Energy 141 (2017) 537e549

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Performance comparison of conventional, hybrid, hydrogen and

electric urban buses using well to wheel analysis
G. Correa a, *, P. Mun
~ oz b, c, T. Falaguerra a, C.R. Rodriguez c
a
CONICET, CITCA, Universidad Nacional de Catamarca - FACEN, Prado 366, K4700BDH, San Fernando Del Valle de Catamarca, Argentina
b
INFIQC, Facultad de Ciencias Químicas, Universidad Nacional de Co rdoba, CONICET, Ciudad Universitaria, Haya de La Torre Esq. Medina Allende, 5000,
Cordoba, Argentina
c rdoba, Av.Velez Sarsfield 1611, X5016GCA, Co
Facultad de Ciencias Exactas, Físicas y Naturales, Universidad Nacional de Co rdoba, Argentina

a r t i c l e i n f o a b s t r a c t

Article history: This work presents a new method to compare energy and environmental performances of five types of
Received 16 May 2017 urban passenger buses powertrains using a multiphysic index on the basis of a well to wheel analysis.
Received in revised form The well to tank step was made for present and future (year 2030) scenarios using different assumptions
22 August 2017
for the years to come and obtaining various energy and environmental parameters. Additionally, the tank
Accepted 16 September 2017
Available online 20 September 2017
to wheel analysis was performed using dynamic models of vehicles, two different driving cycles and four
ranges. Later both stages were integrated in a well to wheel stage where relevant indexes were proposed
and discussed. In order to properly asses the different hypotheses for systems, range, cycles and sce-
Keywords:
Well to wheel analysis
narios; a multiphysics indicator (Integrated Sustainability Index), valued between zero and one was used.
Dynamic vehicle model The best results were achieved by hybrid electric vehicles for short and medium terms. In the long term
Electric battery electric vehicles are convenient only for short driving range, while the fuel cell buses yield good
Diesel and hybrid buses performances for more extended driving ranges. For the cleaner powertrains to be competitive, hydrogen
Hydrogen production must be fed with clean and renewable energies and the renewable energy share in the
Diesel electric energy matrix should be considerably high.
Battery © 2017 Elsevier Ltd. All rights reserved.

1. Introduction At present, in the transport sector, internal combustion engine

(ICE) vehicles are the headlines, but it is inevitable that a new
Worldwide, 18 million barrels of oil each day are consumed in technology will eventually replace them and everything seems to
vehicular traffic [1]. Our vehicles emit 2.7 billion tons of dioxide indicate that the technology will be electric vehicles [4]. The in-
each year [2], nevertheless in recent years, in end-use sectors, the terest in electric vehicles has increased rapidly over the past few
transport sector delivered by far the largest emissions reduction, years. New registrations of electric cars (including both battery
achieved by tightening fuel-economy standards. However, the electric and plug-in hybrids) increased by 70% between 2014 and
share of this sector in emissions remains very high and also is one 2015, with over 550.000 vehicles being sold worldwide in 2015 [2].
major consumer of fossil resources (In 2013 accounted for 63.8% of Fuel cell buses and battery electric buses have some key advantages
world oil consumption [2]), being one of the biggest contributors to over ICE vehicle as, not producing any pollutant emissions directly
the global consumption of energy. The employment of innovative from their operation, noiseless and highly efficient [5]. Their
powertrains such as full BEV (Battery Electric Vehicles); Hydrogen emissions are entirely upstream related to production of electricity
enriched Compressed Natural Gas (HCNG) powered vehicles; Fuel and hydrogen. This is especially advantageous in city centers where
Cell Hybrid Electric Vehicles (FCHEV) and Hybrid Electrics Vehicles typically there is heavy traffic and the air quality can be poor [6].
(HEV) seems a very promising step towards the energy reduction, Therefore, the use of fuel cells for transit reduces dependence on
global and local environment protection and more sustainable petroleum and adverse effects of price fluctuations. For these rea-
economic growth [3]. sons, Fuel Cell Electric Vehicles are progressing towards commer-
cialization and the number of FC bus and FC manufacturers are
increasing steadily [7]. From a transportation service cost point of
* Corresponding author. view, Lin et al. [8] studied the people's willingness to pay for the
E-mail address: [email protected] (G. Correa). adoption of new energy buses in the four most developed cities of

https://doi.org/10.1016/j.energy.2017.09.066
0360-5442/© 2017 Elsevier Ltd. All rights reserved.
538 G. Correa et al. / Energy 141 (2017) 537e549

Notation BEV Battery electric vehicle

CI Compression ignition
CNG Compressed natural gas
Latin symbols CT Charging time
A Frontal area, m2 DC/AC Booster buck converter
Ax Concentration of emission gas in the local DC/DC Inverter
environment, m g h km
1 DIFF Final drive/Differential
Ax,st Ambient air quality standard for each gas emission, m g DV Diesel vehicle
m
3 EI Emission index
CD Aerodynamic drag coefficient EM Electric motor
En Energy, J EV Electric vehicle
F Faraday constant, C mol
1 FC Fuel cell
f0 Rolling factor FC AUX Fuel cell auxiliaries
g Gravitational constant, ms
2 FCHEV Fuel cell hybrid electric vehicle
I Current, A FCS Fuel cell system
m Mass, kg GB Gear box
M Molar mass, kg mol
1 GHG Green house gases
NC Number of cells H2 Compressed gaseous hydrogen
Paux Power, W HCNG Hydrogen enriched compressed natural gas
V Speed, ms
1 HEV Hybrid electric vehicle
WEIi Weight of contaminant i for the emission index ICE Internal combustion engine
ISI Integrated sustainability index
Greek symbols LCA Life cycle analysis
a Road slope, rad LHV Low heating value
h Efficiency MC Mechanical coupling
r Air density, kgm
3 NG Natural gas
t Quantity of each air contaminant by km, m g km
1 PEMFC Proton exchange membrane fuel cell
f Residence time of each air contaminant, h PT Powertrain
RLED Reciprocal of linear energy density
Abbreviations and acronyms SI Spark ignition
aux Auxiliaries SOD State of discharge
elec Electronic components TEE Total energy efficiency
mec Mechanical components TTW Tank to wheel
req Required VGE Vehicle gravimetric energy density
AUX VEHVehicle auxiliaries WTT Well to tank
BAT Battery WTW Well to wheel

China. The results show that approximately eighty percent of the the systems under different scenarios, allowing to examine the
respondents in the four cities would like to pay a higher fare to performance of vehicles, energy consumption, range autonomy and
support the adoption of buses powered with renewable energies. environmental impact, when they are fed with varied energy
However there are significant technological barriers, such as the sources and driven along different types of roads, such as the Well-
limited driving range of those vehicles and the lack of a battery and to-Wheel (WTW) analysis. A WTW analysis is also called a fuel-
hydrogen charging infrastructure, that still prevent the widespread cycle analysis in the fuels transportation field and a life-cycle
usage of EVs [9]. For BEVs, technical barriers are mostly associated analysis (LCA) for consumer products [15].
with battery technology [10]. A significant challenge is the rela- There are many studies in literature based in WTW analysis for
tively low energy density of batteries, which means that, for a detailed examination of the transport systems and diverse gener-
reasonable range, they have to be large, heavy and expensive. For ation types applied to different countries. The WTW analysis,
example, with present technology a range of 200 km requires developed by the Argonne National Laboratory, is a useful approach
roughly 150 kg of lithium ion cells or more than 500 kg of lead acid for evaluating greenhouse gas emissions, among other important
batteries. With FCHEVs the infrastructure problem is truly a sig- indicators, produced by various means of transport using fuels
nificant one [11], there is very little commercial hydrogen-refueling produced through different pathways. It takes into account all the
infrastructure in the world and it exists only in very localized areas processes from Natural Resources extraction and/or exploitation
[12]. This means that even if an individual wishes to buy a FCHEV until the vehicle operation. This analysis can be broken down in
they are prohibited from doing so due to the lack of support two stages, well to tank (WTT) and tank to wheel (TTW). The first
infrastructure [13]. On the other hand, Diesel hybrid city buses are stage, WTT, includes the energy costs of natural resources extrac-
estimated around 30e50% more expensive that conventional Diesel tion, exploitation, transportation, processing and delivery. The
buses [6]. The variation can be partly explained by the different concept of TTW refers to the efficiency of the vehicle itself, since
hybrid technologies [14]. fuel is loaded until it is transformed into mechanical energy and
When it is planned to concretely install some of these new heat.
transport technologies, the subjects discussed before shows that it Wang [15] studied the impact of a fuel cell vehicle using the
is necessary a global vision with the aim to analyze the behavior of GREET model, and evaluated WTW energy and emissions; Mizsey
 G. Correa et al. / Energy 141 (2017) 537e549 539

