A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based range extender system

A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based range extender system

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2 Available online at www.sciencedirect.com ScienceDi...

2MB Sizes 53 Downloads 147 Views

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based range extender system   ndez*, Fernando Beltra  n Cilleruelo, Roberto Alvarez Ferna ~ Inaki Villar Martı´nez Universidad Nebrija, Pirineos 55, 28040 Madrid, Spain

article info

abstract

Article history:

Sometimes technology and development of society run slightly different roads. This situ-

Received 27 November 2015

ation is now happening in the case of hydrogen as an energy carrier in the automotive

Received in revised form

world. In the article presented here, the authors propose a change in the structure of the

8 January 2016

power plant of Battery Electric Vehicles (BEV). The objective is that these vehicles can be

Accepted 8 January 2016

presently used until the development of an electric and/or hydrogen recharge/refuel

Available online xxx

network allows being useful with the current status. In this paper a new concept of Extended Range Electric Vehicle (EREV) based in a Fuel Cell Electric Vehicle (FCEV) set

Keywords:

model is presented. A study is then developed in order to determine the working condi-

Hydrogen

tions that will lead to better efficiency and performance, referring to capacity of both en-

Electric vehicle

ergy sources: electricity stored in a Lithium-Ion battery and hydrogen gas in high pressure

Fuel cell

tanks. The possibilities here shown open the door to strategic advantages and innovation

Extended range

for car designers in the future.

Simulation

Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

Model

Introduction Nowadays, when the traditional transport model has become to its depletion, manufacturers and governments are betting hard on newer and greener technologies as a solution. Not only the progressive depletion of fuel reserves, but also the environment evolution indicate that the mobile fleet must probably change in no more than the next twenty e thirty

reserved.

years [1,2]. Many manufacturer companies agree that the Battery Electric Vehicle (BEV) is the one to beat [3,4], but differ on the specific way [5]. This has much to do with the characteristics of the different technologies of energy storage available. It is known that batteries offer a good dynamic response, while their discharge time, shorter than desired, and the recharge time, longer than desired, makes consequently that BEVs available in the market today are not suitable for many customers.

Abbreviations: BEV, Battery Electric Vehicle; EREV, Extended Range Electric Vehicle; FCEV, Fuel Cell Electric Vehicle; PHEV, Plug-in Hybrid Electric Vehicle; RE, Range Extender; ICE, Internal Combustion Engine; PDU, Power Distribution System; AFV, Alternative Fuel Vehicles; SoC, State of Charge; PEM, Proton Exchange Membrane; NEDC, New European Driving Cycle. * Corresponding author. Tel.: þ34 914521100; fax: þ34 914521111.  Ferna  ndez). E-mail address: [email protected] (R.A. http://dx.doi.org/10.1016/j.ijhydene.2016.01.035 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.  et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

