging
power and actual charging duration for each transaction. A
distinction is made between shared EVs and regular EVs based
on the car ID of the We Drive Solar EVs. All We Drive Solar
EVs use one of the logged EV charging stations as their home
station. Table I provides an overview of the number of logged
charging transactions for regular and shared EVs.
TABLE I: Overview of logged charging transactions in the
input data.
No. of logged charg-
ing transactions
No. of charging transactions in
which charging power is logged
Regular EVs 28,621 11,570
Shared EVs 8,722 1,963
A residential grid in the Lombok district in Utrecht, the
Netherlands serves as a case study grid to analyze the grid
impact of EV charging. This grid is connected to the medium-
voltage (MV) grid through a 400 kVA transformer and serves
340 grid connections, of which the majority is households. It
is assumed that 25% of the households will use a full-electric
heat pump (HP) for heating in the future. The assumed future
installed photovoltaic (PV) capacity in the grid equals 200
kWp. The profiles of HP demand, residential load and PV
generation are created using the methods in [8].
III. CHARGING CHARACTERISTICS OF SHARED AND
REGULAR EVS
Fig. 1 compares key characteristics of charging transactions
of shared and regular EVs for the selected case study. It
shows considerable differences in the arrival times of both
EV categories. The arrival hours of regular EVs in Fig. 1a
show a distinct peak between 17:00-19:00, induced by EV-
owners returning home from work. The minor peak between
8:00-10:00 in the arrival hours of regular EVs are caused by
people arriving at work and charging their car in a residential
area close to their working location. The majority of charging
transactions of shared EVs start in the late afternoon or early
evening, but the arrival times do not peak in one or two specific
hours, while a substantial share of shared EVs arrive in the
late evening. In addition, Fig. 1b shows that the number of
charging transactions is substantially higher in weekends for
shared EVs. Both trends could indicate that shared EVs are
used relatively often for leisure and/or family/friend visits.
While regular EVs charge in some cases up to 75 kWh
in Fig. 1c, the charging volume remains below 50 kWh in
all charging transactions with shared EVs, due the absence
of shared EVs with a battery capacity above 54 kWh in the
studied fleet. Regular EVs show a peak between 3 and 8 kWh
in the distribution of charging volume in Fig. 1c and between
3 and 4 kW in the distribution of charging power in Fig. 1d,
which can be attributed to plug-in hybrids in the fleet of regular
EVs, which have a relatively low battery capacity (<12 kWh)
and a maximum charging power of 3.7 kW [4].
The relatively high share of transactions of shared EVs with
a connection time below one hour in Fig. 1e demonstrates high
utilization of shared EVs at specific moments. However, the
utilization is not always high, as the connection time exceeds
72 hours in a substantial share of the charging transactions.
IV. SCENARIO DEVELOPMENT
This section shows how large-scale adoption of shared EVs
may affect the aggregated EV charging behavior in a LV grid.
This section presents different scenarios for future adoption of
shared EVs based on scientific literature, after which a method
is provided to generate future EV charging transactions.
A. Scenario Overview
Studies on expected future trends in car sharing show high
ambiguity in the future adoption of shared cars. Shaneen
& Cohen [16] conducted expert interviews and arrived at
a market potential of shared cars of 2-10%, while 70% of
the experts interviewed in [15] expected a market potential
of shared cars between 11-25%. Other studies estimate the
future adoption of shared cars using activity-based models.
Martinez et al. [17] estimate that 2.4% of the trips in Lisbon,
Portugal will be conducted using shared cars when introducing
a car sharing scheme, while the model in [18] estimated a
future market share of car sharing services of 1-3.8%. A stated
preference study by Namazu et al. [14] in Vancouver, Canada
showed that 25% of the inhabitants cannot be convinced to
move to car sharing. Liao et al. [13] indicated that 13.9% of
the car owners in the Netherlands will use shared cars for all
car trips if a car sharing scheme is available, while 63.4% of
the car owners will not use a car sharing scheme. Rotaris &
Danielis [19] indicated that only 3.7% of the inhabitants in a
Fig. 1: Histograms comparing key charging characteristics of shared and regular EVs in the input data.
rural area in Italy have a probability of >50% of adopting a
shared EV.
