ArticlePDF Available

Beijing passenger car travel survey: implications for alternative fuel vehicle deployment

Authors:

Abstract and Figures

A survey of Beijing China private passenger car driving behavior was conducted based on global positioning system (GPS) data loggers. The survey focused on the distribution of daily driving distance, number of trips, and parking time. Second-by-second data on vehicle location and speed for 112 private cars were collected. The data covered 2,003 travel days, from June 2012 to March 2013, and nearly 10,000 km for a total of 4,892 trips. The trips covered six major urban and suburban areas in Beijing. The survey results showed average daily driving distances of 31.4, 39.1, and 48 km, and average single trip distances of 13.1, 15.1, and 17.2 km, respectively, on workdays, weekends, andholidays in Beijing urban areas. Average daytime parking times were 5.78, 3.39, and 3.12 h, and average numbers of daily trips were 2.3, 2.6, and 2.8; about 60 % of the vehicles parked last at home, starting from 17:30 to 22:30. These results were used to evaluate electric vehicle (EV) and plug-in hybrid electric vehicle (PHEV) deployment. A vehicle with a 60-km all-electric range (AER) could meet 70 % of daily driving demands. However, EVs with double the AER, such as the Nissan Leaf and Honda Fit, could only increase daily travel by EVs by 20 %. Based on Beijing’s daily driving distance distribution, the estimated average fuel consumptions for the PHEV10 (Toyota Prius) and PHEV40 (Chevrolet Volt) are 2.92 and 1.08 L per 100 km (L/100 km), respectively. These estimates are 20 and 58 % lower, respectively, compared with fuel consumption for the same vehicles used in the USA.
Content may be subject to copyright.
ORIGINAL ARTICLE
Beijing passenger car travel survey: implications
for alternative fuel vehicle deployment
Hewu Wang &Xiaobin Zhang &Lvwei Wu &Cong Hou &
Huiming Gong &Qian Zhang &Minggao Ouyang
Received: 8 April 2014 /Accepted: 12 September 2014
#Springer Science+Business Media Dordrecht 2014
Abstract A survey of Beijing China private passenger car driving behavior was conducted
based on global positioning system (GPS) data loggers. The survey focused on the distribution
of daily driving distance, number of trips, and parking time. Second-by-second data on vehicle
location and speed for 112 private cars were collected. The data covered 2,003 travel days,
from June 2012 to March 2013, and nearly 10,000 km for a total of 4,892 trips. The trips
covered six major urban and suburban areas in Beijing. The survey results showed average
daily driving distances of 31.4, 39.1, and 48 km, and average single trip distances of 13.1,
15.1, and 17.2 km, respectively, on workdays, weekends, andholidays in Beijing urban areas.
Average daytime parking times were 5.78, 3.39, and 3.12 h, and average numbers of daily trips
were 2.3, 2.6, and 2.8; about 60 % of the vehicles parked last at home, starting from 17:30 to
22:30. These results were used to evaluate electric vehicle (EV) and plug-in hybrid electric
vehicle (PHEV) deployment. A vehicle with a 60-km all-electric range (AER) could meet
70 % of daily driving demands. However, EVs with double the AER, such as the Nissan Leaf
and Honda Fit, could only increase daily travel by EVs by 20 %. Based on Beijings daily
Mitig Adapt Strateg Glob Change
DOI 10.1007/s11027-014-9609-9
H. Wang (*):X. Zhang :L. Wu :C. Hou :Q. Zhang :M. Ouyang
State Key Laboratory of Automotive Safety and Energy, China Automotive Energy Research Center,
Tsinghua University, Beijing, China
e-mail: wan ghw@tsinghua.edu.cn
X. Zhang
e-mail: zhangxiaobin08@gmail.com
L. Wu
e-mail: wulvwei@163.com
C. Hou
e-mail: houc09@mails.tsinghua.edu.cn
Q. Zhang
e-mail: cauzq@sina.com
M. Ouyang
e-mail: ouymg@tsinghua.edu.cn
H. Gong
The China Sustainable Energy Program, the Energy Foundation, Rm. 2403 CITIC Bldg., Jianguomenwai,
Dajie No. 19, Beijing 100004, Peoples Republic of China
e-mail: gonghuiming@efchina.org
driving distance distribution, the estimated average fuel consumptions for the PHEV10
(Toyota Prius) and PHEV40 (Chevrolet Volt) are 2.92 and 1.08 L per 100 km (L/100 km),
respectively. These estimates are 20 and 58 % lower, respectively, compared with fuel
consumption for the same vehicles used in the USA.
Keywords Daily driving distance .Private passeng er car .GPS logger.Electric vehicle .Plug-in
hybrid electric vehicle .Charging strategy.All electric range .Fuel saving
1 Introduction
The rapid growth of vehicle populations in China during recent years has resulted in significant
challenges for energy security and environmental pollution, such as massive petroleum imports
and PM
2.5
(particulate matter with an aerodynamic diameter of 2.5 μm or less) pollution in
Beijing and Shanghai (Wang et al. 2011,2013; Hao et al. 2011a,b,c,d,2012; Huo et al.
2007). As a national strategy, the Chinese government has implemented research, develop-
ment, demonstration, and deployment programs for clean vehicle technologies for 20 years to
build a sustainable transportation system (Wang and Ouyang 2007;Gongetal.2013). The
national strategy highlights new energy vehicles (NEVs), including electric vehicles (EVs),
fuel cell vehicles (FCVs), and hybrid electric vehicles (HEVs), with more than 1,000 FCVs
deployed in 2011 (Dixon et al. 2011) and 27,000 NEVs in 2012 (MOST 2013).
Facing the challenges of global energy shortage and climate change, automotive powertrain
electrification has been widely regarded as a feasible solution to future transportation charac-
terized by low emission and high energy efficiency (Greene et al. 2010). Life cycle analysis
(LCA) on the cost of manufacturing, owning, and operating passenger vehicles showed that
the adoption of HEVs and plug-in hybrid vehicles (PHEVs) with a smaller size battery offers
more societal benefits in terms of reducing air emissions and oil dependency per dollar spent
(Michalek et al. 2011). However, because of the limitation of battery performance, EVs and
PHEVs with smaller batteries have a shorter driving range and require a longer recharging time
than conventional vehicles. For such range-related vehicles, driving patterns, that is daily
driving distance, trip frequency, starting and parking time, and fueling features, become
important issues for vehicle design (Vilimek et al. 2012;Al-Alawietal.2013;Duetal.
2012a,b). Such patterns also raise issues regarding energy consumption (SAE 1999,2009;
Bradley and Quinn 2010; Gonder et al. 2007;Neubaueretal.2013), environmental impacts
(Karabasoglu et al. 2013; Elgowainy et al. 2010,2012), and infrastructure construction (Dong
et al. 2014).
The research methods for daily driving patterns included a questionnaire survey and
onboard instrument records (usually global positioning system (GPS) data loggers)two
methods used throughout the world. In the United States (U.S.), the National Household
Travel Survey (NHTS) provides comprehensive national data on travel and transportation
patterns for transportation researchers (NHTS 2014). GPS devices have also been adopted in
regional travel surveys on travel behavior characteristics, such as the data on 407 vehicles in
the greater Chicago area in Illinois, U.S., in 2007, and data on 1,325 vehicles in Atlanta,
Georgia in 2011 (NREL 2014). In Europe, the Danish National Travel Survey (DNTS 2014)
and GPS-based data that track vehicles have been adopted to extract driving distances and
driving time periods. The NHTS and DHTS can provide average travel distance per day, such
as 29.1 miles in the USA and 18.3 miles in Denmark. However, the information collected by
onboard GPSs is more suitable for evaluating EV driving requirements/availability (Wu et al.
2010) and PHEV energy analysis/charging infrastructure (Lin et al. 2012).