and Newson [16] compared five powertrain/fuel combinations, driving conditions.

considering WTW efficiency, greenhouse gases (GHG) emissions In this context, all buses systems, relevant fuels, and primary
and investment costs; the best efficiency was obtained for the energy sources were compared with the aim to answer the next
hybrid electric vehicle (HEV) with an internal combustion engine questions:
fed with diesel, while the best WTW GHG emissions was obtained
for the FCHEV, operated with compressed H2, produced on a  What are the alternative uses of a particular resource and how it
centralized plant. In Simmons et al. (2014) [17] a Ballard fuel cell can be used in an efficient way?
stack was used. The model for the fuel cell used in this study is a  What are the alternative ways to produce a given energy vector
static model which neglects dynamic behavior. The research is and which of these can keep the best prospects?
founded on an energy consumption analysis, which is carried out
on the basis of extensive simulations in different bus routes. Five
different full size hybrid and electric city bus configurations were 2. Methods
considered in this study; two parallel and two series hybrid buses,
and one electric city bus. Overall, the simulation results indicate The method used is based on the description of individual
that plug-in hybrid and electric city buses have the best potential to discrete processes, which are steps or complete relevant pathways
reduce energy consumption and emissions. Hu et al. [18] made a for the selection of energy and emissions data. This process was
TTW analysis of a series plug-in hybrid electric bus characterized by carried out considering the energy point of view and harmful
the recuperation and fuel-to-traction efficiencies, which are emissions to the environment.
quantified and compared for two optimization-based energy For the WTT analysis two scenarios for the production of energy
management strategies. Campanari et al. [19] presents a study of vectors in Argentina were proposed, the current one (year 2018) for
the energy and environmental balances for FCHEV and BEV through all the energy vectors and a future scenario (year 2030) for elec-
the method of WTW analysis, applied to ECE-EUDC driving cycle tricity generation, based on the work of Di Sbroiavacca et al. [28],
simulations, using efficiency maps models. Yazdanie et al. [20] and hydrogen production. In both cases Diesel, Compressed
presents a WTW analysis for different passenger car drivetrain Gaseous Hydrogen (H2), hydrogen enriched compressed natural gas
technologies and energy carrier production pathways in (HCNG) and Electricity were considered as the output energy
Switzerland. Torchio and Santarelli [21] proposes a WTW global vectors.
index that takes into account the energy and environmental as- Fig. 1, shows the primary energy sources, the transport and
pects, through the assignment of the costs associated to the energy distribution process, and the relevant fuels and energy vectors to
and to the pollutant emissions. In Svensson et al. [22] work, a WTW supply all propulsion systems used, giving a visual description of
approach was applied in order to evaluate the energy and envi- the pathways.
ronmental impacts of introducing hydrogen in the transportation For the Diesel, the emissions and primary energies needed to
sector under conditions relevant for the Norwegian energy system. produce it from crude oil were considered along with its distribu-
In Garcia et al. [23] examined the environmental impact caused by tion by truck and barges.
the life cycle of the process of production, conditioning, and The H2 was obtained from natural gas reforming and trans-
transporting of the fuels used by buses (diesel, biodiesel (B100), a ported by virtual pipelines, for the current scenario and electrolytic
blended biodiesel at 20% (B20), and natural gas) is examined, where hydrogen from wind farms delivered through pipelines for a future
a WTW analysis was also included. Sharma & Stresof [24] per- scenario.
formed the environmental and economic life cycle analysis of the The HCNG was obtained as a mixture of the Compressed
impacts of alternative transport fuels and a comparison with con- Gaseous Hydrogen and Compressed Natural Gas (CNG). The latter
ventional fuels for Australian conditions. Karabasoglu and Michalek was considered as the U.S.A. CNG produced using the Argentinian
[25] report that the driving conditions affect the performance of Electric Matrix.
different powertrains producing efficiencies and GHG emissions For the Electricity production and distribution the Argentinian
changes, hence using significantly different driving cycles would electricity matrix was taken into account as a case study consid-
allow a more comprehensive analysis and reduce the bias yielded ering all the available generating methods in Argentina in its proper
by the use of a single driving cycle. In Zhou et al. study [26], three share: thermal (stem and gas turbine, diesel engine, combined
BEV models were tested on-road while participating in the cycle), nuclear, renewables (solar and wind) and hydraulic. The
demonstration project in Macao and LCA and WTW analyses were electricity production needs as inputs natural and enriched ura-
applied in the energy and environmental assessments for alterna- nium, renewable sources such as solar and wind, and fossil fuels
tive fuel options and battery systems. produced domestically and imported such as oil, gas and coal. Also
In this work, a general method to compare energy and envi- the use of biofuels was considered since they are promoted by Law
ronmental performances of different types of powertrains and 26.093 00 Regulation and Promotion Regime for the Production and
energy vectors using a single multiphysic index is proposed. The Sustainable Use of Biofuels”, to make blends with biodiesel.
study is carried out within the WTW scope and applied to an urban In the TTW stage five powertrains, that use the energy vectors
passenger bus with five different propulsion systems, fed with their analyzed in the WTT stage, were proposed for the usage in a bus for
respective fuels (or energy vectors) obtained from different sources. urban passenger transport:
The transport sector is analyzed as an essential constituent part of
the growth of a smart city. In turn, in the WTT stage different pri-  Internal Combustion Engines (ICE) fed with Diesel.
mary energies for the fuels and energy scenarios for the production  ICE fed with HCNG (30% V/V hydrogen on CNG).
of electricity, both current and future, are evaluated framed to  Fuel Cell Hybrid Electric Vehicle (FCHEV) fed with hydrogen.
Argentina. The buses energy consumption and emissions (TTW  Battery Electric Vehicle (BEV) fed with electricity.
stage) vary significantly due to driving conditions (i.e. congestion,  Hybrid Electric Vehicle (HEV) fed with diesel.
geography, and number of stops) and propulsion configurations
(i.e. degree of hybridization, battery type, and fuel cell type) [27]. These powertrains were studied on a bus for urban passenger
Therefore in the TTW stage dynamic models were used in this work transport for the WTW analysis from the perspective of mass
along with two different driving cycles that impose different transit. The main features of the bus are listed in Table 1. As shown
540 G. Correa et al. / Energy 141 (2017) 537e549

Fig. 1. Energy sources, transport and distribution process, relevant fuels and energy vectors to supply all propulsion systems used.