2

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

A temporary solution may be the Plug in Hybrid Electric Vehicle (PHEV), as it can be charged with electricity like BEVs, run on gasoline with an Internal Combustion Engine (ICE) and use batteries to improve fuel efficiency [6]. The combination offers increased driving range with potential large fuel cost savings and emission reductions. There are two main PHEV technologies: parallel hybrids, in which both, the electric motor and the combustion engine, are mechanically coupled to the wheels through a transmission (i.e. Toyota Prius), and series hybrids, also known as Extended Range Electric Vehicles (EREV), in which the electric motor is directly coupled to the wheels and the combustion engine is only used to charge the batteries (i.e. BMW i3). Although PHEVs possess many advantages, they also have certain limitations. The main concerns include increased cost due to the introduction of engines, energy storage systems, and power converters [7], and also, fossil fuels are used. At best, a 2 to 2.5 fold fuel efficiency gain can be hoped for the world car fleet out to 2030. Most of this gain would be the result of a switch to hybrid technologies [8], and depending on the percentage of electricity derived from renewable energy that could replace most petroleum-based fuels. On the other hand, Fuel Cell Electric Vehicles (FCEVs) are powered by gaseous hydrogen, stored onboard in high pressure tanks, which is converted into electricity by multiple individual cells serial connected (fuel cell stack). A small battery pack is still used. It is typically smaller than BEV's one and it is charged by an excess of energy from the hydrogen fuel cell or through regenerative braking techniques (also often available on BEVs) which returns energy from the kinetic force when braking, by switching the motor to operate in reverse, flipping the route of the electricity and charging the battery. Hyundai Tucson ix35 Fuel Cell and Toyota Mirai are two examples of FCEVs: both are zero tailpipe emissions and enjoy good characteristics when it comes to range, as it is determined by the capacity of the tank, which can be refilled as simply and fast as a gasoline tank. Several policy initiatives have been adopted in order to promote the development of a hydrogen refuelling network: i.e. California State has committed funding for the development of 100 hydrogen fuelling stations, Japan's government proposed $71 million to build hydrogen fuelling stations, the U.K. announced over $752 million of new capital investment between 2015 and 2020 in support of ultra-low emission vehicles, including FCEVs [9]. Germany, alone, expects to have 400 hydrogen fuelling stations in 2020. Norway, Sweden and Denmark are developing the Scandinavian Hydrogen Highway to make the Scandinavian region the first in Europe where hydrogen is commercially available in a network of refuelling stations [10]. Italy is establishing a similar highway, designed to connect the country in a hydrogen way to Germany and Scandinavia. Nevertheless the slow development of refuelling infrastructure and current vehicle cost are clearly the most important hurdles keeping FCEVs from storming the market en masse [11]. In the present paper the authors have started to combine both vehicle concepts, EREV and FCEV, in order to solve these particular problems and obtain a mixed response and an improved vehicle range with easy refill.

Problem statement A configuration scheme for an EREV and a FCEV is very similar. An EREV is characterized by a powertrain composed by an electric engine, a power converter and an energy storage battery pack, that compound the vehicle propulsion subsystem (see Fig. 1). It also has a second subsystem, Range Extender (RE), composed by an Internal Combustion Engine (ICE), a fuel tank and an electric generator. That subsystem it is exclusively used to charge the batteries [7]. FCEV configuration is that similar. Like Battery Electric Vehicles, Fuel Cell Electric Vehicles use electricity to power an electric engine, but in contrast to other electric vehicles, FCEVs produce their primary electricity using the fuel cell powered by hydrogen. The vehicle uses the fuel cell as a generator to power what is otherwise a battery electric car (see Fig. 2). The power plant has also a small battery pack that helps the fuel cell to boost and also to recover energy during regenerative braking periods. Energy flow is controlled by a Power Electronic Distributor Unit (PDU). See the arrows inside the box in Fig. 2: when the vehicle is in a transient of hard acceleration, the PDU distributes the power generated from the onboard fuel cell and the battery to cover the power demand. When this transient finishes, the PDU allows the energy flowing from the fuel cell stack to the primary motor through the power converter and, at the same time, the battery could be recharged. In case of braking, the PDU manages the regenerative braking. This recovered electricity is stored in the battery. There are few models of FCEV available currently in the market but with limited distribution. Details of these models’ specifications, shown in Table 1, illustrate the efficiency of FCEVs. The models are: Midsize Car Honda FCX Clarity, Mercedes B Class FCell, Toyota Mirai and Hyundai Tucson (ix35 Fuel Cell in Europe) respectively. All these vehicles have similar characteristics. Similar power levels (about 100 kW) and hydrogen gas/electrical energy storage systems technologies are similar too, fuel tank and battery technologies. Nickel Metal Hydride and Lithium-Ion battery packs, two hydrogen pressures: 350 bar and 700 bar with fibre wrapped composite tanks. But the most important common characteristic is that all of them use a very low capacity battery storage and also that no one of them is plugin. This is obvious, as the batteries have capacities lower than 2 kWh, and it creates an auto-generated product design drawback to commercialize these vehicles, as consumers will not feel comfortable without the availability of a full refuelling infrastructure before purchasing a hydrogen fuel cell vehicle. Refuelling has been a historical problem for Alternative Fuel Vehicles (AFV) [11], but this problem is more pronounced for those AFVs that operate exclusively on a single alternative fuel, such as hydrogen fuel cell vehicles or battery electric vehicles [12]. EREVs were born as one possible solution to cope with some of the BEV limitations in this sense. Some studies have explored and compared different EREVs taking into account their energy consumption [13] or the different range