Mitig Adapt Strateg Glob Change
Compared with driving pattern studies in other countries, many studies in China focused
primarily on the total annual driving mileage estimation using the questionnaire survey method
(Hao et al. 2011d;Huo2005; Huo et al. 2012;Wangetal.2002,2006). A few studies focused
on Chinese daily driving patterns, especially driving range distribution. In the years of 2012
and 2013, a questionnaire survey on passenger car drivers in Beijing was conducted focusing
on the daily driving range distribution (Hou et al. 2012,2013). To obtain more detailed and
accurate data on Chinese driving patterns, a similar effort using onboard instruments recording
the data second by second should be done as in the U.S. and other countries in some
representative regions.
This study is aimed to reveal the real world using features of passenger cars in a specific
region, Beijing, China. The statistical results of the Beijing driving pattern survey were firstly
presented. Based on the driving patterns, the mileage applicability of battery electric vehicles
(BEVs) and energy consumptions of PHEVs were then studied..
2 Data collection and sample characteristics
2.1 Selection and analysis of samples
GPS loggers were adopted to collect the travel data, including the position and
vehicle speed second by second. In our survey, 112 voluntary vehicles were recruited
from June 2012 to March 2013, and each vehicle was observed for continuing days.
Therefore, nearly 100,000 km of travel distance in 2,003 travel days for about 4,892
trips were recorded.
Beijing is the transportation center in northern China, and the road system mainly
consists of 4 ring roads and 12 national highways, which is shown in Fig. 1.In
Beijing, regions within the 5th ring road (S50 in Fig. 1) are regarded as urban areas.
There are currently over five million vehicles running in Beijing. In order to guaran-
tee the representativeness and randomness of samples, two sampling methods, strati-
fied sampling as well as random sampling, were conducted to recruit voluntary cars.
The distribution of voluntary cars is also shown in Fig. 1. For stratified sampling, we
selected one organization in each of the six main administrative regions/districts of
Beijing, which were basically evenly distributed in the ring (see Table 1); then,
voluntary vehicles from these six organizations were randomly selected. For random
sampling, voluntary vehicles in schools and communities were randomly selected,
which covered Beijings major urban areas and roads.
The survey of travel origin and destination (OD) points was adopted based on the GPS-
based survey platform. Vehicle owners included corporate employees, civil servants, university
teachers, researchers, IT engineers, and freelancers. For the distribution of travel days and trips,
non-holiday (holidays refer to non-working periods for more than 3 days, including the
National Day holiday and Spring Festival holiday) travel days accounted for 84.6 % of the
total days, and non-holiday trips accounted for 82.4 % of the total trips. Travel days on
workdays accounted for 72.3 % of the total non-holiday days. This proportion is very close to
71.4 %, which is the proportion of workdays per week, while the number of trips accounted for
69.5 % of non-holiday trips.
Statistical analysis of the OD points of the urban samples was conducted. The distribution
of OD points within the 2nd, 3rd, 4th, and 5th ring roads of Beijing (these four ring roads were
regarded as approximately rectangular loops in order to locate OD points by latitude and
longitude, which can be seen in Fig. 1) was analyzed; in addition, the proportion of the driving
Mitig Adapt Strateg Glob Change
distance within the four ring roads in total driving distance, respectively, is also shown in
Table 2.Table2shows that the proportion of travel mileage within each ring road was
generally lower than the OD point distribution, which indicates that vehicles traveling out of
the 5th ring road usually traveled longer distances with fewer OD points. The distribution of
OD points within the five ring roads reveals the travel starting and ending features of Beijing
passenger vehicles, which is essential for siting charging infrastructures for BEVs and PHEVs.
However, the relevant study has yet to be conducted.
Fig. 1 Schematic diagram and map of the four ring roads and sampling organizations in Beijing
Tab le 1 The voluntary cars are chosen according to the organizations and working locations in Beijing, China
Volunteer organization Districts Working location
Government agency Dongcheng Inner 2nd ring
State-owned business Xicheng Between 2nd and 3rd rings
High-tech enterprise Chaoyang Between 4th and 5th rings
Internet company Haidian Between 3rd and 4th rings
Industry association Fengtai Between 2nd and 3rd rings
Research institute Shijingshan Outside 5th rings
The locations are shown as blue spots in Fig. 1, which lie between two ring roads of the Beijing road system
Mitig Adapt Strateg Glob Change
2.2 General findings of total samples
Six main indicators were derived to reflect the general characteristics of the total samples, as
shown in Table 3. First, the aggregate average daily driving distance of the total samples was
about 46 km. Specifically, it decreased to 37.2 km on non-holidays but increased significantly
on holidays. Second, the average driving distance was about 19 km. It was 15.6 km on non-
holidays but doubled on holidays. Third, the aggregate average daily travel time was about
100 min. Fourth, the aggregate average travel time per trip was about 40 min, and it varied
little between different travel days. Fifth, the aggregate average daytime parking time was
about 4.7 h, but on holidays, it decreased to 3 h. And, sixth, the average daily travel frequency
for all days was less than 2.5 times, and it increased slightly on holidays and weekends. These
factors indicate that people tend to travel much longer trips and daily distance and park less on
holidays than on non-holidays. However, the daily travel frequency and travel time per trip
vary less between different travel days.
2.3 Findings of the urban samples
Table 4presents statistics for the urban samples. The aggregate average daily driving distance
was about 35 km, and it was 37 % longer on holidays. The aggregate average travel mileage
per trip was about 15 km and kept stable between different travel days. The aggregate average
daily travel time and travel time per trip were about 95 and 37 min respectively, both of which
vary little between different travel days. The aggregate daily parking time was about 4.8 h, and
it decreased to 3.39 and 3.12 h on weekends and holidays. The aggregate average daily travel
frequency was less than 2.5 times. Compared with the total samples, urban samples show
similar daily travel time, trip time, daily parking time, and travel frequency, but less variations
in trip distance and daily distance between different travel days.
Comparisons between statistical results for urban and total samples indicated that vehicles
beyond urban areas tend to travel longer and faster both within a day and within a trip. The
aggregate daily driving distance of total samples was 30.5 % longer than that of urban samples,
Tab le 2 Distributions of travel OD points and distances of the Beijing urban samples are shown
Inner 2nd ring Inner 3rd ring Inner 4th ring Inner 5th ring
Origination (O) point 13.2 % 33.9 % 51.9 % 76.5 %
Destination (D) point 12.6 % 33.3 % 50.9 % 77.4 %
Both O and D points 12.9 % 33.6 % 51.4 % 76.9 %
Travel mileage 9.8 % 26.8 % 46.7 % 69.6 %
Tab le 3 The general characteristics of the total samples are reflected by six indicators used in this study
Workdays Weekends Non-holidays Holidays All days
Average daily distance (km) 35.4 41.7 37.2 96.0 46.2
Average distance per trip (km) 15.5 16.0 15.6 34.5 18.9
Average daily travel time (h) 1.51 1.92 1.62 2.31 1.73
Average travel time per trip (h) 0.63 0.69 0.65 0.77 0.67
Average daytime parking time (h) 5.70 3.35 5.01 2.95 4.68
Average daily trips (times) 2.29 2.61 2.38 2.78 2.44
Mitig Adapt Strateg Glob Change
while the daily travel time only increased by 9.5 %. The aggregate trip distance and time
showed similar trends. A significant difference emerged in the results on holidays. The daily
and trip driving distances of total samples were double those of the urban samples. This
indicates that urban residents usually visit the suburbs or travel to satellite counties on
holidays.
3 Analysis of travel mileage and characteristics
3.1 Distribution of driving distance
3.1.1 Distribution of daily travel mileage
Figure 2shows the cumulative percentages of daily driving distance on workdays, weekends,
non-holidays (workdays and weekends integrated), and holidays, illustrated by curves. Dis-
tributed percentages of non-holidays are illustrated as bars. The non-holiday distributed
percentages reveal that the vast majority of non-holiday driving distances were less than
100 km and evenly distributed below 25 km, accounting for about 6 % per 3 km.