in Fig. 2, each model used allows to analyze the performance of Table 1

buses providing a large number of output data from which only a Bus parameters.

few of them are used for the purpose of this work and are described Parameter Value Unit
in section 2.3.
Bodywork weight 12754.4 kg
The powertrains have been studied with four different ranges: Aerodynamic drag coefficient 0.79 e
100 km, 200 km, 300 km and 400 km, and two driving cycles Rolling factor 0.0094 e
designed to assess the emission levels and fuel efficiency in vehi- Wheel Radio 0.486 m
Passengers load 1500 kg
cles: EUDClow (Extra Urban Driving Cycle for Low Powered Vehi-
cles) [29] and UK-BUS (London Transit Bus Drive Cycle) [30]. The
driving patterns affect fuel consumption significantly, as the anal-
ysis made in Karabasoglu et al. work shows [25]. Fig. 3 graphically transported via pipelines (Forecast: Argentina 2030).
displays the frequency of speed of the two driving cycles used in
this work. UK-BUS cycle represents a real life cycles with more
starts and stops that the EUDClow cycle but with less final speed, as 2.1.2. Electricity generation
shown in Fig. 3. Table 2 shows the mix of technologies, efficiencies and partici-
pation rates in the electricity generation from Argentinian mix of
primary energies for the current scenario [34] including the targets
2.1. Well to tank model description
to achieve by 2018 of 8% (already tendered) share of renewable
energy proposed by the Argentine Republic Ministry of Energy and
The model used is the well-known GREET [15], which is an
Mining; and the proposed scenario for the year 2030.
analytical tool for estimating fuel-cycle energy use and emissions.
Each technology used for power generation from fossil sources
The model was used to study the fuel-cycle energy use and emis-
account with different raw materials: in the case of the combined
sions for four different energy vectors, Diesel, H2, HCNG, and
cycle, it has 88.42% of natural gas (NG) and 11.58% of gas oil; the
Electricity.
nuclear has 68.31% of Slightly Enriched Uranium and 31.69% of
Natural Uranium; the simple cycle gas turbine has 91.18% of NG and
2.1.1. Hydrogen production 8.82% of gas oil and finally the steam turbine has 16.06% of carbon,
Steam reforming of hydrocarbons is the most economical and 59.1% of fuel oil and 24.84% of NG. The Argentinian mix of primary
widely used process for the production of hydrogen [31]. Actually, energies for 2030 scenarios were taken from the work Di Sbroia-
approximately 90% of hydrogen generated in the world is produced vacca et al. [28].
from fossil fuels, mainly through steam methane reforming [32,33]. In both scenarios, the losses due to electricity distribution and
Natural gas, consisting primarily of methane, is commonly used as transportation in the grid were taken into account.
the main feed. Electricity is also required for the compression,
storage, and dispensing of hydrogen gas. The production of H2 for
the present scenario was considered from natural gas reforming 2.1.3. Diesel production
with sources from the Argentinian mix of primary energies. The Diesel fuel pathways, were based on the updated inputs of
Currently in Argentina, H2 is produced following the world Argentina and included fuel oil imported transported in barge,
trend of steam reforming of hydrocarbons. Since almost all the shale oil and traditional oil from domestic reserves, crude oil
production is captive, new methods can be proposed for a future transportation, diesel refining, diesel transportation and distribu-
scenario. Therefore, for the future scenario its assumed that the tion, and finally serving as fuel for the Diesel Vehicle and the Hybrid
hydrogen will be produced by wind powered electrolysis and Electric Vehicle.
 G. Correa et al. / Energy 141 (2017) 537e549 541

Fig. 2. TTW analysis scheme.

2.2. Tank to wheel model description

This section provides a description of the components used for

each one of the urban passenger bus powertrains and the different
mathematical models used to simulate them.

2.2.1. Powertrain model

The BEV and FCHEV were studied using models developed by
the authors and the remaining Buses systems were simulated using
ADVISOR [36] which approximates the continuous behavior of a
vehicle through a series of discrete steps. During each step, steady
state of the components was assumed. This assumption allows
using efficiency maps of components derived from steady state
tests in the laboratory.
This type of models, however, does not allow a detailed inves-
tigation of short-term dynamic power sources responses, which led
Fig. 3. Speed frequency of driving cycles.
to design new models of the propulsion system and control of the
fuel cells and battery powered vehicles. Fig. 4 shows the energy
2.1.4. Compressed natural gas production flow diagram of the different proposed powertrain architectures,
The natural gas includes: extraction, processing, pipeline were H2 is the hydrogen storage system, DC/DC are the Booster
transport and intermediate compression. The stages considered for Buck Converters, DC/AC are Inverters, EM are the Electric Motors,
the feedstock of this fossil fuel are: Natural Gas import from MC is the Mechanical Coupling, GB are the Gearboxes, DIFF are the
bordering countries, shale and traditional gas from domestic re- Final drives and Differentials.
serves, electricity consumption in intermediate compression stages To compute the energy required for the bus motion, several
and natural gas transport by piping. effects were taken into account: the effects of rolling of the wheel,
the force exerted by gravity on slopes, the air resistance, the effi-
ciency of electronic components, the efficiency of the electric motor
as the drive power, the mechanical efficiency and the auxiliary
2.1.5. Hydrogen enriched compressed natural gas production energy required by the vehicle [37]. This model is useful for all
Since Argentina has an extensive fleet of CNG vehicles and a kinds of vehicles by adjusting the various input parameters such as
widespread CNG distribution structure, this way of H2 utilization is vehicle weight, rolling factors, drag coefficient, efficiency tables of
considered as a natural first step for penetration as a transportation electric motors, etc. equation (1) below expresses the electrical
fuel [35]. Thus Compressed Natural Gas enriched with 30% of power needed to feed the electric motor in the authors bus model,
Hydrogen (in volume) is considered as fuel. considering rolling, gravity and aerodynamic effects, the electronic
components, electric motor and mechanical efficiencies. The pa-
rameters values were specified in Table 1.
Table 2
Mix of technologies, efficiencies and participation rates in the electricity generation h  i
from Argentinian mix of primary energies. dV
dt
þ f0 cosðaÞ þ g sinðaÞ m þ 12CD rAV 2 V
Preq ¼ þ Paux (1)
Efficiency 2018 shares 2030 shares helec hEM hmec
Hydro 25.095% 27%
The auxiliary system is considered to work at a constant power
Steam turbine 32.37 12.843% 8.8%
Nuclear 46.39 4.181% 10% of 6 kW, which is consumed by the air conditioning systems,
Combined cycle 56.54 43.235% 29.65% pumps, lights, instruments, etc; and are powered with electricity
Simple cycle 29.12 6.646% 4.55% generated on board.
Wind 7% 15.5% Table 3, shows the weights of all the systems in all the cycles
Solar 1% 4.5%
used, where empty bus refers to vehicle bodywork and FCS means
542 G. Correa et al. / Energy 141 (2017) 537e549