In some urban areas, the car sharing potential could be much
higher. The large future role of car sharing in specific areas is
highlighted by the recently published plans for the Merwede
district in the city of Utrecht [20]. This to be developed
neighborhood will become car-free, meaning that there are no
parking spots available for private cars. One shared car will
be available to every three households in the district.
TABLE II: Overview of annual EV charging demand of shared
and regular EVs per scenario.
Regular EVs Shared EVs
Reference scenario 530 MWh 0 MWh
Scenario 1 450 MWh 64 MWh
Scenario 2 0 MWh 285 MWh
Scenario 3 132 MWh 214 MWh
Given this high ambiguity about the role of shared EVs in
the future car fleet, this study distinguishes different scenarios
for the adoption of shared EVs when determining the future
grid impact of EV charging. All scenarios assume that the fu-
ture car fleet is completely electrified. Each scenario identifies
the annual charging demand of all regular and shared EVs in
the investigated grid (Eann). These values are used in section
IV-B to simulate future charging transactions for both EV
groups. Table II presents the assumed annual charging demand
of all regular and shared EVs per scenario. The scenarios were
generated as follows:
Reference scenario - Shared EVs are not part of the car
fleet in this scenario. The total annual charging demand
of regular EVs is based on the average mileage per car of
13,000 km/year [21], a driving efficiency of 0.2 kWh/km
and the car ownership ratio of 0.6 cars/household [22] in
the city of Utrecht.
Scenario 1: Limited adoption of shared EVs - Based on
the literature study above, this scenario assumes that 15%
of the car-owners adopt a shared EV, corresponding to a
reduction in charging demand of regular EVs of 15%.
Adoption of car sharing in the Netherlands leads to a
reduction of vehicle kilometers travelled (VKT) of around
20% [23]. This has been considered when determining the
charging demand of shared EVs.
Scenario 2: Only shared EVs - This scenario is inspired
by the newly-developed car-free districts and assumes
one shared EV per three households. The average annual
charging demand of a single We Drive Solar shared EV
equals 2519 kWh. It is assumed that the usage of shared
EVs in car-free districts is not considerably different
than the current usage in other districts. This assumption
implies that people living in this district will mostly
commute using other means of transport.
Scenario 3: Mix of shared and regular EVs - This scenario
assumes 25% of the car owners are non-adopters of
shared EVs, based on [14]. The charging demand of
regular EVs equals 25% of the charging demand in the
reference scenario. The charging demand of shared EVs
equals 75% of the charging demand in Scenario 2.
B. Generating Future EV Charging Transactions
Charging transactions for shared EVs and regular EVs are
simulated for one year to obtain insight in EV charging
behavior in the different scenarios. A new EV charging trans-
action n∈ N is created until the total charging requirement
of all simulated transactions of a specific category j
{Shared EV, Regular EV}(Ereq,n,j ) equals the predetermined
charging demand of all EVs of category j(Eann,j):
N
X
n=1
Ereq,n,j =Eann,j j. (1)
The simulated charging transaction set is created using a
probabilistic model which considers a set of historical charging
transactions Sjfor each EV category. In case the charging
power over time is not logged at all charging stations, the
subset Spl,j of Sjis created with all charging transactions of
which the charging power is logged. Charging transactions are
simulated using the following steps:
1) The arrival day of the week of charging transaction n
(darr,n,j ) is determined based on the probability density
function of the arrival day of the week in Sj. Subse-
quently, this transaction is randomly assigned to a date
with the specific weekday in the simulation period.