As can be seen from the cumulative percentages, the non-holiday curve was steeper than
the holiday curve, and the weekday curve was steeper than the weekendscurve,which
Tab le 4 The general characteristics of the urban samples are reflected by six indicators used in this study
Workdays Weekends Non-holidays Holidays Aggregate
Average daily distance (km) 31.4 39.1 33.5 48.0 35.4
Average distance per trip (km) 13.7 15.1 14.1 17.2 14.6
Average daily travel time (h) 1.44 1.88 1.56 1.70 1.58
Average travel time per trip (h) 0.60 0.68 0.62 0.56 0.61
Average daytime parking time (h) 5.78 3.39 5.07 3.12 4.82
Average daily trips 2.29 2.60 2.37 2.79 2.43
Fig. 2 The cumulative and distributed percentages of daily travel mileage for all samples are illustrated in curves
and bars
Mitig Adapt Strateg Glob Change
indicate that driving distances on non-holidays were shorter than on holidays, and on week-
days were shorter than weekends. Specifically, mileages of 11.5, 22.3, and 49.2 km can cover
25, 50, and 80 % of travel demand on weekdays. On weekends, coverage rates of 25, 50, and
80 % of travel mileage demand per day were 14.1, 30.2, and 63.3 km, respectively. Mileage of
105 km had the same 95 % coverage rate on both workdays and weekends. The cumulative
percentage on holidays was closer to the case of the weekends. The coverage of 90 % demand
was 90 km on both holidays and weekends, but there was a significant difference after 90 km.
This indicates that long-distance travel above 90 km was distributed more on holidays; thus, a
longer driving distance is needed to get the same coverage ratio.
3.1.2 Distribution of single trip driving distance
Figure 3shows the cumulative percentages of single trip driving distance as curves, while the
distributed percentages are illustrated as bars. The distributed percentages revealed that the
single minimum mileage significantly concentrated in the 0- to 6-km range, accounting for
nearly half the proportion. The cumulative percentages show that mileages of 8.8, 22.1, and
45.3 km could cover 50, 80, and 95 % of the non-holiday single travel demand. When taking
just the maximum trip of a single day into consideration, the coverage rates of 50, 80, and
95 % of single trip demand were 14.8, 31.6, and 60.2 km, respectively, which are 68, 43, and
33 % higher than the average situation. There was not a large difference between holidays and
non-holidays when the cumulative percentage of a single trip was within 30 km. The gap
gradually emerged after more than 30 km, but the two cumulative percentages simultaneously
tended to 100 % after more than 100 km. It indicates that compared with that on non-holidays,
the distribution of single trip mileage on holidays was similar within 30 km and distributed
widely after more than 30 km.
The average driving distances of the first and last trips on non-holidays were 16.8 and
15.9 km, respectively, slightly more than the average driving distance per trip, which was
14.1 km. The average driving distances of the first and last trips on workdays were 15.9 and
15.1 km, respectively, also slightly more than the average driving distance per trip, which was
13.7 km. The average driving distances of the first and last trips on weekends were 19.4 and
17.8 km, respectively, also slightly more than the average driving distance per trip, which was
Fig. 3 The cumulative and distributed percentages of single trip driving distance for the urban samples are
illustrated in curves and bars
Mitig Adapt Strateg Glob Change
15.1 km. The results above indicate that the first and the last trips were often longer than other
trips in a single day, and the mileage was usually the distance between home and the
workplace.
3.2 Distribution of travel duration and trip starting and ending time
3.2.1 Distribution of daily travel duration and single trip duration
The distribution of daily travel duration as well as single trip duration was studied with an
interval length of 20 min; Fig. 4shows the results. With respect to daily travel duration, the
most probable interval on workdays was 40 to 60 min, accounting for about 17 %. The
probability that a travel duration was within 60 min was 41.4 %, and within 120 min was
41.4 %. The most probable interval on weekends was 80 to 100 min, accounting for about
14 %. The probability that a travel duration was within 60 min was 30.6 %, and within
120 min was 63.8 %. This result shows that the travel duration was usually within 2 h on
workdays but was more dispersed on weekends, with 17.1 % of the durations longer than 3 h.
With regard to travel duration per trip, the most probable interval for both workdays and
weekends was 0 to 20 min. The former probability was about 36 % and the latter was about
38 %. The probability of travel duration within 60 min on workdays and weekends was 85 and
79 %, respectively, which indicates that the probability of the travel duration of more than 1 h
per trip on weekends was more than that on workdays. In other words, vehicles traveled longer
on weekends than on workdays.
3.2.2 Distribution of trip starting and ending time
Figure 5shows the distributions of O (starting) and D (ending) time on workdays and
weekends. There were clear peaks with travel OD time. About 15 % of travel originated at
7:00 to 8:00 and ended at 8:00 to 9:00, and approximately 12 % of travel originated at 17:00 to
18:00 and ended at 18:00 to 19:00. Travel OD time on weekends was more decentralized than
Fig. 4 The distribution of daily and single trip travel durations for the urban samples is illustrated in bars
Mitig Adapt Strateg Glob Change
that on workdays. In the weekends, the OD time was distributed between 8:00 and 22:00 and
relatively concentrated between 11:00 and 16:00.
3.2.3 Distribution of last parking time
The distribution of last parking time on non-holidays was studied based on all the vehicle
samples, as shown in Fig. 6. Obviously, the highest peak of vehicle parking time emerged
between 18:30 and 19:00, accounting for 8.8 %. The second peak appeared between 20:30 and
21:00, accounting for 8.1 %; 40.3 % of the vehicles parked from 18:00 to 21:00. The
proportion of parking from 12:00 to 1:00 of the next day was more than 85 %. Another small
peak emerged from 8:00 to 9:30, accounting for 6.8 %.
3.3 Distribution of travel frequency and parking duration time
3.3.1 Distribution of travel frequency
Figure 7shows the daily travel frequency in trips for all the vehicle samples in urban
areas. Statistics show that vehicles in Beijing traveled once on more than 1/4 of
workdays; 40 % of the respondents traveled two times, and 17 % traveled three times
per day. In total, the proportion of travel within three times per day was 85.5 %. On
weekends, the proportion was 22, 34, and 22 % for travel frequencies of one, two,
and three times, respectively. On weekends, the proportion for travel frequency of
more than three times was 21.5 %.
For all the non-holidays, the proportion of vehicles traveling one, two, and three times per
day was 26, 39, and 19 %, respectively; vehicles traveling more than four times per day
accounted for less than 8 %. Compared with those on non-holidays, travel times on holidays
tended to be more and the distribution was more scattered. The proportion of vehicles traveling
one, two, three, and four times per day was 22, 27, 21, and 18 %, respectively. The probability
of more than seven times per day was about 3 %.
Fig. 5 The origin and destination (OD) times for the urban Beijing samples on non-holidays are illustrated in curves
Mitig Adapt Strateg Glob Change
3.3.2 Distribution of maximum parking duration time
In order to study charging opportunities between trips in the daytime, the maximum parking
durations were discovered based on the distribution of trips. Then trips before the maximum
parking duration were summed to the new first trip of the day, and trips after the duration were
summed to the second trip of the day. In this way, the original daily trips can be regarded as the
first trip before the maximum parking duration and the second trip back home. Table 5shows
the mean value of maximum parking duration on different days, and the mean distance of the
first and second trips.
The mean of maximum parking duration on holidays and weekdays was more than 6 h, and
the mean on holidays and weekends was more than 4 h. The mean values of the first and
second trip driving distances were nearly equal, which indicates that the maximum parking
duration was basically located at the midpoint of daily driving distance. Therefore, it is
reasonable to charge a second time at the maximum parking duration. On the other hand,
the maximum parking duration is enough to basically get the EV battery fully charged. Thus,
Fig. 6 The last parking time for the Beijing urban samples on non-holidays is illustrated in bars
Fig. 7 The distributed percentage of daily travel frequency for the Beijing urban samples is illustrated in bars
Mitig Adapt Strateg Glob Change
similar to the first charge at home, assuming that vehicles can be fully charged at the second
charge in the daytime, the first and second mileages (non-zero) were both regarded as full
power driving distance.