fuel cell system. previously validated and published in Ref. [37], includes the effect
of temperature in voltage and current using a lumped thermal
2.2.2. Diesel vehicle model for heat generation and dissipation. The model is semi-
This vehicle operates with a conventional powertrain consisting empiric and quasi-static using experimental results of a new bat-
of a compression-ignition Caterpillar 3126E Diesel Engine, a Rock- tery, without considering the aging of the battery. Within the
well RM10-145A gearbox and a standard catalyst for compression model, the code defines a surface of working points of the battery
ignition (CI) engines all of which add up to 1262 kg. using the experimental data and matches the required power with
an appropriate voltage and current output or input during the
2.2.3. Hybrid electric vehicle simulations. Fig. 5 shows the workflow of the battery model.
Hybrid vehicles make use of two or more power sources e.g. As inputs the model needs the ambient temperature, the power
electric motors and internal combustion engines. There are several required by the vehicle to complete the driving cycle and the power
powertrain configurations available for vehicles such as series, delivered by the regenerative brake of the vehicle, giving as output
parallel, series parallel, etc. For the hybrid bus a parallel configu- the voltage and current delivered, the heat transferred to the at-
ration is adopted (see Fig. 4). The engine is a Mercedes OM611 mosphere, the battery temperature, the state of discharge and the
reaching 92 kW at 4200 rpm connected to a Rockwell RM10-145A losses due to the processes of charging and discharging.
gearbox with a standard catalyst for CI engine. The electric pro- The batteries cells connection can be in series or parallel
pulsion comprises 100 modules of 6 Ah Saft Lithium Ion batteries depending the goals. Using batteries in series it is possible to in-
and a 100 kW electric motor. The powertrain elements listed above crease the voltage bus maintaining the capacity of the batteries
weight 895 kg. stack equal to the capacity of a single battery, while a parallel
arrange increases the capacity and keeps the voltage bus of the
2.2.4. Hydrogen enriched compressed natural gas fueled vehicle batteries stack equal to the voltage of a single battery. Thus a stack
Argentina has a very developed infrastructure for CNG, since in of 56 Li-ion battery cells in series is defined as the target voltage
the transport sector a fleet of nearby 2 million vehicles is feed with bus, and the number of stacks in parallel varies as the bus range is
this fuel. The very wide net of pipes and service stations implies a increased.
big opportunity for implementing a future Hydrogen economy in Table 4 shows the parameters used in the model of battery.
the country [38]. The HCNG bus uses a modified Daewoo (186 kW)
SI Engine which operates on a 30% hydrogen and 70% natural gas 2.2.6. Fuel cell hybrid electric vehicle
volume mixture. The engine was tested on a dynamometer bench The powertrain architecture of the hybrid fuel cell vehicle
to obtain the fuel economy and emissions maps [39]. The power- consist of a fuel cell system with a li-ion battery. The main source of
train is completed with an Eaton Fuller RTLO-12610B gearbox and a energy is hydrogen stored in a pressure vessel (350 bar) which is
standard catalyst for spark ignition (SI) engines adding up to transformed into electricity in the fuel cell (FC Stack), with a
1269 kg. lithium-ion battery (see model in Section 2.2.5) to help in moments
in which it does not achieves to generate the required power, which
2.2.5. Battery electric vehicle can be due to delays in the response of the fuel cell or a power
These vehicles are powered by electricity stored in Li-ion bat- request that exceeds the FC maximum power. Battery charging was
teries, especially designed for this type of vehicles. Since the power also considered through regenerative braking.
source is electricity, TTW emissions are zero. The emissions in the The Ballard FCvelocity-HD6, specifically designed for electric
electricity generation process are considered in the WTT stage. drive buses, delivers 150 kW (2 stack of 75 kW) of gross power with
Among the many advantages of electric vehicles is that of having a system weight of 400 kg. The system includes air humidification,
regenerative brakes. In traditional, friction-based brake systems, H2 recirculation and condenser for water management [7]. Table 5
the kinetic energy of the vehicle is lost as heat. Regenerative brakes shows the main characteristics of this FC and the number of bat-
allow a significant fraction of the vehicle's kinetic energy to be teries used for the EUDC and for the UK cycles.
transformed in electrical energy and store it in the batteries as The PEMFC stack dynamic model was extracted from the work
electrochemical energy. On the other hand, when a car stops at a of Correa et al. [40,41] and modified by introducing appropriate
traffic light, there is simply no fuel consumption. This contrasts parameters [42]. This model takes into account the main electro-
with internal combustion vehicles, where fuel is consumed even chemical, fluid-dynamic and thermal phenomena to predict the
when the vehicle is idle. The model used for the batteries, power output and it is coupled with the balance of plant model

Fig. 4. Energy flow of powertrain architectures.

 G. Correa et al. / Energy 141 (2017) 537e549 543

Table 3
Weights of all the systems in all the cycles in kg.

Range [km] Cycle BUS weight FCHEV bus BEV bus HCNG bus

Empty bus Cargo FCS/EM BAT H2 tank Total BAT weight Total PT HCNG tank Total

100 UK 12754 1500 750 108 209 15621 1404 15659 890 253 15658
100 EUDC 12754 1500 1000 385 166 16105 1138 15393 890 219 15624
200 UK 12754 1500 750 108 420 15832 2988 17242 890 657 16062
200 EUDC 12754 1500 1000 385 329 16268 2338 16592 890 410 15815
300 UK 12754 1500 750 108 638 16051 4854 19109 890 960 16365
300 EUDC 12754 1500 1000 385 497 16436 3662 17916 890 615 16020
400 UK 12754 1500 750 108 864 16277 7143 21397 890 1314 16719
400 EUDC 12754 1500 1000 385 665 16603 5145 19399 890 876 16281

Fig. 5. Battery model scheme.

which includes compressor, cooling devices, and water manage- Table 4

ment systems. In order to obtain the stack power output, the Parameters used in the battery model.
temperature of the whole system is computed (stack FC and water Parameter Value Unit
management system) and the input of the code needs data of the
Battery Capacity 5 Ah
external environment (ambient temperature and pressure), the test Nominal Voltage 3.7 V
drive profiles (altitude and speed). Moreover, since the voltage Max. Discharge Current 600 A
depends on the reactant pressures at the catalyst layer, the con- Max. Charge Current 30 A
Cut-off voltage 2.7 V
centration needs to be described as a function of the cells operating
Number of cells 56
condition. Depth of discharge 70%
The total consumption of hydrogen (mH2 ) is given by: Initial SOD 10%
Maximum SOD 80%
Ztf Minimum SODa 10%
MH2
mH2 ¼ 2FðIFC ðtÞ þ Iaux ðtÞ Þ dt (2) a
Minimum SOD at which the regenerative brake and FC are allowed to charge the
NC battery.
t0

where MH2 is the molar mass of hydrogen, NC is the number of cells,

studied, indicators of different nature were selected, because of
IFC is the stack current, Iaux is the auxiliaries current, F is the Faraday
their importance in the multiphysical (energy and environmental)
constant (96485 C/mol).
performance of these vehicles, for the two driving cycles described
above. The indexes proposed are listed on Table 6.
2.3. Relevant indexes
2.3.1. Vehicle gravimetric energy density
In order to compare the behavior of the five types of vehicles This analysis goes beyond fuel efficiency and consider the
544 G. Correa et al. / Energy 141 (2017) 537e549

Table 5
FCS parameters. En
hEnTTW ¼ Pwheelk (4)
Parameter EUDC UK k Eni
i
Number of Stacks 2 1
Active Cell Area 419 419
En
Stack cells no. 370 370
hEnWTW ¼ P wheel
Eni
k
(5)
Gross power [k W] 150 75 k

H2 purge percentage 8% 8%
hEnWTT
i i
where:

gravimetric density of energy (stored energy per unit mass) as an  i: Energy vectors.
efficiency indicator, in order to measure the amount of energy that  j: Primary energy sources.
the vehicle stores as a whole per unit of the powertrain mass. In this  k: Diesel, HEV, HCNG, FCHEV and BEV.
way a holistic perspective of it is obtained.