2) The arrival hour harr,n,j is determined using a probability
density function of the arrival hours of all EVs charging
transactions in Sjarriving at weekday darr,n,j. The
charging transaction is randomly assigned a starting
minute in the arrival hour.
3) The required charging volume Ereq,n,j and connection
time to the charging station Tcon,n are determined
by randomly selecting a charging transaction from all
EV charging transactions in Sjarriving at weekday
darr,n,j and arriving at harr,n,j. The charging volume and
connection time of this specific charging transaction are
assigned to the simulated charging transaction.
4) The average and maximum charging power (Pav,n and
Pmax,n) of a simulated charging transaction are deter-
mined by using the charging power of a randomly
selected charging transaction from all charging trans-
actions in Spl,j arriving at weekday darr,n and at hour
harr,n. If the condition in eq. 2 is not met, the steps to
determine Pav,n and Pmax,n are repeated:
Pav,n Ereq,n,j
Tcon,n
.(2)
If no transaction in Spl,j arriving at weekday darr,n and
at hour harr,n can meet this condition, Pav,n and Pmax,n
are determined using eq. 3:
Pav,n, Pmax,n =Ereq,n,j
Tcon,n
.(3)
Since Spl,j Sj, it can occur that there are no or a
very limited number of transactions in Spl,j arriving at
weekday darr,n and at hour harr,n. In that case, the set of
charging transactions from which a transaction is ran-
domly selected is expanded to all charging transactions
arriving at hour harr,n for all weekdays if darr,n is a
weekday, or all weekend days if darr,n is a weekend day.
V. RE SU LTS
A. Charging Patterns
Fig. 2 compares the average EV charging power during
week and weekend days for the four considered scenarios.
Both in the Reference Scenario and Scenario 1 a charging de-
mand peak occurs around 19:00 on weekdays, caused by EVs
arriving home from work. Comparing the charging demand
peak between both scenarios indicates that limited adoption of
shared EVs reduces the average peak in charging demand on
weekdays by 8%, due to a lower overall charging demand and
a lower simultaneity in arrival hours in Scenario 1. The average
overall charging demand on weekdays decreases from 1433
to 1366 kWh when transitioning from the Reference Scenario
to Scenario 1. Despite the lower overall charging demand in
Scenario 1, the average charging peak and overall charging
demand in weekends are similar in both scenarios due to the
large utilization of shared EVs in weekends.
A large scale transition towards shared EVs in Scenario 2
and 3 has substantial impact on the charging peak and overall
charging demand on weekdays. The average charging peak
on weekdays in Scenario 2 reduces by 75 kW compared to
the Reference Scenario, while the average overall charging
demand on weekdays reduces from 1433 to 681 kWh. A large
share of the charging demand is shifted towards weekends with
high adoption of shared EVs; the average overall charging de-
mand on weekend days is 51% higher compared to weekdays
in Scenario 2, while this is 4% in the Reference Scenario.
Charging demand peaks and overall charging demand are
similar between Scenario 2 and 3, indicating that a limited
share of regular EVs in the car fleet still leads to a considerable
reduction in the peak charging demand compared to a scenario
without shared EVs.
B. Charging Flexibility
TABLE III: Average share of charging demand that can be
delayed for at least 6 and 12 hours at different moments of
the day in each considered scenario.
Ref. scenario Scenario 1 Scenario 2 Scenario 3
>6h >12h >6h >12h >6h >12h >6h >12h
00:00-06:00 24% 7% 26% 9% 45% 25% 35% 16%
06:00-12:00 9% 2% 10% 4% 25% 20% 15% 9%
12:00-18:00 20% 15% 22% 18% 35% 33% 30% 27%
18:00-24:00 36% 13% 37% 16% 47% 36% 43% 28%
Total 27% 11% 28% 14% 41% 31% 35% 24%
Flexibility in charging demand provides insight in the
potential of EVs to provide ancillary services, as EVs have
more room to alter their charging behavior to provide ancillary
services with high charging flexibility. It can be expressed as
the difference between the actual charging time Tch and the
Fig. 2: Average charging power and average flexibility in EV charging demand on weekdays and weekend days in the studied
LV grid for the four scenarios introduced in section IV-A.