4 Mileage applicability of EVs based on driving distance distribution
Two basic assumptions can be made when analyzing the mileage applicability of EVs. First,
people will transfer their daily driving habits from traditional internal combustion engine
vehicles to EVs, which means that driving patterns will remain the same. Second, EVs start to
charge after the last trip every day and can be fully charged before they travel the next day.
Here, EVs mainly refer to pure BEVs and PHEVs.
4.1 Satisfaction and utilization of EVs
To study the satisfaction and utilization of EVs with different AERs, the distribution of daily
driving distance was fitted using gamma distribution based on actual survey data. Figure 8
shows the cumulative probability of daily driving distance on holidays, non-holidays, work-
days, and weekends.
BEVs with an AER of 100 km could meet 96 % of the daily travel demand (96 % fill rate)
on non-holidays, which could cover almost all of the daily travel demand; on holidays, the
travel fill rate was close to 90 %. On workdays and weekends, the fill rate showed little
Tab le 5 Under the scenario of charging twice a day, the mean of maximum parking durations and driving
distances of urban samples are listed
Non-holidays Holidays Workdays Weekends
Mean of maximum parking duration (h) 6.2 4.1 6.9 4.3
Standard deviation of maximum parking duration (h) 3.7 3.2 3.6 3.3
Mean driving distance of 1st trip (km) 19.4 26.6 17.9 23.3
Mean driving distance of 2nd trip (km) 19.2 28.7 18.7 20.4
Fig. 8 The cumulative probability of daily driving distance for the urban samples is illustrated in curves
Mitig Adapt Strateg Glob Change
difference, at 95 % or so, which could basically cover the majority of the daily travel demand.
If the AER was reduced to 50 km, the fill rate on non-holidays was 78 %; on holidays, it was
only 65 %. On workdays and weekends, the fill rate was 80 and 72 %, respectively. When the
AER was halved, the fill rate was not halved, which indicates that the more the AER increases,
the more difficult it is to improve the fill rate.
Several typical BEVs with different AERs were analyzed based on the derived distribution
of daily driving distance. Table 6shows their fill rates on non-holidays and holidays in Beijing.
The AER of the Nissan Leaf and Honda Fit EV was more than 120 km; thus, their fill rate was
around 98 % on non-holidays and more than 90 % on holidays. The fill rate of the Scion iQ
EV with a 60-km AER on non-holidays was close to 85 %, while only about 70 % on holidays.
The utilization rate is defined as the percentage of daily driving distance during which the
battery is fully utilized. For PHEVs, when the charge depleting (CD) range was set to be 32 km
(PHEV20), 40 % of daily driving distances exceeded the CD range on non-holidays, which
means that 40 % of daily travel fully utilized the battery; so the utilization rate was 40 %. The
utilization rate was 37 and 47 %, respectively, on workdays and weekends. On holidays, the
utilization rate of the AER was more than 50 %. When the AER was set to be 16 km
(PHEV10), the full utilization rate exceeded 65 % on non-holidays. The utilization rate was
63 % and nearly 10 % higher, respectively, on workdays and weekends. While on holidays, the
utilization rate reached 72 %.
The Toyota Prius 2013 and Chevrolet Volt 2013 were investigated and compared in terms
of utilization rate. The former is a PHEV10 product, whose full utilization rate of AER is
around 60 to 70 %. While the Volts AER is set up to be 60 km, its utilization rate is only 15 to
30 %.
4.2 Mileage applicability under different charging scenarios
The applicability of EVs under different charging scenarios was studied. Two charging
patterns, once a day and twice a day, were considered. Charging once a day means only
charging at home before the first trip every day. Charging twice a day means charging at home
and at the workplace, which can be regarded as charging before the first trip every day as well
as at the longest interval between two trips. Figure 9shows the distribution of driving distance
considering charging once a day and twice a day on non-holidays. In the case of charging
twice a day, the full power driving distance significantly concentrated toward a short distance.
The proportion reached 48 % within 12 km, while the proportion of charging once a day was
only 25 %. According to the fitted gamma distribution of cumulative probability (red line in
Fig. 8), under the case of charging twice a day, when the AER of the BEV was set to be
100 km, its mileage fill rate was 99.5 %. When the AER was set to be 50 km, the mileage fill
rate was still as high as 93.0 %, and when the AER was 30 km, it still can cover close to 80 %
of the travel demand. Compared with the case of charging once a day (blue line in Fig. 8),
charging twice a day has more advantages on the mileage fill rate at the same AER. When the
Tab le 6 The fill rates of travel demand of different EVs in Beijing under the conditions of this study
Manufacturer Model Type AER
(km)
Non-holiday fill rate Holiday fill rate
Nissan Leaf 2013 BEV 120 97.9 % 92.0 %
Honda Fit EV 2013 BEV 131 98.6 % 93.7 %
Scion iQ EV 2013 BEV 60 84.2 % 71.4 %
Mitig Adapt Strateg Glob Change
AER increases, the advantages increase first and then decrease, with the maximum occurring
at 23 km. When the AER is 23 km, the absolute difference between the two fill rates reaches
the maximum. The fill rate of charging twice a day is 69.4 %, while that of charging once a day
is 47.0 %, which indicates that when the AER of the BEV is set to be 23 km, charging twice a
day could bring the largest increase to mileage fill rate. When the AER surpasses 65 km, the
advantage of charging twice a day over once a day is 10 %; when the AER surpasses 90 km,
the advantage is less than 5 %.
The cumulative probabilities of charging once a day and twice a day on holidays,
workdays, and weekends, respectively, were compared in the same way. The results showed
that charging twice a day is better than once in various types of travel days, and the variation of
the advantages is similar to the situation of non-holidays. Table 7shows the AERs according to
the mileage fill rate of 95, 80, and 50 % in the case of charging once a day and twice a day,
respectively.
5 Evaluation of energy consumption of PHEVs based on driving distance distribution
5.1 UF and PHEV energy consumption in Beijing
To estimate the average energy consumption of a PHEV in a given driving pattern, the Utility
Factor (UF) was introduced as an important index, as standardized in SAE J1711 (SAE 1999)
and SAE J2841 standards (SAE 2009). The UF of Beijing was derived and compared with that
of the USA, then the energy consumption of typical PHEVs is presented.
5.1.1 UF of Beijing and comparison with that of the USA
The Beijing UF of different types of travel days is shown in Fig. 10, which was derived based
on driving distance distribution on holidays, workdays, weekends, and holidays. Since the
distribution of daily travel distance indicates that most of the daily travel distances are within
100 km, the CD range of 300 km could cover nearly all the daily travel, and UF at 300 km is
near to 1.
Fig. 9 The distributions of Beijing driving distances to be met by EVs in different charging scenarios on non-
holidays are compared
Mitig Adapt Strateg Glob Change
The curve was fitted in order to achieve continuous UF values and to facilitate the
calculation of the average fuel and electricity consumption of PHEVs. The UF of Beijing
was expressed as a function and compared with that of the USA (Table 8). Note that the unit of
R
CD
in China is kilometers, and the range of R
CD
is between 0 and 300 km. The unit of R
CD
in
the USA is miles.
UF RCD
ðÞ¼1exp a1RCD=400ðÞþa2RCD=400ðÞ
2þa3RCD=400ðÞ
3þa4RCD=400ðÞ
4þa5RCD=400ðÞ
5þa6RCD=400ðÞ
6
h in o
Based on the above fitted function, Fig. 11 shows the comparison of the two UF curves.