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

3

Fig. 1 e Schematic of an EREV.

extender technologies [14] and it is demonstrated that an EREV will consume, on average, less than half of the oil of a PHEV in the real world, if overnight charging is assumed, reducing regulated emissions by more than 70% when compared to a PHEV [15]. But oil is still used as fuel. Opel Vauxhall Ampera and BMW i3 are the most popular/sold EREVs in the market today. Table 2 resumes the main characteristics of these two concepts of EREV, BMW i3 has significant weight advantage, better electric range (50 km upper) and 14 g. CO2 emissions less per kilometre, but the achieved range is 120 km, so it is ideal for city use. For consumers that want to jump into electric car ownership, but to only want one car, then the Ampera is the model to go for. Its 35 L fuel tank allows a 500 km all-electric range. Nevertheless Ampera's total range is close to FCEVS’. The production of hydrogen through a home unit, which sits outside a house and reforms natural gas to produce hydrogen to power the car is today a utopia. Refuelling the vehicles through a network of charging stations is a complex but closer idea. See the example of California, the state in which there is a remarkable refuelling infrastructure for such vehicles [18,19]. Table 3 summarizes the hydrogen fuelling station in the United States. It is important to remark that 18 of the global 40

stations are located in California. The number of stations in Europe (36), Japan (21) and Korea (13) are not very encouraging, but referring to the EU situation, a recent studies [21,22] show that deploying a 25% share of FCEVs in road transport by 2050 can contribute up to 10% of all cumulative transport-related carbon emission reductions and concludes that by the end of 2025 an appropriate number of hydrogen refuelling stations needs to be in place within those Members States which adopted the use of hydrogen for road transport as one of their national policies. Therefore, in the near future two hypothetical scenarios can be posed:  A first optimistic scenario in which a large network of hydrogen refuelling will be developed.  A second and less optimistic scenario in which the hydrogen refuelling network is not fully developed or development occurs very slowly. None of these scenarios is good for the commercialization of hydrogen-powered vehicles. In scenario one there would be a refuelling network similar to the current fossil fuel one, but the possibility of electric charging at home would not exist. In the second scenario it is evident that the hydrogen-powered vehicle would be infeasible or with a negligible market

Fig. 2 e Schematic of a FCEV.  et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

4

0.97 1.03 1.07 100 kW 100 kW 100 kW 3.7 kg - 700 bar 5 kg - 700 bar 122.4 L 5.64 kg 700 bar e 144 L 385 km (NEDC) 483 km (EPA Test Data) 525 km (NEDC) Lithium-Ion- 1.4 kWh Nickel-Metal-Hydride (NiMH) - 1.6 kWh Lithium-Ion Polymer (Li-Po) e 0.95 kWh

100 kW 386 km (EPA Test Data)

Mercedes B Class FCell Toyota Mirai Hyundai ix35

DC Permanent Magnet 100 kW 100 kW 114 kW AC Induction 100 kW

Lithium-Ion (Li-Ion)

3.92 kge350 bar

1015

Table 2 e Main characteristics of the best-selling EREVs [16,17].

Honda FCX Clarity

Electric energy storage Electric engine power Model

Table 1 e Technical specifications of commercial FCEVs.