connection time to the charging station Tcon, as introduced
in [4]:
Tflex,n = ∆Tcon,n Tch,n.(4)
Table III and the colors in Figure 2 indicate that charging
flexibility is generally high. In all considered scenarios, at least
27% of charging can be delayed with 6 hours or more, at least
11% of charging can be delayed with 12 hours or more, while
only up to 12% of charging can be delayed with maximum
one hour. EVs parking overnight cause that charging flexibility
is highest during the charging demand peak in the evening.
Flexibility in charging demand increases with a higher share
of shared EVs in the car fleet. Since car sharing companies
must guarantee availability of EVs to their customers, there
will always be a slight overcapacity of EVs. This results in
a high connection time of shared EVs in different cases (see
Section III) leading to higher charging flexibility.
C. Grid Impact
Fig. 3 presents the transformer load in the investigated
grid for each considered scenario. The grid load exceeds
the transformer capacity for more than 2.5% of the year in
all scenarios. Transformer congestion levels decrease with
higher adoption of shared EVs, as this reduces the overall
charging demand in the grid. The difference in transformer
congestion levels is minor between the Reference Scenario
and Scenario 1 (476 versus 446 hours/year). Large scale
adoption of shared EVs can reduce the number of hours with
transformer congestion with over 50%, with 235 hours/year
of transformer congestion in Scenario 2 and 248 hours/year in
Scenario 3.
Fig. 3: Load duration curve of transformer load in the inves-
tigated grid for the four considered scenarios.
VI. DISCUSSION
This study presented a generic method to use historical
charging transactions to get insight in the future charging
patterns of shared EVs. However, different methodological
considerations should be taken into account when interpreting
the results. With higher adoption of car sharing, the utilization
of shared EVs could increase and different user groups could
adopt shared EVs, potentially leading to different arrival times
or to lower connection times. On the other hand, one can
expect that there will always be an overcapacity in the number
of shared EVs to guarantee availability of shared EVs to users,
causing that flexibility values will remain similar in the future.
As there is uncertainty about the departure times of EVs in
absence of perfect foresight and not all EV users will allow
their EV to be used for the provision of ancillary services, the
reported flexibility values will be lower in practice. Shared
EVs can only be used through a reservation system, causing
that the departure time of a shared EV is more predictable and
that a larger share of the charging flexibility of shared EVs can
be used for the provision of ancillary services.
Future developments, including increased charging at fast-
charging stations or a transition towards autonomous transport
or hydrogen vehicles, could reduce the overall charging de-
mand in LV grids, lowering grid congestion levels. Similarly,
these congestion levels are affected by different external
factors, including the installed PV capacity and the integration
of HPs in the grid.
An implication of this analysis is that DSOs can consider
lower charging demand peaks in grids with a considerable
share of shared EVs in the local car fleet. The high charging
flexibility of shared EVs shows that grid operators and market
parties should target shared EVs for the provision of ancillary
services or for participation in electricity markets.
VII. CONCLUSIONS AND FUTURE RESEARCH
This study compared the charging behavior of EVs in a
station-based car sharing scheme with the charging behavior
of regular EVs to get insight in the future grid impact of EV
charging with a transition towards car sharing. Results of the
study indicated that adoption of shared EVs leads to lower
EV charging peaks, since shared EVs show less simultaneity
in arrival hours than regular EVs. The charging demand peak
on week days decreases with 8% with limited adoption of
shared EVs compared to a scenario without shared EVs, while
the charging demand peak decreases with over 50% with
high adoption of shared EVs. Consequentially, the number of
hours with grid congestion decreases substantially with higher
adoption of shared EVs. The potential of EVs to provide
ancillary services is substantially higher with high adoption
of shared EVs, in particular during the day, causing that a
transition to shared EVs provides grid operators with more
flexibility resources for the provision of ancillary services.