The dotted red line in the figure represents the raw data of the Beijing UF, and the solid red line
and the blue line are the fitted curves of the Beijing UF and the US UF, respectively. The fitted
and original values of the Beijing UF coincide very well. The Beijing UF curve is always
above the curve of the US UF, which indicates that the Beijing UF is higher than the US UF at
the same charge depleting (CD) mileage, namely that compared with the USA. The CD stage
of PHEVs in Beijing occupies a greater proportion of mileage. The maximum difference
between the two UF curves emerged at 47.6 km, where the Beijing UF was 0.24 more than the
US UF. The comparison indicates when the CD range is set to be 47.6 km. A PHEVin Beijing
has the most advantages in fuel economy over that in the USA. If the UF is 0.5, the CD range
Tab le 7 Under different charging scenarios and mileage fill rates, the AERs are compared for this Beijing study
Mileage fill rate AER (km)
Non-holidays Holidays Workdays Weekends
Charging
once a
day
Charging
twice a
day
Charging
once a
day
Charging
twice a
day
Charging
once a
day
Charging
twice a
day
Charging
once a
day
Charging
twice a
day
95 % 94.2 56.3 142.2 82.2 88.2 52.4 108.4 66.0
80 % 53.2 31.0 77.1 43.6 49.8 29.1 61.7 35.4
50 % 25.0 13.8 33.6 18.5 23.2 13.3 29.3 15.2
Fig. 10 The UF of Beijing based on actual mileage distribution under the conditions of this study
Mitig Adapt Strateg Glob Change
of a PHEV in Beijing is 22.8 km, while in the US, the value is 45 km, nearly double that of
Beijing.
5.1.2 Energy consumption of typical PHEVs
Several typical PHEV modelsToyota Prius 2013, Chevrolet Volt 2013, Fisker Karma
2012were analyzed. The average fuel and electricity consumption were calculated using
the formula (SAE 2009), in which FC
CD
and FC
CS
are the fuel consumptions at CD and
charge sustaining (CS) stage, while EC
CD
and EC
CS
are the electricity consumptions at CD and
CS stage. These energy consumption results directly refer to the EPA fuel economy website.
While it is acknowledged that the driving cycles of Beijing and the USA differ, examination of
that distinction was not within the scope of this study. The results are shown in Table 9.
Compared with that in the USA, the average fuel consumption in Beijing was generally
lower. The Prius in Beijing reduced fuel consumption by 20 %, and the Volt and Karma by
50 %. Accordingly, the average electricity consumption in Beijing was higher. The Prius in
Beijing increased electricity consumption by 70 % approximately, while the Volt and Karma
increased consumption by 40 %. Obviously, PHEVs operating in Beijing have greater fuel
economy potentials than in the USA.
FC ¼UF RCD
ðÞFCCD þ1UF RCD
ðÞ½FCCS
EC ¼UF RCD
ðÞECCD þ1UF RCD
ðÞ½ECCS
Tab le 8 Parameters of fitted functions of Beijing UF and US UF are specified under the conditions of this study
a
1
a
2
a
3
a
4
a
5
a
6
Beijing UF 12.3168 2.4786 19.0467 38.1964 31.0399 8.8085
US UF 10.52 7.282 26.37 79.08 77.36 26.07
Source: SAE (2009)
Fig. 11 The UFs of Beijing, China and U.S. are compared
Mitig Adapt Strateg Glob Change
5.2 Energy consumption under different charging scenarios
With the deployment of more charging infrastructures in Beijing, EVs can get charged both at
home and at the workplace. The fuel and electricity consumption of PHEVs under different
charging scenarios was studied to evaluate the impact of charging patterns on energy
consumption.
5.2.1 Beijing UF under repeated charging scenario
If PHEVs get charged twice a day, the influencing factor of energy consumption will be the full-EV
driving distance, not the daily driving distance. That is, the full-EV driving distance will change if a
PHEV gets charged once a day or twice a day, and the corresponding UF will also change.
The original curve and fitted curve of the UF in the case of charging twice a day and the fitted
curveoftheUFinthecaseofchargingonceaday (i.e., Beijing UF) were compared; Fig. 12 shows
the results. The red line above the blue one in the figure indicates that at the same CD range, the UF
of charging twice a day was significantly higher than that of charging once a day, with a maximum
difference of 0.1547 at 24.7 km. In other words, for the same PHEV, the average fuel consumption
was lower charging twice a day than charging once a day. When the PHEVs CD range was set to
24.7 km, the advantage of charging twice a day reached the maximum. When the UF is 0.5, namely
the CD mileage and the CS mileage each accounts for half of the daily driving range, the PHEVs
CD range should be 14.6 and 22.8 km, respectively, in the case of charging twice a day and once a
day. The results indicate that to achieve the same average fuel consumption, increasing the charging
frequency helps to reduce the PHEV battery size.
5.2.2 Fuel efficiency under repeated charging scenarios
To compare the improvement of fuel efficiency brought by charging twice a day, the three
PHEV modelsToyota Prius 2013, Chevrolet Volt 2013, and Fisker Karma 2012were
investigated regarding average energy consumption. Table 10 gives the results. In the case of
charging twice a day, the average fuel consumption was reduced significantly. The Prius in
Beijing decreased fuel consumption by about 20 %, and the Volt and Karma reduced
consumption by more than half. However, the average electricity consumption increased
significantly. The Prius increased by more than 30 %, and the Volt and Karma increased by
over 10 %.
Tab le 9 The average energy consumptions of different PHEV models between Beijing, China and the U.S. are
compared
Toy ota
Prius 2013
Chevrolet
Vo l t 2 0 1 3
Fisker
Karma 2012
UF FC EC UF FC EC UF FC EC
Beijing 0.42 2.92 7.56 0.83 1.08 20.53 0.79 2.47 30.42
USA 0.25 3.64 4.50 0.60 2.55 14.84 0.55 5.29 21.18
Relative difference (%) 68.0 19.8 68.0 38.3 57.6 38.3 43.6 53.3 43.6
FC = L/100 km; EC = kW·h/100 km
Mitig Adapt Strateg Glob Change
6Conclusions
A Beijing passenger car travel survey platform was established and adopted in this study. The
goal was to produce the general features and distribution of daily driving patterns. The mileage
applicability and energy consumption of electric vehicles (BEVs and PEHVs) with different
AERs were studied. Analysis of collected data from 112 voluntary cars for 2,003 travel days in
2012 to 2013 revealed several findings on Beijing vehicle usage.
The GPS-based research on driving patterns of Beijing passenger cars indicated that daily
driving distance on weekdays, weekends, and holidays was 31.4, 39.1, and 48 km, respec-
tively. Driving time and parking time were 1.44 and 5.78 h, respectively, on weekdays, while
on weekends, they were 1.88 and 3.39 h, respectively. There were 2.29 trips per day on
weekdays and 2.60 trips on weekends. The distribution of driving distance revealed that 80 %
of the daily driving distances on non-holidays were within 52.5 km, and 80 % of the single trip
distances were within 22.1 km. The distribution of OD time shows clear double peaks with
travel OD time on weekdays. The distribution of travel frequency showed that the proportion
of travel within three trips per day was 85.5 % on weekdays, while on weekends, it was 78 %.
Based on the analysis of driving patterns, the mileage applicability of BEVs and the energy
consumption of PHEVs under different charging scenarios were studied. The results showed
that for BEVs, a 60-km AER can meet 70 % of daily travel demand. However, EVs with
Fig. 12 The UF curves for the Beijing urban samples are shown under different charging scenarios
Table 10 The average energy consumptions of different PHEV models in various Beijing scenarios are
compared
Toyota Prius 2013 Chevrolet Volt 2013 Fisker Karma 2012
UF FC EC UF FC EC UF FC EC
Charge twice a day 0.57 2.31 10.18 0.92 0.51 22.76 0.90 1.20 34.57
Charge once a day 0.42 2.92 7.56 0.83 1.08 20.53 0.79 2.47 30.42
Relative difference (%) 35.7 20.9 34.7 10.8 52.8 10.9 13.9 51.4 13.6
FC = L/100 km; EC = kW·h/100 km
Mitig Adapt Strateg Glob Change
double the AER, such as the Nissan Leaf and Honda Fit EV, can only increase daily travel by
20 %. Considering charging patterns, the 50-km AER can meet 78 % of travel demand if
charged once a day but increases to 93 % if charged twice a day. The Beijing UF of different
charging patterns was derived. The average fuel consumption of PHEVs in Beijing is generally
lower compared with that in the USA. The estimated fuel consumptions of the PHEV10
(Toyota Prius) and PHEV40 (Chevrolet Volt) are 2.92 and 1.08 L/100 km, respectively, which
are 20 and 58 % lower compared with those in the USA.