Driving range (Cycle)

Fuel capacity (massepressureevolume)

Fuel cell power

Fuel consumption (kg hydrogen/100 km)

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

Electric engine power Lithiumeion battery capacity/weight Extended range oil tank Range exender petrol engine power Electric-only range All electric range CO2 emissions Unladen weight

BMW i3

Opel Ampera

125 kW 18.8 kWh/230 kg

111 kW 16 kWh/198 kg

9L 647 cc 2-cylinder (25.4 kW) 131 km 250 km 13 g/km 1390 kg

35 L 1.4i 16v (63 kW) 83 km 580 km 27 g/km 1715 kg

penetration rate. Part of this problem could be solved with a change in the vehicle power plant architecture design, including a higher capacity battery and plug-in technology. A power plant based on an EREV concept that used electric energy storage in a high electric battery and an extended range fuel cell stack system powered by hydrogen would allow the vehicle to be used in both scenarios successfully. In scenario one, the vehicle could be recharged with electricity at home and refuel hydrogen using the network refuelling points. The second scenario would allow the driver to preferably recharge batteries at home with electricity and using the poorly developed network of hydrogen refuelling for long trips. So the authors of the present paper, aiming to optimize the design of two current concepts: EREV and FCEV, and propose a power plant architecture that mixes the following concepts: Plug-in battery, ER (Extended Range), FC (Fuel cell stack) and EV (Electric Vehicle) shown in Fig. 3. The aim of Plug-in ERFCEV vehicles is to satisfy the customer specifications defined by today's car user profile in order to cover two specific requirements of vehicle's customers: Range and refuel. This power plant structure is not a novelty for medium and heavy duty vans, trucks and buses. Renault Kangoo ZE H2 model is an example of adaptation of the standard Kangoo Z.E. electric van featuring the same 22 kW on-board battery pack and 44 kW electric motor, but also including a small range extender using a hydrogen fuel cell and hydrogen fuel tank. However in light duty vehicles this approach is novel and especially designing a switching control method on the rangeextender strategy, allowing the driver to manage hydrogen consumption versus battery charge depletion is a novelty.

Methodology A Matlab/Simulink vehicle model has been developed to estimate the energy consumption in battery range attending to the comparison of different FCEV Plug-in configurations. The purpose of the study is to find out trends on the range of the

Table 3 e Alternative fuelling station counts in U.S.A [20].

Public Public þ private

Electric (stations/charging outlets)

Hydrogen

10708/26623 12642/30826

12 40

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

5

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

Fig. 3 e Plug-in ERFC-EV power train.

vehicle with the new power plant architecture working. Matlab-Simulink block set series allows modelling, in a unique simulation environment, both the electrical and mechanical systems. The Plug-in ERFC-EV proposed is based on a recent topology, such as the Hyundai ix35, as starting point. The electric power plant consists of five major components: Battery model, Fuel Cell Stack model, Power Mixture Management System, Vehicle Dynamics and Electric Engine. The preliminary design of these components is described below, with emphasis on the available state of the art in each one and challenges related to the present application.

methodology to estimate State of Charge is the Coulomb Counting (known as Ah method too), which is widely used. The model, shown in Fig. 4, calculates the SoC by measuring the battery current and integrating it in time. One of the most important and indispensable parameters of a Battery Model is an accurate estimation of the State of Charge as it is a classic method to estimate the range of the vehicle. Battery model presented by the authors of this paper have been tested in previous works [23e26]. The model returns every relevant curve representing the performance of the battery, including the State of Charge

Battery model A battery is characterized by having a capacity, measured in ampere-hours, which indicates the strength that is capable of providing per hour from the full State of Charge (SoC 100%) to the point at which the voltage at its terminals reaches the limit called “cut-off voltage” for defining the state of charge 0%. A battery has different capacity in ampere-hours depending on the intensity with which it is discharged: with less capacity demand for high intensities, whereas for small currents the capacity increases. Considering the working conditions of the battery packs on BEVs, the selected