Future research could use more sophisticated models (e.g.,
activity-based models) to get more realistic insight in the
utilization of shared EVs by new user groups. This allows
for more detailed analyses in the charging behavior and grid
impact of shared EVs with high adoption of car sharing. To be
able to quantify the potential of shared EVs to provide ancil-
lary services, charging optimization models for the provision
of ancillary services should be applied to transaction data of
shared EVs in future research.
REFERENCES
[1] IEA, “Global EV Outlook 2019,” Tech. Rep., 2019.
[2] Netherlands Enterprise Agency (RVO), “Statistics Electric Vehicles in
the Netherlands (upto June 2019),” Tech. Rep. July, 2019.
[3] E. Ucer, M. C. Kisacikoglu, and A. Cafer Gurbuz, “Learning EV
Integration Impact on a Low Voltage Distribution Grid,” IEEE Power
and Energy Society General Meeting, vol. 2018-Augus, pp. 1–5, 2018.
[4] M. K. Gerritsma, T. A. AlSkaif, H. A. Fidder, and W. G. van Sark,
“Flexibility of Electric Vehicle Demand: Analysis of Measured Charging
Data and Simulation for the Future,” World Electric Vehicle Journal,
vol. 10, no. 1, p. 14, 2019.
[5] S. Habib, M. Kamran, and U. Rashid, “Impact analysis of vehicle-to-grid
technology and charging strategies of electric vehicles on distribution
networks - A review,” 2015.
[6] N. Brinkel, W. Schram, T. AlSkaif, I. Lampropoulos, and W. van Sark,
“Should we reinforce the grid? Cost and emission optimization of
electric vehicle charging under different transformer limits,Applied
Energy, vol. 276, no. October, p. 115285, 2020.
[7] M. Raoofat, M. Saad, S. Lefebvre, D. Asber, H. Mehrjedri, and
L. Lenoir, “Wind power smoothing using demand response of electric
vehicles,” International Journal of Electrical Power and Energy Systems,
2018.
[8] N. Brinkel, M. Gerritsma, T. AlSkaif, I. Lampropoulos, A. van Voorden,
H. Fidder, and W. van Sark, “Impact of rapid PV fluctuations on power
quality in the low-voltage grid and mitigation strategies using electric
vehicles,” International Journal of Electrical Power & Energy Systems,
vol. 118, 2020.
[9] K. Frenken, “Political economies and environmental futures for the
sharing economy,Philosophical Transactions of the Royal Society A:
Mathematical, Physical and Engineering Sciences, vol. 375, no. 2095,
2017.
[10] D. Wosskow, “Unlocking the sharing economy: An independent review,”
2014, no. November, p. 43.
[11] S. A. Shaheen, N. D. Chan, and H. Micheaux, “One-way carsharing’s
evolution and operator perspectives from the Americas,Transportation,
vol. 42, no. 3, pp. 519–536, 2015.
[12] S. Shaheen, A. Cohen, and M. Jaffee, “Innovative
Mobility: Carsharing Outlook,” Tech. Rep., 2018. [Online].
Available: http://innovativemobility.org/?project=innovative-mobility-
carsharing-outlook-spring-2018
[13] F. Liao, E. Molin, H. Timmermans, and B. van Wee, Carsharing: the
impact of system characteristics on its potential to replace private car
trips and reduce car ownership. Springer US, 2018, vol. 47, no. 2.
[14] M. Namazu, D. MacKenzie, H. Zerriffi, and H. Dowlatabadi, “Is carshar-
ing for everyone? Understanding the diffusion of carsharing services,
Transport Policy, vol. 63, no. January, pp. 189–199, 2018.