The general features and distribution of driving patterns in Beijing help to understand the
implications of the deployment of electric vehicles, including BEVs and PHEVs. Due to the
driving patterns in Beijing, BEVs with a moderate AER (e.g., 100 km) can meet most of the
travel demand, and the average fuel consumption of PHEVs in Beijing is generally lower
compared with that in the USA. On the other hand, repeated charging can increase the utility of
electric vehicles. The comparative advantages of electric vehicles in Beijing justify the
deployment and infrastructure construction, which can help reduce fuel consumption and
GHG emissions from transportation sector.
Acknowledgments This work was supported by the Ministry of Science and Technology of China under
contract Nos. 2010DFA72760, 2011DFA60650, 2012DFA81190, and 2011BAG02B12, 2013BAG06B02, and
by the Energy Foundation China under contract No. G-1204-15951. The authors wish to thank Dr. Michael Wang
of Argonne National Laboratory for his help and guidance. Special thanks go to the volunteer vehicle owners
whose names are not listed here due to privacy. Finally, we thank Mr. Xihao Li and Ms. Qingxiu Meng for
contacting the volunteers.
References
Al-alawi Baha M, Bradley TH (2013) Total cost of ownership, payback, and consumer preference modeling of
plug-in hybrid electric vehicles. Appl Energy 103:488506, Beijing Municipal Commission of Transport,
Beijing Transportation Research Center
Bradley TH, Quinn CW (2010) Analysis of plug-in hybrid electric vehicle utility factors. J Power Sources
195(16):53995408
Dixon RK, Wang X, Wang MQ et al (2011) Development and demonstration of fuel cell vehicles and supporting
infrastructure in China. Mitig Adapt Strateg Glob Chang 16(7):775789
DNTS (Danish National Travel Survey) (2014) http://www.modelcenter.transport.dtu.dk/english/TU/Latest_
dataset. Cited 18 February 2014.
Dong J, Liu C, Lin Z (2014) Charging infrastructure planning for promoting battery electric vehicles: an activity-
based approach using multiday travel data. Transp Res Part C: Emerg Technol 38:4455
Du J, Wang H, Ouyang M (2012a) Parameters optimization of PHEV based on cost-effectiveness from life cycle
view in China. In: Proceedings of the FISITA 2012 world automotive congress, pp 697704. FISITA 2012
World Automotive Congress, Beijing, China, 2012
Du J, Ouyang M, Wang H (2012b) Battery electric vehicle parameters design targeting to cost-benefit objective.
In: Proceedings of vehicle power and propulsion conference (VPPC), 2012 IEEE, pp 11601164. 2012 I.E.
Vehicle Power and Propulsion Conference, Seoul, Korea, 2012
Elgowainy A, Han J, Poch L, Wang M, Vyas A, Mahalik M, Rousseau A (2010) Well-to-wheels analysis of
energy use and greenhouse gas emissions of plug-in hybrid electric vehicles, ANL/ESD/10-1. Argonne
National Laboratory, Argonne
Elgowainy A, Zhou Y, Vyas AD et al (2012) Impacts of charging choices for plug-in hybrid electric vehicles in
2030 scenario. Transp Res Rec: J Transp Res Board 2287(1):917
Gonder J, Markel T, Thornton M et al (2007) Using global positioning system travel data to assess real-world
energy use of plug-in hybrid electric vehicles. Transp Res Rec: J Transp Res Board 2017(1):2632
Gong HM, Wang M, Wang HW (2013) New energy vehicles in China: policies, demonstration and progress.
Mitig Adapt Strateg Glob Chang 18(2):207228
Greene DL, Boudreaux PR, Dean DJ et al (2010) The importance of advancing technology to Americasenergy
goals. Energ Policy 38(8):38863890
Mitig Adapt Strateg Glob Change
Hao H, Wang HW, Yi R (2011a) Hybrid modeling of Chinas vehicle ownership and projection through 2050.
Energy 36(2):13511361
Hao H, Wang HW, Ouyang MG (2011b) Predictions of Chinas passenger vehicle and commercial vehicle
stocks. J Tsinghua Univ Sci Technol 51(6):868872
Hao H, Wang HW, Ouyang MG (2011c) Comparison of policies on vehicle ownership and use between Beijing
and Shanghai and their impacts on fuel consumption by passenger vehicles. Energ Policy 39:10161021
Hao H, Wang HW, Ouyang MG (2011d) Fuel conservation and GHG (greenhouse gas) emissions mitigation
scenarios for Chinas passenger vehicle fleet. Energy 36(11):65206528
Hao H, Wang HW, Ouyang MG (2012) Fuel consumption and life cycle GHG emissions by Chinas on-road
trucks: future trends through 2050 and evaluation of mitigation measures. Energ Policy 43:244251
Hou Cong, Wang Hewu, Ouyang Minggao (2012) A comparison study on the evaluation methodologies of
PHEV energy consumption between China and U.S.. accepted by Automotive Engineering (in Chinese)
Hou Cong, Wang Hewu, Ouyang Minggao (2013) Survey of daily vehicle travel distance and impact factors in
Beijing. In: Advances in Automotive Control, Vol. 7, Part 1, pp 3540. 7th IFAC Symposium on Advances
in Automotive Control, Tokyo, Japan, 2013. (DOI) 10.3182/20130904-4-JP-2042.00002
Huo H (2005) Study on links light-duty vehicle emissions based on traffic characteristics. Dissertation, Tsinghua
University, Beijing, China
Huo H, Wang M, Johnson L, He DQ (2007) Projection of Chinese motor vehicle growth, oil demand, and CO
2
emissions through 2050. Transp Res Rec: J Transp Res Board 2038:6977
Huo H, QiangZ, Kebin H, Zhiliang Y, Michael W (2012) Vehicle-use intensity in China: current status and future
trend. Energ Policy 43:616
Karabasoglu O, Michalek J (2013) Influence of driving patterns on life cycle cost and emissions of hybrid and
plug-in electric vehicle powertrains. Energ Policy 60:445461
Lin Z, Dong J, Liu C et al. (2012) PHEV energy use estimation: validating the gamma distribution for
representing the random daily driving distance. Transportation Research Board 2012 Annual Meeting.
http://pubsindex.trb.org/view/2012/C/1129781 Washington, D.C., USA, 2012.
Michalek JJ, Chester M, Jaramillo P et al (2011) Valuation of plug-in vehicle life-cycle air emissions and oil
displacement benefits. Proc Natl Acad Sci 108(40):1655416558
MOST (Ministry of Science and Technology) (2013) (in Chinese) http://www.most.gov.cn/ztzl/kjzjbxsh/
kjzjbxmt/201302/t20130204_99599.htm. Cited 18 February 2014.
Neubauer J, Aaron B, Wood E (2013) Sensitivity of plug-in hybrid electric vehicle economics to drive patterns,
electric range, energy management, and charge strategies. J Power Sources 236(2013):357364
NHTS (National Household Travel Survey) (2014) http://nhts.ornl.gov. Cited 18 February 2014.
NREL (National Renewable Energy Laboratory (2014) Transportation secure data center. http://www.nrel.gov/
vehiclesandfuels/transportationdata. Cited 18 February 2014
SAE (2009) Utility factor definitions for plug-in hybrid electric vehicles using 2001 U.S. DOT national
household travel survey data, Hybrid Committee, J2841
SAE (Society of Automotive Engineers) (1999) Recommended practice for measuring the exhaust emissions and
fuel economy of hybrid-electric vehicles, Hybrid Committee, J1711
Vilimek R, Keinath A, Schwalm M (2012) The MINI E field studysimilarities and differences in international
everyday EV driving. In: Stanton NA (ed) Advances in human aspects of road and rail transport, oca Raton,
FL, pp 363372
Wang HW, Ouyang MG (2007) Transition strategy of the transportation energy and powertrain in China. Energ
Policy 35(4):23132319
Wang W, Xiang Q, Chang Y et al (2002) Analysis method of urban transport energy consumption and
environmental Impact. Science Press, Beijing
Wang M et al. (2006) Projection of Chinese motor vehicle growth, oil demand, and CO
2
emissions through 2050.