Table 4 e Battery parameters. Energy content Capacity (per cell) Technology Number of cells Nominal voltage Max. voltage Min. voltage Discharge cont. (power) Discharge peak (power) Max. charging current

16 kWh 40 Ah LieIon 108 400 V 448 V 324 V 200 A (80 kW) 400 A (160 kW) 80 A (32 kW)

Fig. 4 e Battery model overview.  et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

6

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

(SoC) per time, device voltage per time and per SoC, intensity running through the battery per time, power per time and effective energy delivered in the cycle. In this paper, the modelled battery has bigger capacity than the one installed in the Hyundai ix35 FCEV, because the power plant will work as a plug-in EREV. The characteristics of the battery are summarized in Table 4. This battery has to achieve a minimum range value in battery mode driving: 100 km.

Fuel cell stack model Modelling and control for a Proton Exchange Membrane (PEM) fuel cell stack system follows textbook procedures [27e29]. The equivalent circuit is shown in Fig. 5. The model gives data about the stack efficiency and consumption, and therefore the one able to be used in combination with a fuel tank subsystem, whose capacity is 142 L, pressured at 700 bar, the same characteristics than Hyundai ix35's. The model is fit to the data proceeding from datasheets [30] such as voltage at 0A, voltage at 1A, voltage and intensity of the nominal operating point and the maximum operating point, stack efficiency, operating temperature, air flow rate, supply pressure and nominal composition. As outcomes it offers voltage produced at a determined current, consumption and efficiency.

Power Mixture Management System The unified model consists of both, the battery and the fuel cell stack models, connected through a Power Mixture Management System (PMMS). The developed PPMS essentially works as follows: power demand arrives from the vehicle or from the complete dynamic plus the electric motor model; a converter adjusts that demand according to the battery instant working voltage and transforms it into a current; the PMMS decides whether the depletion of the battery requires a demand of energy coming from the fuel cell stack system or not; if it is required this intensity is derived and the corresponding power delivered by the fuel cell is reflected in a hydrogen consumption. The only input of the unified model, which is presented in Fig. 6, is therefore the drive cycle, and it offers the following outcomes: SoC evolution in the battery; current through the battery; voltage produced in the battery; power delivered by the battery; effective energy delivered by

Fig. 5 e Fuel cell stack model overview.

Fig. 6 e Unified model overview.

the battery; voltage produced in the fuel cell; intensity through the fuel cell; consumption of the stack; power delivered by the fuel cell; effective energy delivered by the fuel cell; fuel reserve remaining; total power delivered by the set and an estimation about how many driving cycles would the configuration stand. In the present paper, the New European Driving Cycle (NEDC) has been used as a driving cycle in the tests. NEDC cycle is composed of two parts: ECE-15 (Urban Driving Cycle), repeated 4 times, is plotted from 0 s to 780 s; EUDC (Extra Urban Driving Cycle) is plotted from 780 s to 1180 s. The complete cycle is shown in Fig. 7. The Power Mixture Management System architecture has been developed as a block which presents several working options. It takes into account demanded current, SoC and capacity of the battery. In the present paper the PMMS system prioritizes the battery range. It decides when the demand should start the fuel cell and its amount, considering the limits in the battery operation. The vehicle initially operates in battery-only mode and switches to Extended Range Fuel Cell powered after the battery reaches a low state of charge threshold. In this phase, the fuel cell supplies the power needed to recharge the battery vehicle and increase the SoC. To prevent the battery from supplying current above the maximum discharge current, a protection circuit will be considered. This solution can be used to control, not only the discharge current, but also the charge current. The maximum discharge and charge currents are provided by battery manufacturer's datasheets. The overcharge effect can be avoidable by applying an upper limit to the state of charge of the battery. According to previous documentation [31] this limit can be located in the 90% of SoC. The PMMS disconnects the fuel cell when the 80% of the SoC is reached, to give a margin to a possible energy recovering. It would even be possible that users can handle this limit according to their preferences. The regenerative braking charge can be regulated by the use of the brake system. When the SoC reaches the upper

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

7

Fig. 7 e New European Driving Cycle.

limit, the hydraulic break system takes over the braking needs, avoiding an overcharge of the battery. Appropriate safety measures could also be incorporated to handle realworld situations about fails or underperformance.