2.3.4. Reciprocal of linear energy density

2.3.2. Charging time The reciprocal of linear energy density calculates the ratio be-
Another important feature to make a correct comparison be- tween the distance covered by the buses and the amount of energy
tween mobility systems is the recharge time of storage systems, stored in them as electricity or fuel used to complete the driving
according to their autonomy range. In general, these factors are the cycle. The liquid and gaseous fuels energy is computed using its low
key barrier that reduces the attractiveness of electric mobility in heating value (LHV).
many contexts of choice. Range anxiety is a relatively new phe- This parameter allows us to appreciate the efficiency of the
nomenon in car development [43] related to the inadequacy of the vehicle but, in contrast to the energy efficiency parameter, it takes
electric range for daily travel activities and tries to point out the into account the vehicle weight as the aptitude of the vehicle to
problem that electric vehicles manufactures has to overcome to cover distances and not its performance in converting energies is
compete with gasoline and diesel vehicles [44]. For a modern EV being evaluated.
the autonomy range run between 100 km and 500 km usually,
while, the refueling time for an EV, ranging from 30 min to 10 h or
2.3.5. Emission index
more [45], depending on vehicle application (Overnight or Oppor-
This environmental impact index addresses the effects of the
tunity BEV [46]). Hybrid vehicles provide an overall autonomy
road transport sector on the environment (pollutant emissions,
range similar to diesel buses and the parallel hybrid used in the
global warming, etc.). In order to take into account environmental
Mahmoud et al. work [46] provides an additional all electric range
aspects, the pollutant emissions associated to both the production
of 10 km. With 350 or 700 bar H2 storage, the FCHEV provides a full
procedure (WTT) step and the final utilization of a fuel step (TTW)
electric autonomy range similar to diesel bus [47].
has been evaluated. The WTT emissions in the extraction, chemical
processing and transport was computed using the Greet software
and the TTW emissions with the author's models and Advisor
2.3.3. Total energy efficiency models. The indexes were obtained following Hacatoglu et al., 2016
Three different energy efficiencies can be calculated, the WTT [48] and are defined as:
efficiency, the TTW efficiency and the WTW efficiency. The WTT
energy efficiency is the energy vector output divided by the energy EI ¼ bNOx WEINOx þbCO WEI (6)
CO
consumed from the sources. The TTW energy efficiency calculates
the ratio between the energy needed to move the vehicle (wheel
1
energy) and the amount of energy supplied to the vehicle as energy WEINOx ¼ WEICO ¼ (7)
2
vectors. The WTW energy efficiency is the ratio between the energy
needed to move the vehicle and the amount of energy supplied by
Ax;st
the energy vectors to the vehicle, plus the energy consumed to bx ¼ (8)
produce the energy supplied to the vehicle, i.e. the energy supplied Ax
to the vehicle divided by the WTT efficiency. Equations (3)e(5)
where Ax,st is the ambient air quality standard for each gas emission
were developed considering the fact that the vehicles are fed by, at
(EPA, 2011), and Ax represents the concentration of gas emission in
most, two energy vectors.
the local environment and is calculated as shown:
Eni
hEnWTTi ¼ P (3) Ax ¼ ft (9)
Enj
j
where f is the residence time and t is the quantity of each air
contaminant per km. The U.S.A. Environmental Protection Agency

Table 6
Relevant indexes.

Index Accounting for

1. EI: Emission Index Environmental aspects

2. VGE: Vehicle Gravimetric Energy Density (kWh/kg) Energy stored per kg of power train components
3. CT: Charging Time (min) Charging or refueling time of energy storage
4. TEE: Total Energy Efficiency (valued between 0 and 1) Efficiency in energy conversion
5. RLED: Reciprocal of Linear Energy Density (km/kWh) Efficiency of the vehicle to cover distances
 G. Correa et al. / Energy 141 (2017) 537e549 545

(EPA) has identified air contaminants to be monitored as part of its Table 8

national ambient air quality standards [49]. In this work only two Weight values involved in ISI index.

air contaminants (NOx and CO) were considered and are shown in Ij TEE RLED VGE EI CT
Table 7. Wj 0.3 0.3 0.1 0.2 0.1
The results are expressed as linear density of emissions (g/km).

2.4. The integrated sustainability index and HEV) have significant differences with respect to the UK. The
HEV improves the efficiency and CO emission rates significantly.
Seeking to address a more comprehensive approach in the The diesel improves in energy efficiencies and NOx emissions, but
assessment of the sustainability of the different buses studied here, the VGE is reduced and the CO emissions increase. The HCNG im-
able to include efficiency ratios, autonomy ranges and the envi- proves on all indexes except for the VGE.
ronmental friendliness degree, an Integrated Sustainability Index The UK400 presents a more balanced graph between the
(ISI) (Hacatoglu et al. [48]) is proposed. The ISI index assess his different configurations of buses, although again the BEV and the
performance considering appropriate weighting factors associated FCHEV appear to have the best performances. They practically do
with those relevant indexes (described above and used as in- not modify their indexes from 100 to 400 km. In contrast the HEV
dicators) with normalized values ranging from zero to one, where presents an evident improvement since it substantially increases
one is the best possible evaluation achievable for the buses. The the VGE index. The diesel and HCNG also increases the VGE while
value of the indicator j (Ij) is multiplied by its weighting factor (Wj). maintaining the other indexes. The EUDC400 presents the most
The ISI of the system is obtained as the sum of this values. even scenario. Bus configurations with ICE improve on the VGE
All weighting factors were taken based in the criterion of index with respect to the EUDC100, but decrease slightly with
reference [50] and are shown in Table 8. respect to the UK400. They also improve CO emission rates for the
UK400 and EUDC100. The buses with electric motor (BEV and
3. Results and discussion FCHEV) practically do not modify their indexes with respect to the
cycle UK or to the range 100 km.
In order to analyze the results, they will be divided into two Figs. 7e10 show the WTW emissions (NOx and CO) and effi-
parts TTW and WTT. While the calculations were performed for all ciencies (TEE and RLED) for each one of the bus configurations and
ranges (100, 200, 300 and 400 km), the figures used in this section for the 100 and 400 km ranges of each cycles in the 2018 scenario.
show the performance of the five bus configurations in the lowest In the case of the WTT only the efficiency and emissions indexes are
and highest ranges (100 and 400 km) for the two cycles and are evaluated since the CT and VGE are a vehicle-specific characteristic.
plotted for each of the proposed scenarios. Fig. 7A shows the behavior of the TEE efficiency for the EUDC
In Fig. 6 the results of the TTW analysis are shown and each of cycle with 100 km range in the 2018 scenario, where, in general, the
the indexes explained in the previous section (TEE, CT, VGE, etc.) are buses with ICE have a superior performance for the WTT, on the
graphed in bars. The emission index was not used in this figure, other hand the electric motor buses have better performance for
instead, each GHG emissions was plotted separately. the TTW. In the case of FCHEV it has an average performance for
The hight of the indexes shown in the graphs is constructed by WTT and TTW. In the WTW analysis, HEV and BEV show the best
performing a canonical normalization of the results and thus are behaviors.
expressed on a scale of 0e1, with 1 being the best. In the case of the In Fig. 7B the range is changed (400 km) and behaviors similar to
efficiency indexes (TEE, VGE and RLED) the higher efficiency is the EUDC100 can be seen. Fig. 7C shows the UK cycle for 100 km,
closer to one, in the case of CT and the environmental indexes the where a remarkable deterioration can be seen in all the bushes
higher the parameters the lower its index. Thus, for the NOx and CO using ICE. This behavior is due to the fact that the UK cycle has more
emissions the value of 1 is achieved when there are no emissions stops than the EUDC cycle, therefore two things happen: ICE en-
(FCHEV and BEV). The graph is divided into four parts. The lowest gines work much longer at lower rpm and hence lower efficiencies,
and highest ranges (100 and 400 km) for the two cycles were used and regenerative braking increases the performance of systems
to display the results variation. with electric motors. Fig. 7D increases the range to 400 km and
For the UK100 it can be seen that the BEV has an excellent similar behaviors to the UK100 can be seen.
performance in almost all indexes, except in the CT that is the A comparison of Reciprocal of Linear Energy Density (RLED)
lowest. Then, the FCHEV has very good indexes, a little more even TTW and WTW for 100 km and EUDC cycle is shown in Fig. 8A and
than the BEV since it has a high rate of energy efficiencies (TEE and 400 km and EUDC cycle in Fig. 8B and 100 km and UK cycle en
RLED), but its gravimetric efficiency index (VGE) is very low. As the Fig. 8C and finally 400 km and UK cycle is shown in Fig. 8D. Fig. 8
BEV, the FCHEV has excellent emission rates (zero emissions). In outlines a different behavior of BEV and ICE systems: Energy
relation to the CT, the FCHEV has similar index to Diesel, HCNG and required per km is strongly dependent on range for BEV, while
HEV. Diesel and HEV have similar rates, although HEV improves Diesel, HEV, FCHEV and HCNG has only a slight dependency.
efficiency rates. Finally the HCNG has very low NOx (high index) Fig. 9 reveals how the emissions related to the generation (WTT)
and high CO (low index) emissions. In the case of EUDC100 the BEV of the electricity, hydrogen and HCNG energy vectors draw the
stands out with low rates in VGE and CT and very high in emissions WTW CO emissions performance to be the worst in vehicles that
and efficiencies as in the UK100 case. The FCHEV also behaves have little or no emissions at all. The NOx emissions plotted in
similarly to the UK100. In contrast the buses with ICE (Diesel, HCNG Fig. 10 shows the opposite effect, in which the ICE high emissions
outweigh the potential benefits of low emissions during the energy
vector production (WTT).
Table 7 Figs. 11 and 12 were performed to show the bus configurations
Emission index parameters.
and relevant indexes (see section 2.3) for the range of 100 and
t[h] Ax,st [mg,m
3 ] 400 km and for the two cycles analyzed. In order to display the
NOx 24 100 multivariate data, the indexes were plotted in two-dimensional
CO 840 10000 radar charts, which pretend to show a comparison between buses
546 G. Correa et al. / Energy 141 (2017) 537e549