ANL/ESD/06-6. Argonne National Laboratory, Argonne, USA
Wang Y, Teter J, Sperling D (2011) Chinas soaring vehicle population: even greater than forecasted? Energ
Policy 39:32963306
Wang HW, Hao H, Ouyang MG (2013) Scenario analyses of Chinas vehicle ownership and vehicle traffic
services. In: sustainable automotive energy system in China, Springer Berlin, Heidelberg, pp 2758
Wu Q, Nielsen AH, Østergaard J et al. (2010) Driving pattern analysis for electric vehicle (EV) grid integration
study. In: Proceedings of innovative smart grid technologies conference Europe (ISGT Europe), 2010 I.E.
PES. IEEE, pp 16. Gothenburg, Sweden, 2010.
Mitig Adapt Strateg Glob Change
Article
Nearly ten years have passed since the comeback of battery electric vehicles (BEVs). However, the percentage of such vehicles on the roads is still low and there is widespread concern about the long-term success of this technology. On the other hand, vehicles fueled with hydrogen, which it is beginning to be praised by several car companies as the fuel of future, do not yet have broad market acceptance, being even lower than battery electric vehicles and facing significant barriers for a widespread diffusion. In this research work, a novel type-approval test procedure is developed aiming to stablish an impartial simulation-based comparative analysis of technical feasibility, performance, range, and greenhouse gas emissions of two hybrid powertrains that manage electricity stored in a lithium-ion battery and hydrogen in tanks: Plug-In Fuel Cell Hybrid Electric Vehicle (FC-PHEV) and Extended Range Fuel Cell Hybrid Electric Vehicle (FC-EREV). The simulation framework developed considers the barriers and how to avoid them through the implementation of a set of fuel management rules. The findings indicate that hybrid powertrains achieve greater ranges and lower greenhouse gas emissions, opening the door to introduce these vehicles onto the market without the need to wait till the main barrier, the hydrogen refuelling infrastructure, is fully developed.
Book
Full-text available
This book mainly focuses on the innovations in intelligent transportation infrastructure and management. The content of the book is selected in such a way that it will cover a wide range of areas to integrate advanced technologies and provide best and innovative solution to problems faced by the rapidly growing transportation sector. The topics of the book primarily address the needs of the students in civil, electrical, and mechanical engineering. It is equally useful as reference material for innovators, inventors, practitioners, and policymaker for an innovative and classified solution in the field of transportation and infrastructure management. Looking at the global electric and hybrid vehicles market, the book focuses on and discuss recent developments in electric mobility across the globe. In this edition, we try to feature toward an update on the performance and costs of batteries which is the current demand of the consumers. The book inspires researchers, innovators, industry experts, and policymakers to consider the solutions into the practice.
Article
The actual energy saving effect of plug-in hybrid vehicles (PHEVs) is usually evaluated by the electricity utility factor (UF) in a standardized charging pattern. To further evaluate the impacts of the charging pattern heterogeneity of PHEV, actual vehicle travel data are adopted to classify the charging pattern in seven typical Chinese cities and derive its impacts on actual UF. Additionally, UF curves are fitted as power exponential functions. The result shows that during daily usage, UF can reach over 0.8 for 50 km all-electric-range PHEVs for the 77% frequently charging adopters, while it is as low as 0.1 for the 3% rarely charging adopters. Comprehensive UF values at an actual charging pattern are 0.53 and 0.68 for the 50 km and 80 km all-electric-range PHEVs, and the values are 0.03 and 0.04 smaller than the standard UF, respectively.
Article
Promoting Plug-in Hybrid Electric Vehicles (PHEVs) is a promising strategy to mitigate the climate change and improve the environment in the road transportation sector. To evaluate the complex energy saving and GHG emission reduction contribution of on-board fuel and electricity, utility factor (UF) is the most widely adopted parameter to weight the distance driven by electricity and on-board fuel, respectively. However, few researches focus on city-level actual utility factors based on quantitative analysis of the influencing factors, especially in China. In this research, a probability-distribution-based PHEV UF function is constructed from travel pattern parameters, charging frequency and PHEV charge depleting range. Then the urban city-level UF at actual travel and charging patterns are derived from PHEV driving big data in seven typical cities in China. The sales-weighted actual UF is 0.49 and 0.64 for a 50 km-electric-range PHEV and a 80 km-electric-range PHEV; therefore, PHEVs with over 80 km electric range are recommended in China urban areas due to better energy saving and emission reduction potential. Finally, carbon and pollutant emissions are analyzed, actual PHEV carbon emission is 133–137 g/km currently while it could drop to 71–88 g/km at a grid with 80% renewable energy; therefore, cleaner grid would largely contribute to PHEV environmental effects.
Chapter
Atmospheric pollution is extremely raised in recent years due to the growth of fuel-based vehicles number and the increase of energy generation coming from nonrenewable sources. To alleviate this issue, electric vehicles (EVs) and renewable sources are applied in many countries. This chapter presents an optimal model able to forecast photovoltaic (PV) power generation and optimize power flows between grid, PV system, and EVs at workplace charging station. Considering multimarket platforms, the optimal model aims to maximize the selling cost of the surplus energy discharged by EVs and the PV system while minimizing the EVs’ charging energy cost. The developed model consists of three components: back-propagation neural network model to predict PV power generation, the EV user’s behavior uncertainties by the probability distribution functions, and mixed-integer linear programming optimal framework that allocates power to maximize the selling cost of energy and to minimize purchasing energy cost. The results show that the developed approach can minimize the purchasing energy cost while maximizing the selling cost of energy.
Article
The sales of electric vehicles (EVs) are rapidly increasing and their integration in the power system is becoming a crucial issue. However, there is a scarcity of necessary data to derive charging profiles and analyze their impact on the power system. The purpose of this manuscript is to provide a comprehensive review of published data sources that can be useful for EV studies in the context of smart grids and power systems. The manuscript focuses on the last two decades of published data, as this is more complete and reliable in terms of user and vehicle behavior. Data sources are categorized into three classes: surveys, internal combustion engine vehicles and EVs trials, and charger trials. Data from the different sources are summarized, including information regarding how and what kind of data has been collected and their availability. Based on the reviewed sources, five parameters are identified as essential to derive charging profiles: battery capacity, charging power, plug-in state of charge, plug-in/out time and charged energy. In order to observe individual behavior it is important to consider sets of charging sessions per charger, otherwise important correlations may be neglected. Depending on the source and data availability, in many cases this is not possible. To this end, this manuscript discusses how to use data from various sources to complement missing information and concludes with guidelines and limitations about data usage in EV studies.
Article
* Full text available till August 20, 2021. https://authors.elsevier.com/a/1dKok4rgZijM6c Digitalizing documents is evaluated as a key value-added factor in greening maritime transportation. Covid-19 accelerates the need for digitalizing shipping documents. However, very few studies analyze digitalization in maritime transportation from the environmental aspect, which is an important topic considering the global emphasis on decarbonization. The methods to measure document-related emissions are mainly conceptual guidelines. Therefore, this research aims to develop an estimation framework with concrete methods to quantify the greenhouse gas reductions from digitalizing shipping documents. Taking Singapore and China as examples, this study compares blockchain with centralized systems and an integrated platform with separate platforms. The major results are 1) digitalization can reduce over 99% of document-related emissions; 2) passenger road transport to transfer documents is the largest emission source in paper systems constituting over 90% of emissions per shipping event; 3) blockchain and centralized systems have similar effects on emission reductions; so do integrated and separated platforms.