[15] F. Zhou, Z. Zheng, J. Whitehead, R. Perrons, L. Page, and S. Washing-
ton, “Projected prevalence of car-sharing in four Asian-Pacific countries
in 2030: What the experts think,” Transportation Research Part C:
Emerging Technologies, vol. 84, pp. 158–177, 2017.
[16] S. A. Shaheen and A. P. Cohen, “Carsharing and Personal Vehicle
Services: Worldwide Market Developments and Emerging Trends,
International Journal of Sustainable Transportation, vol. 7, no. 1, pp.
5–34, 2012.
[17] L. M. Mart´
ınez, G. H. d. A. Correia, F. Moura, and M. Mendes
Lopes, “Insights into carsharing demand dynamics: Outputs of an agent-
based model application to Lisbon, Portugal,” International Journal of
Sustainable Transportation, vol. 11, no. 2, pp. 148–159, 2017.
[18] S. Le Vine, M. Lee-Gosselin, A. Sivakumar, and J. Polak, “A new
approach to predict the market and impacts of round-trip and point-
to-point carsharing systems: Case study of London,” Transportation
Research Part D: Transport and Environment, vol. 32, pp. 218–229,
2014.
[19] L. Rotaris and R. Danielis, “The role for carsharing in medium to small-
sized towns and in less-densely populated rural areas,Transportation
Research Part A: Policy and Practice, vol. 115, no. August 2017, pp. 49–
62, 2018. [Online]. Available: https://doi.org/10.1016/j.tra.2017.07.006
[20] Municipality of Utrecht, “Omgevingsvisie Merwedekanaalzone,
Tech. Rep. december, 2018. [Online]. Available:
https://www.utrecht.nl/fileadmin/uploads/documenten/bestuur-en-
organisatie/beleid/omgevingsvisie/ deelgebied Merwedekanaalzone/2018-
03-MWKZ-Omgevingsvisie–deel-1-ruimtelijke-agenda 01.pdf
[21] CBS Statline, “Verkeersprestaties personenauto’s; kilometers, brandstof-
soort, grondgebied,” 2019. [Online]. Available: https://www.cbs.nl/nl-
nl/cijfers/detail/80428ned
[22] CBS Statline, “Kerncijfers buurten en wijken 2017,” 2017. [On-
line]. Available: https://www.cbs.nl/nl-nl/maatwerk/2017/31/kerncijfers-
wijken-en-buurten-2017
[23] H. Nijland and J. van Meerkerk, “Mobility and environmental impacts of
car sharing in the Netherlands,” Environmental Innovation and Societal
Transitions, vol. 23, pp. 84–91, 2017.
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
With high electric vehicle (EV) adoption, optimization of the charging process of EVs is becoming increasingly important. Although the CO2 emission impact of EVs is heavily dependent on the generation mix at the moment of charging, emission minimization of EV charging receives limited attention. Generally, studies neglect the fact that cost and emission savings potential for EV charging can be constrained by the capacity limits of the low-voltage (LV) grid. Grid reinforcements provide EVs more freedom in minimizing charging costs and/or emissions, but also result in additional costs and emissions due to reinforcement of the grid. The first aim of this study is to present the trade-off between cost and emission minimization of EV charging. Second, to compare the costs and emissions of grid reinforcements with the potential cost and emission benefits of EV charging with grid reinforcements. This study proposes a method for multi-objective optimization of EV charging costs and/or emissions at low computational costs by aggregating individual EV batteries characteristics in a single EV charging model, considering vehicle-to-grid (V2G), EV battery degradation and the transformer capacity. The proposed method is applied to a case study grid in Utrecht, the Netherlands, using highly-detailed EV charging transaction data as input. The results of the analysis indicate that even when considering the current transformer capacity, cost savings up to 32.4% compared to uncontrolled EV charging are possible when using V2G. Emission minimization can reduce emissions by 23.6% while simultaneously reducing EV charging costs by 13.2%. This study also shows that in most cases, the extra cost or emission benefits of EV charging under a higher transformer capacity limit do not outweigh the cost and emissions for upgrading that transformer.