Fig. 6. TTW result, expressed on a scale of 0e1, with 1 being the best.

Fig. 7. WTW TEE for the EUDC cycle 2018 scenario result.

Fig. 9. WTW Emission performance - CO.

Fig. 8. RLED WTW and TTW results.

without giving any weights to the indexes used. The most desirable
performance in the bus configuration occupy the periphery of the
graph.
The current scenario for Argentina (2018) is shown in Fig. 11 Fig. 10. WTW Emission performance - NOx.
where it can be seen that for the case of EUDC100 the HEV bus
has the best behavior, except in the EI where Diesel Bus shows In the 2030 scenario (Fig. 12), the electric matrix renewables
better performances. participation is increased as stated in section 2.1.2 and the
 G. Correa et al. / Energy 141 (2017) 537e549 547

Table 9
ISI 2018.

ISI EUDC ISI UK

100 200 300 400 100 200 300 400

DIESEL 0.85 0.88 0.9 0.92 0.81 0.83 0.86 0.88

HEV 0.89 0.91 0.93 0.96 0.84 0.87 0.9 0.93
HCNG 0.63 0.66 0.67 0.69 0.45 0.47 0.48 0.5
FCHEV 0.61 0.62 0.62 0.62 0.6 0.61 0.61 0.61
BEV 0.68 0.67 0.66 0.64 0.7 0.68 0.65 0.62

Table 10
ISI 2030.

ISI EUDC ISI UK

100 200 300 400 100 200 300 400

DIESEL 0.58 0.61 0.63 0.65 0.54 0.56 0.59 0.61

HEV 0.63 0.65 0.68 0.7 0.58 0.61 0.64 0.66
HCNG 0.47 0.49 0.51 0.52 0.33 0.35 0.37 0.38
FCHEV 0.67 0.67 0.67 0.67 0.65 0.65 0.65 0.65
BEV 0.66 0.65 0.64 0.62 0.65 0.64 0.61 0.59

the range for both driving cycles, closely followed by Diesel bus. The
BEV, FCHEV, and HCNG turn out to be much less favorable in every
range. The BEV bus stands in the middle point for the UK cycle with
Fig. 11. Radar chart WTW 2018.
a 100 km range.
As shown above BEV, FCHEV, and HCNG buses performance rely
hydrogen production switched to electrolytic, through wind farms strongly on the electric matrix and the hydrogen production which
with distribution via pipeline. This conditions boost the perfor- for the 2018 scenario is very fossil fuel dependent. Thus their ISI
mances of the BEV and FCHEV powertrains, attaining great effi- performance is poor.
ciencies for the BEV and excellent emissions indexes for the FCHEV. The 2030 scenario (Table 10), with its increase in renewable
The ISI evaluations of the 2018 and 2030 scenarios for both energy generation participation evens the performance of the ve-
cycles are shown in Tables 9 and 10 respectively. The cells of the hicles, although it is not enough to place neither BEV nor HCNG
table are colored in a gradient according with the values, green performances over the HEV configuration. Nevertheless, with the
being the best valuation, red the worst, and white the middle point. introduction of electrolytic hydrogen from wind farms the FCHEV
In Table 9 it can be seen at first glance that the ISI for HEV bus increase its performance, beating the HEV for UK cycle in 100 km,
with a 400 km range is the most favorable, from the viewpoint of 200 km and 300 km ranges.
The major ISI difference between ranges is founded in BEV bus
in the 2018 scenario and in the HEV bus in the 2030 scenario, while
FCHEV present the most stable ISI performance in both cycles and
scenarios.
In the first case, this behavior can be explained by the fact that
the weight of the energy storage system increases proportionally to
the range. This impacts negatively on both gravimetric (VGE) and
total energy (TEE) efficiencies. In the second case, this difference is
due to the change in the VGE index. In the third case, the change in
the weight of the energy storage system does not increase as
significantly as in the BEV in function of the range.
Comparing both cycles, it can be observed that the HCNG bus
improves its performance significantly in relation to the other ve-
hicles, as the range increases in the EUDC cycle, but not in the UK
one.