Chapter
Plug-in hybrid electric vehicle (PHEV) technology combining the merits of Battery electric vehicle (BEV) and Hybrid electric vehicle (HEV), has the potential to reduce greenhouse gas (GHG) emissions, and petroleum consumption in the transportation sector. However, the cost-benefit of PHEVs mainly determined by battery technology, optimal powertrain design, and vehicle kilometers daily traveled and charging habits. Targeting to cost-benefit, the optimal design method was presented, taking battery cycle life Vs DOD data, driving data, battery performance data into consideration. The method provided optimal vehicle designs to realize minimum life cycle cost, and maximum petroleum consumption under different scenarios. For A-segment equivalent PHEV (similar to a F3DM), under Shanghai urban driving conditions, it can be find that while PHEVs with present traction battery technology, 30 km AER was most life cycle cost-effective to obtain maximum petroleum displacement based on Shanghai driving data. Large capacity battery lead to petroleum displacement not so much as cost increased. At China electricity price off peak, Li-ion battery pack costs must fall below ¥2.0/Wh to be cost competitive with equivalent internal combustion engine vehicles (ICEs).
Article
A hybrid vehicle ownership model, which comprises three sub-models (private passenger vehicle population model, public traffic vehicle population model, and other vehicle population model), was established to simulate the growth of China’s vehicle population. The passenger vehicle population model links the vehicle population and residents’ income distribution to forecast the growth of the private passenger vehicle population. The public traffic vehicle population model links the vehicle population, human population, and urbanization rate to forecast the growth of the vehicle population of urban public buses and taxis. The other vehicle population model links vehicle population growth and GDP growth to forecast the population of all other vehicles, including passenger vehicles for public affairs, other buses, and trucks. Passenger and freight traffic volume were projected based on the forecast results for the vehicle population. The vehicle population and traffic volume were projected under two scenarios of reference scenario and comprehensive policy scenario. The reference scenarios and comprehensive policy scenarios employed in this section are the same as those in Chap. 12 for the vehicle energy development scenario. Chapter 12 presents the scenario definitions in detail, and the present section describes only the content related to vehicle population and vehicle traffic volume.
Article
The stocks of China's passenger vehicles and commercial vehicles were projected by employing Gompertz model and linear extrapolation model. The results indicate that China's vehicle stock will reach 252-292 million in 2020 and 398-523 million in 2030. The passenger vehicle stock will grow faster with some uncertainty, reaching 220-259 million in 2020 and 351-476 million in 2030. The commercial vehicle stock will reach 32.8 million in 2020 and 46.9 million in 2030. In spite of the fast growth of the vehicle stock in China, the vehicle ownership level of China will still be far behind developed countries.
Article
The evaluation method for the fuel and electricity consumptions of plug-in hybrid electric vehicles (PHEVs) in both China and U.S is studied. Through the comparisons in terms of measurement regulations and indicator calculation, the inadequacies in the evaluation method of energy consumption in China are pointed out, including the lack of energy consumption label, the inability to test a blended PHEV range and the neglect of the effects of trip patterns on PHEV energy consumption etc. Finally a suggestion is put forward on enhancing the research on trip patterns to improve the current evaluation method of PHEV energy consumption in China.
Article
Plug-in hybrid electric vehicles (PHEVs) have received considerable recent attention for their potential to reduce petroleum consumption significantly and quickly in the transportation sector. Analysis to aid the design of such vehicles and predict their real-world performance and fuel displacement must consider the driving patterns the vehicles will typically encounter. This paper goes beyond consideration of standardized certification eyeless by leveraging state-of-the-art travel survey techniques that use Global Positioning System (GPS) technology to obtain a large set of real-world drive cycles from the surveyed vehicle fleet. This study specifically extracts 24-h, second-by-second driving profiles from a set of 227 GPS-instrumented vehicles in the St. Louis, Missouri, metropolitan area. The performance of midsize conventional, hybrid electric, and PHEV models is then simulated over the 227 full-day driving profiles to assess fuel consumption and operating characteristics of these vehicle technologies over a set of real-world usage patterns. In comparison to standard cycles used for certification procedures, the travel survey duty cycles include significantly more aggressive acceleration and deceleration events across the velocity spectrum, which affect vehicle operation and efficiency. Even under these more aggressive operating conditions, PHEVs using a blended charge-depleting energy management strategy consume less than 50% of the petroleum used by similar conventional vehicles. Although true prediction of the widespread real-world use of these vehicles requires expansion of the vehicle sample size and a refined accounting for the possible interaction of several variables with the sampled driving profiles, this study demonstrates a cutting-edge use of available GPS travel survey data to analyze the (highly drive cycle-dependent) performance of advanced technology PHEVs. This demonstration highlights new opportunities for using innovative GPS travel survey techniques and sophisticated vehicle system simulation tools to guide vehicle design improvements and to maximize the benefits offered by energy efficiency technologies.
Article
This study systematically examined the potential impacts of recharging scenarios for multiple plug-in hybrid electric vehicles (PHEVs) in the western United States-in particular, the service area of the Western Electricity Coordinating Council (WECC)-in 2030. The goal of the study was twofold: to examine the impact of scenarios for market penetration and charging of PHEVs on the electric utilities and transmission grid and to estimate the potential reductions in petroleum use and greenhouse gas (GHG) emissions attributable to PHEV miles traveled on primarily grid electricity. Three charging scenarios for PHEVS were examined: (a) begin recharging upon arrival at home at the end of the last daily trip, (b) complete recharging of batteries just before the start of the first daily trip, and (c) any additional charging opportunity during the daytime. The three charging scenarios produced distinct hourly electric load profiles, with the opportunity-charging scenario resulting in a significant increase in load during the daytime. However, when the utility dispatch simulations were run for these charging scenarios in the WECC area, they all exhibited similar marginal-generation mixes (dominated by the natural gas combined-cycle technology) to satisfy the PHEV load, and GHG emissions were within 2% of each other. A well-to-wheel analysis revealed that the marginal-generation mixes produced 40% to 45% lower GHG emissions by PHEVs than did conventional gasoline internal combustion engine vehicles.
Conference Paper
Due to the importance of distribution of the Daily Vehicle Kilometers Travelled (DVKT) to design the powertrain of the Battery Electric Vehicle (BEV) and Plug-in Hybrid Electric Vehicle (PHEV), as well as to predict the energy and emission reduction caused by BEV and PHEV market penetration, a comprehensive survey was carried out in Beijing in the year of 2009, and more than 500 questionnaires including DVKT and the related factors were collected. The analysis results show that the average DVKT of the private passenger vehicle in Beijing was 46.35 km, and 68.2% of the travels were within 50 km while only 9.1% were longer than 100 km. The detailed analysis implied that the travel purpose is the most important impact factor of DVKT. With the same electric range, the BEV can cover approximately 5% more of the daily trips in Beijing than in the U.S. With the same charge depleting range, the average fuel consumption of the PHEV used in Beijing is 10-40% lower than that in the U.S.
Article
During this study a methodology was developed to project growth trends of the motor vehicle population and associated oil demand and carbon dioxide (CO2) emissions in China through 2050. In particular, the numbers of highway vehicles, motorcycles, and rural vehicles were projected under three scenarios of vehicle growth by following different patterns of motor vehicle growth in Europe and Asia. Projections showed that by 2030 China could have more highway vehicles than the United States has today. Three scenarios of vehicle fuel economy were also developed on the basis of current and future policy efforts to reduce vehicle fuel consumption in China and in developed countries. With the vehicle population projections and potential vehicle fuel economy data, it was projected that in 2050 China's on-road vehicles could consume approximately 614 million to 1,016 million metric tons of oil (or 12.4 million to 20.6 million barrels per day) and emit 1.9 billion to 3.2 billion metric tons (or 2.1 billion to 3.5 billion tons) of CO2 each year. Although these projections by no means imply what will happen in the Chinese transportation sector by 2050, they do demonstrate that an uncontained growth in motor vehicles and only incremental efforts to improve fuel economy will certainly result in severe consequences for oil use and CO2 emissions in China.