Article
Full-text available
Cloud transients cause rapid fluctuations in the output of photovoltaic (PV) systems, which can significantly affect the voltage levels in a low-voltage (LV) grid with high penetration of PV systems. These voltage fluctuations may lead to violation of the existing power quality standards. This study estimates the impact of rapid PV output fluctuations on the power quality in an existing LV grid by performing load flow analyses for scenarios in the years 2017, 2030 and 2050 using PV data with 20-second resolution. In this study, we propose a system for the mitigation of PV output fluctuations by altering the charging processes of electric vehicles (EVs) and we assess the effectiveness of the proposed system. Results indicate that PV output fluctuations have minor impact on the voltage levels in the year 2030, but PV output fluctuations induce considerable voltage fluctuations in the year 2050. The magnitude of the voltage fluctuations is dependent on the location in the grid, the installed PV capacity and the grid configuration. These voltage fluctuations can induce visible and annoying light flicker for a significant part of the day in the year 2050. Implementing the proposed system shows that EV technology can contribute in reducing the amount of visible and annoying light flicker considerably, however at the expense of increased charging costs for EV owners.
Article
Full-text available
This paper proposes a method for analyzing and simulating the time-dependent flexibility of electric vehicle (EV) demand. This flexibility is influenced by charging power, which depends on the charging stations, the EV characteristics, and several environmental factors. Detailed charging station data from a Dutch case study have been analysed and used as input for a simulation. In the simulation, the interdependencies between plug-in time, connection duration, and required energy are respected. The data analysis of measured data reveals that 59% of the aggregated EV demand can be delayed for more than 8 h, and 16% for even more than 24 h. The evening peak shows high flexibility, confirming the feasibility of congestion management using smart charging within flexibility constraints. The results from the simulation show that the average daily EV demand increases by a factor 21 between the ‘Present-day’ and the ‘High’ scenario, while the maximum EV demand peak increases only by a factor 6, as a result of the limited simultaneity of the transactions. Further, simulations using the average charging power of individual measured transactions yield more accurate results than simulations using a fixed value for charging power. The proposed method for simulating future EV flexibility provides a basis for testing different smart charging algorithms.
Article
Full-text available
This paper aims to explore the potential of carsharing in replacing private car trips and reducing car ownership and how this is affected by its attributes. To that affect, a stated choice experiment is conducted and the data are analyzed by latent class models in order to incorporate preference heterogeneity. The results show that around 40% of car drivers indicated that they are willing to replace some of their private car trips by carsharing, and 20% indicated that they may forego a planned purchase or shed a current car if carsharing becomes available near to them. The results further suggest that people vary significantly with respect to these two stated intentions, and that a higher intention of trip replacement does not necessarily correspond to higher intention of reducing car ownership. Our results also imply that changing the system attributes does not have a substantial impact on people’s intention, which suggests that the decision to use carsharing are mainly determined by other factors. Furthermore, deploying electric vehicles in carsharing fleet is preferred to fossil-fuel cars by some segments of the population, while it has no negative impact for other segments.
Article
Full-text available
This paper presents a “big-picture view” for policymakers and related stakeholders regarding the future development of car-sharing services. Car-sharing has the potential to significantly disrupt the personal mobility market. Thus, understanding their market penetration and implications is urgently needed. Previous studies in this domain have predominantly focused on the views, opinions, and preferences of consumers. In this study, we complement the current demand modelling research on car-sharing by applying an expert elicitation and aggregation technique that relies on transport experts’ opinions to investigate the role of car-sharing in the future. Specifically, based on the opinions of mobility suppliers, this research elicits experts’ judgment from across government, industry, and academia to gain insights into the future of car-sharing markets in four countries – Australia, Malaysia, Indonesia, and Thailand. The analysis reveals that, from a mobility supplier's perspective, energy and vehicle prices will not have a statistically significant impact on the future adoption of car-sharing. The results also show that the more knowledgeable an expert is, the more pessimistic they are about the market penetration of car-sharing in 2016, and the more optimistic they are about the prevalence of car-sharing in 2030.