4. Conclusions

In this work a new method to compare different buses power-

trains using a WTW analysis is presented. This analysis was divided
in two stages, WTT and TTW; framed to Argentina and performed
for five different buses configurations: Diesel, Hybrid, Compressed
Natural Gas enriched with Hydrogen, Proton Exchange Membrane
Fuel Cell and Battery Electric Vehicles. The WTT step was made for
present (2018) and future (2030) scenarios varying the energy
vector pathways and technologies. The TTW analysis was carried
Fig. 12. Radar chart WTW 2030. out using two different standard driving cycles, four ranges
548 G. Correa et al. / Energy 141 (2017) 537e549

(100 km, 200 km, 300 km and 400 km) and dynamic models of comprehensive hydrogen infrastructure for fuel cell electric cars in view of EU
GHG reduction targets. In: Fuel cell vehicle demonstration and hydrogen
vehicles. In order to compare the powertrains, five relevant indexes
infrastructure project in Japan; 2015.
were proposed, Total Energy Efficiency, Reciprocal of Linear Energy [13] Hardman S, Steinberger-Wilckens R, van der Horst D. Disruptive innovations:
Density, Charging time Emission Index and Vehicle Gravimetric the case for hydrogen fuel cells and battery electric vehicles. Int J Hydrogen
Energy Density. Qualitative analysis was done comparing the re- Energy 2013;38(35):15438e51. https://doi.org/10.1016/j.ijhydene.2013.09.
088. http://www.sciencedirect.com/science/article/pii/S0360319913023112.
sults of the indexes proposed for each step, driving cycle and range. [14] McKenzie EC, Durango-Cohen PL. Environmental life-cycle assessment of
Finally a quantitative analysis was performed using a multiphysics transit buses with alternative fuel technology. Transp Res Part D Transp En-
index (Integrated Sustainability Index). The results show that the viron 2012;17(1):39e47. https://doi.org/10.1016/j.trd.2011.09.008. http://
linkinghub.elsevier.com/retrieve/pii/S1361920911001179.
use of dynamic models in the analysis allows to evidence how the [15] Wang M. Fuel choices for fuel-cell vehicles: well-to-wheels energy and
powertrains behaviors change with the driving style. emission impacts. J Power Sources 2002;112(1):307e21. https://doi.org/
From the ISI analysis it could be concluded that: 10.1016/S0378-7753(02)00447-0.
[16] Mizsey P, Newson E. Comparison of different vehicle power trains. J Power
Sources 2001;102(1):205e9. https://doi.org/10.1016/S0378-7753(01)00802-
 The HEV is the best choice for the present scenario and for short 3.
and medium terms. [17] Simmons K, Guezennec Y, Onori S. Modeling and energy management control
design for a fuel cell hybrid passenger bus. J Power Sources 2014;246:736e46.
 In the long term, FCHEV for all ranges and BEV for shorter ranges https://doi.org/10.1016/j.jpowsour.2013.08.019. http://linkinghub.elsevier.
seem to be competitive choices if the renewable energy pro- com/retrieve/pii/S037877531301358X.
duction share of the energy matrix increases reasonably. [18] Hu X, Murgovski N, Johannesson L, Egardt B. Energy efficiency analysis of a
series plug-in hybrid electric bus with different energy management strate-
 For the cleaner powertrains to be competitive within the WTW
gies and battery sizes. Appl Energy 2013;111:1001e9. https://doi.org/
scope, the hydrogen production should be powered by clean and 10.1016/j.apenergy.2013.06.056.
renewable energies and the renewable energy share in the [19] Campanari S, Manzolini G, Garcia de la Iglesia F. Energy analysis of electric
electric matrix needs to be at least 47% as shown in the scenario vehicles using batteries or fuel cells through well-to-wheel driving cycle
simulations. J Power Sources 2009;186(2):464e77. https://doi.org/10.1016/
proposed for the year 2030. j.jpowsour.2008.09.115. http://linkinghub.elsevier.com/retrieve/pii/
S0378775308018934.
It should be noted that for the construction of the scenario 2030 [20] Yazdanie M, Noembrini F, Dossetto L, Boulouchos K. A comparative analysis of
well-to-wheel primary energy demand and greenhouse gas emissions for the
the TTW technologies were the same as the scenario 2018, and operation of alternative and conventional vehicles in Switzerland, considering
therefore no advances were considered in terms of the charging various energy carrier production pathways. J Power Sources 2014;249:
time and gravimetric density of the batteries. 333e48. https://doi.org/10.1016/j.jpowsour.2013.10.043.
[21] Torchio MF, Santarelli MG. Energy, environmental and economic comparison
of different powertrain/fuel options using well-to-wheels assessment, energy
Acknowledgments and external costs - european market analysis. Energy 2010;35(10):4156e71.
https://doi.org/10.1016/j.energy.2010.06.037.
[22] Mari Svensson A, Møller-Holst S, Glo €ckner R, Maurstad O. Well-to-wheel
This work was partially supported by the project PICT 2015 study of passenger vehicles in the Norwegian energy system. Energy
Nº0145, the project PIO Nº15920150100001CO and project SECyT- 2007;32(4):437e45. https://doi.org/10.1016/j.energy.2006.07.029. http://
UNC 2014 (Nº30720130100644CB). The authors acknowledge the linkinghub.elsevier.com/retrieve/pii/S0360544206001939.
[23] García Sa nchez JA, Lo pez Martínez JM, Lumbreras Martín J, Flores
contribution of CONICET-UNCA, MINCyT, SECyT-UNC and the Holgado MN. Comparison of Life Cycle energy consumption and GHG emis-
FCEFyN-UNC Aeronautical Department. sions of natural gas, biodiesel and diesel buses of the Madrid transportation
system. Energy 2012;47(1):174e98. https://doi.org/10.1016/j.energy.2012.
09.052. http://linkinghub.elsevier.com/retrieve/pii/S0360544212007311.
References [24] Sharma A, Strezov V. Life cycle environmental and economic impact assess-
ment of alternative transport fuels and power-train technologies. Energy
[1] Mitchell WJWJ, Borroni-Bird C, Burns LD. Reinventing the automobile : per- 2017;133:1132e41. https://doi.org/10.1016/j.energy.2017.04.160. http://
sonal urban mobility for the 21st century. Massachusetts Institute of Tech- linkinghub.elsevier.com/retrieve/pii/S0360544217307375.
nology; 2010. [25] Karabasoglu O, Michalek J. Influence of driving patterns on life cycle cost and
[2] World energy outlook 2016. 2016. emissions of hybrid and plug-in electric vehicle powertrains. Energy Policy
[3] Nanaki EA, Koroneos CJ. Comparative economic and environmental analysis of 2013;60:445e61. https://doi.org/10.1016/j.enpol.2013.03.047. http://www.
conventional, hybrid and electric vehicles the case study of Greece. J Clean sciencedirect.com/science/article/pii/S0301421513002255.
Prod 2013;53:261e6. https://doi.org/10.1016/j.jclepro.2013.04.010. [26] Zhou B, Wu Y, Zhou B, Wang R, Ke W, Zhang S, et al. Real-world performance
[4] Pilkington A, Dyerson R, Tissier O. The electric vehicle:: patent data as in- of battery electric buses and their life-cycle benefits with respect to energy
dicators of technological development. World Pat Inf 2002;24(1):5e12. consumption and carbon dioxide emissions. Energy 2016;96:603e13. https://
https://doi.org/10.1016/S0172-2190(01)00065-5. http://www.sciencedirect. doi.org/10.1016/j.energy.2015.12.041. http://linkinghub.elsevier.com/
com/science/article/pii/S0172219001000655. retrieve/pii/S0360544215016837.
[5] Mekhilef S, Saidur R, Safari A. Comparative study of different fuel cell tech- [27] Ang B, Fwa T. Study on the fuel-consumption characteristics of public buses.
nologies. Renew Sustain Energy Rev 2012;16(1):981e9. Energy 1989:797e803. Oxf, http://pesquisa.bvsalud.org/oncologiauy/
[6] Lajunen A, Lipman T. Lifecycle cost assessment and carbon dioxide emissions resource/en/rep-99875.
of diesel, natural gas, hybrid electric, fuel cell hybrid and electric transit buses. [28] Di Sbroiavacca N, Nadal G, Lallana F, Falzon J, Calvin K. Emissions reduction
Energy 2016;106:329e42. https://doi.org/10.1016/j.energy.2016.03.075. scenarios in the Argentinean energy sector. Energy Econ 2016;56:552e63.
http://www.sciencedirect.com/science/article/pii/S036054421630319X. https://doi.org/10.1016/j.eneco.2015.03.021.
[7] Hua T, Ahluwalia R, Eudy L, Singer G, Jermer B, Asselin-miller N, et al. Status of [29] Mahlia T, Tohno S, Tezuka T. A review on fuel economy test procedure for
hydrogen fuel cell electric buses worldwide. J Power Sources 2014;269: automobiles: implementation possibilities in Malaysia and lessons for other
975e93. https://doi.org/10.1016/j.jpowsour.2014.06.055. countries. Renew Sustain Energy Rev 2012;16(6):4029e46. https://doi.org/
[8] Lin B, Tan R. Are people willing to pay more for new energy bus fares? Energy 10.1016/j.rser.2012.03.032.
2017;130:365e72. https://doi.org/10.1016/j.energy.2017.04.153. http:// [30] Diego-Ayala U. An investigation into hybrid power trains for vehicles with
linkinghub.elsevier.com/retrieve/pii/S0360544217307296. regenerative braking. Mech Eng 2007;240:240 (February).
[9] Liu K, Wang J, Yamamoto T, Morikawa T. Modelling the multilevel structure [31] García L. 4 Hydrogen production by steam reforming of natural gas and other
and mixed effects of the factors influencing the energy consumption of nonrenewable feedstocks. In: Compendium of hydrogen energy; 2015.
electric vehicles. Appl Energy 2016;183:1351e60. https://doi.org/10.1016/ p. 83e107. https://doi.org/10.1016/B978-1-78242-361-4.00004-2.
j.apenergy.2016.09.082. [32] Dufour J, Serrano D, Galvez J, Moreno J, Garcia C. Life cycle assessment of
[10] Offer GJ, Howey D, Contestabile M, Clague R, Brandon NP. Comparative processes for hydrogen production. Environmental feasibility and reduction of
analysis of battery electric, hydrogen fuel cell and hybrid vehicles in a future greenhouse gases emissions. Int J Hydrogen Energy 2009;34(3):1370e6.
sustainable road transport system. Energy Policy 2010;38(1):24e9. https:// https://doi.org/10.1016/j.ijhydene.2008.11.053. http://linkinghub.elsevier.
doi.org/10.1016/j.enpol.2009.08.040. com/retrieve/pii/S0360319908016078.
[11] Thomas CE. Fuel cell and battery electric vehicles compared. Int J Hydrogen [33] Ewan B, Allen R. A figure of merit assessment of the routes to hydrogen. Int J
Energy 2009;34(15):6005e20. https://doi.org/10.1016/j.ijhydene.2009.06. Hydrogen Energy 2005;30(8):809e19. https://doi.org/10.1016/
003. j.ijhydene.2005.02.003.
[12] Weeda M, Wurster R, Bünger U, Schaap G, Wallmark C, Floris M. Towards a [34] Balance Energe tico Nacional de la República Argentina. 2015. https://www.
 G. Correa et al. / Energy 141 (2017) 537e549 549