Article
Full-text available
The sudden rise of the sharing economy has sparked an intense public debate about its definition, its effects and its future regulation. Here, I attempt to provide analytical guidance by defining the sharing economy as the practice that consumers grant each other temporary access to their under-utilized physical assets. Using this definition, the rise of the sharing economy can be understood as occurring at the intersection of three salient economic trends: peer-to-peer exchange, access over ownership and circular business models. I shortly discuss some of the environmental impacts of online sharing platforms and then articulate three possible futures of the sharing economy: a capitalist future cumulating in monopolistic super-platforms allowing for seamless services, a state-led future that shifts taxation from labour to capital and redistributes the gains of sharing from winners to losers, and a citizen-led future based on cooperatively owned platforms under democratic control. The nature and size of the social and environmental impacts are expected to differ greatly in each of the three scenarios. This article is part of the themed issue ‘Material demand reduction’.
Article
Large penetrations of wind power is prevented in most power systems, mainly because it is highly stochastic. One important aspect of this uncertainty is the large power gradients that wind farms may impose to the grid. To mitigate such undesirable fluctuations, this paper proposes a power smoothing service using the demand response of electric vehicles connected to the adjacent networks. A hierarchical controller is proposed, in which the top layer calculates the ramp rate and sends a request signal to all participant vehicles. In the second layer, a fuzzy controller is developed introducing two fuzzy indices that measure how ready each vehicle is to participate in mitigating large positive and negative fluctuations. These indices are inferred from the state-of-charge and time-to-departure of the vehicle, and are used as the participation factors of the vehicle to provide this service. As much as possible, the proposed controller tries to supply the required service by controlling the vehicles charging load instead of using V2G, which wears out their expensive batteries. Numerical studies on microgrids with high penetrations of wind power corroborate the success of the proposed algorithm in limiting the power fluctuations as well as charging the vehicles in proper time.
Article
The recent strong growth of carsharing (CS) has taken place worldwide mainly in the large cities. We discuss whether there is a potential role for CS in medium to small-sized towns and in less-densely populated rural areas. In large cities CS is provided by large profit oriented firms, capable of offering efficient and flexible services, using the most advanced technologies with highly differentiated and customized prices. In lower density areas and small-to-medium size cities the challenges are much greater. The lower demand resulting from a complex of unfavorable factors (higher car ownership level, larger parking availability, low public transport service, fewer or absent taxi service and so on) requires a different business model for the provision of CS: more socially oriented, with a greater involvement of local municipalities and public transport operators to offer a service, most likely at favorable prices. The paper reports the results drawn from a survey on the potential demand for CS conducted in the Friuli Venezia Giulia region, Italy, a region characterized by small-sized towns and less-densely populated rural areas where CS is not currently offered. It is found that there is a not negligible demand (3.7% of the population holding a driving license) and that the likely users would be the ones that are better informed about CS, environmentally conscious and young. They would be more frequent among students or unemployed people, rather than among professionals, as in the more urban environments. It is also found that CS would mostly be used for non-commuting and longer trips.
Article
This research aims to quantify the effects of car sharing on car ownership, car use and CO2 emissions. The results are based on a survey amongst 363 car sharing respondents in the Netherlands. We found over 30% less car ownership amongst car sharers and they drove 15% to 20% fewer car kilometres than prior to car sharing. The shared cars mostly replace a second or third car. Due to reduced car ownership and car use, car sharers emit between 240 and 390 fewer kilograms of CO2 per person, per year. This is between 13% and 18% of the CO2 emissions related to car ownership and car use.