minem.gob.ar/www/830/25604/balances-energeticos.html. operated in the project: clean Urban Transport for Europe. Energy 2008;33(5):
[35] Martínez P, Dawidowski L, Go mez D, Pasquevich D. Life cycle greenhouse 689e711. https://doi.org/10.1016/j.energy.2008.01.001. http://linkinghub.
emissions of compressed natural gashydrogen mixtures for transportation in elsevier.com/retrieve/pii/S0360544208000029.
Argentina. Int J Hydrogen Energy 2010;35(11):5793e8. https://doi.org/ [43] Hidrue MK, Parsons GR. Is there a near-term market for vehicle-to-grid
10.1016/j.ijhydene.2010.02.097. http://linkinghub.elsevier.com/retrieve/pii/ electric vehicles? Appl Energy 2015;151:67e76. https://doi.org/10.1016/
S0360319910003976. j.apenergy.2015.04.051.
[36] Markel T, Brooker A, Hendricks T, Johnson V, Kelly K, Kramer B, et al. [44] Jang J-H, Yan W-M, Chiu H-C, Lui J-Y. Dynamic cell performance of kW-grade
ADVISOR: a systems analysis tool for advanced vehicle modeling. J Power proton exchange membrane fuel cell stack with dead-ended anode. Appl
Sources 2002;110(2):255e66. https://doi.org/10.1016/S0378-7753(02)00189- Energy 2015;142:108e14. https://doi.org/10.1016/j.apenergy.2014.12.073.
1. http://linkinghub.elsevier.com/retrieve/pii/S0378775302001891. [45] Schücking M, Jochem P, Fichtner W, Wollersheim O, Stella K. Charging stra-
[37] Munoz PM, Moschen EW, Perelmuter GC, Mathe L. Seleccio  n y dimensiona- tegies for economic operations of electric vehicles in commercial applications.
miento de un vehículo ele ctrico híbrido propulsado por celdas de combus- Transp Res Part D Transp Environ 2017;51:173e89. https://doi.org/10.1016/
tible. Comparacio n y ana
lisis con un vehículo de combustio
n interna. In: 2014 j.trd.2016.11.032. http://linkinghub.elsevier.com/retrieve/pii/
IEEE biennial congress of Argentina, argencon 2014; 2014. p. 804e9. https:// S1361920915301358.
doi.org/10.1109/ARGENCON.2014.6868592. [46] Mahmoud M, Garnett R, Ferguson M, Kanaroglou P. Electric buses: a review of
[38] Sigal A, Cioccale M, Rodríguez C, Leiva E. Study of the natural resource and alternative powertrains. Renew Sustain Energy Rev 2016;62:673e84. https://
economic feasibility of the production and delivery of wind hydrogen in the doi.org/10.1016/j.rser.2016.05.019.
province of Co rdoba, Argentina. Int J Hydrogen Energy 2015;40(13):4413e25. [47] Ball M, Weeda M. The hydrogen economy Vision or reality. Int J Hydrogen
https://doi.org/10.1016/j.ijhydene.2015.01.149. Energy 2015;40(25):7903e19. https://doi.org/10.1016/
[39] Burke A, McCaffrey Z, Miller M, Collier K, Mulligan N. Hydrogen bus tech- j.ijhydene.2015.04.032.
nology validation program, tech. Rep. Davis: Institute of Transportation [48] Hacatoglu K, Dincer I, Rosen MA. Sustainability of a wind-hydrogen energy
Studies, University of California; 2005. system: assessment using a novel index and comparison to a conventional
[40] Correa G, Borello F, Santarelli M. Sensitivity analysis of temperature uncer- gas-fired system. Int J Hydrogen Energy 2016;41(19):8376e85. https://
tainty in an aircraft PEM fuel cell. Int J Hydrogen Energy 2011;36(22): doi.org/10.1016/j.ijhydene.2016.01.135. http://linkinghub.elsevier.com/
14745e58. https://doi.org/10.1016/j.ijhydene.2011.08.036. retrieve/pii/S0360319916002123.
[41] Correa G, Borello F, Santarelli M. Sensitivity analysis of stack power uncer- [49] US-EPA. National ambient air quality standards, tech. Rep. 2014. http://epa.
tainty in a PEMFC-based powertrain for aircraft application. Int J Hydrogen gov/air/criteria.html.
Energy 2015;40(32):10354e65. https://doi.org/10.1016/ [50] Hacatoglu K. A systems approach to assessing the sustainability of hybrid
j.ijhydene.2015.05.133. community energy systems by. 2014. Ph.D. thesis.
[42] Saxe M, Folkesson A, Alvfors P. Energy system analysis of the fuel cell buses