Vehicle dynamics This block calculates the required torque and the speed of the electric motor. The model considers the rolling resistance, the aerodynamic drag and the gravitational resistance to calculate the resistance force, according to equation, deduced from the one-directional movement fundamental equations and Newton's second law. However, the regenerative braking is limited in order to prevent the front wheels from becoming locked [23]. The optimal braking energy in the front wheels depends on the acceleration requirement and the dynamic weight distribution. Using this resistance force, it is possible to calculate the power and torque. The torque and the speed in the drive shaft are essential for the determination of the electric motor operational point. Table 5 shows the Hyundai ix35 vehicle dimensions needed to complete the simulation requirements of this block [32].

fuel cell vehicle for Hyundai's 2nd generation hydrogen fuel cell in 2005. This engine offers constant maximum torque 200 Nm (0e2000 rpm) and constant power 100 kW (3600e7200 rpm) [32]. An adjustable-frequency drive has been chosen to control the rotational speed of the electric motor. This drive controls the slip by changing the frequency of the supply of the motor. When changing the frequency, the parameter (V/f) is constant until it reaches the nominal speed. During this phase, the torque produced by the electric motor is constant. At speed greater than the nominal speed, the voltage cannot be increased and it is fixed to its nominal value, but the frequency could still be changed. In this second phase the torque decreases but the power produced by the electric motor is constant. These two electric engines have been modelled in the present paper to establish a comparison in terms of range. A power converter unit converts voltage direct current from the fuel cell stack into alternating current, which is then used to operate the electric motor. It also controls the rotating speed and torque of the motor.

Results Electric engine Hyundai ix35 2015's configuration includes a powerful electric engine rated at 100 kW (134 horsepower). It is an induction engine, bigger than the 80 kW one included in the Tucson test

Table 5 e Hyundai ix35 vehicle dynamics data set [32]. Overall Length (mm) Overall Width (mm) Overall Height (mm) Wheelbase (mm) Front Wheel Tread (mm) Rear Wheel Tread (mm) Front Over Hang (mm) Rear Over Hang (mm) Cargo Area (VDA) (litre) Lightest Curb Weight (kg) Heaviest Curb Weight (kg) Gross Vehicle Weight (kg)

4410 1820 1650 2640 1585 1586 880 890 551 2250 2290 2290

Therefore, the unified model will be run using a continuous repetition of NEDC cycles, firstly as a current demand and as a power demand afterwards. The objective of the study is to determine which working conditions, in terms of dimensions, will lead to a better efficiency and performance in the combined energy storage system in this vehicle. It has been considered interesting to carry out the test divided in the two fundamental dimensions of both devices, battery capacity and recharge power, with the aim of observe how the proposed variations in each one affects system range. The results section has been structured as follows: First, a test is performed without using the range extender (battery mode). The goal is to achieve the battery capacity that allows 100 km of range using only the energy stored in the battery. Secondly, the range extender is connected performing different recharging strategies based on the variation of the recharging amperage: beginning with high values (80 A) up to low values (10 A). The amperage will be analysed to find the better result for vehicle

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

8

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

range. NEDC speed profile will be used as input data for the vehicle in both tests. Finally, a real-world driving cycle is considered to study the behaviour of the powertrain when performing a driving cycle different to NEDC. a) Performing the tests without considering the effects of fuel cell range extender. As a reference model shape and behaviour of discharge curve of the battery, Fig. 8 presents the variation of battery voltage obtained using a vehicle operating in pure electric driving (battery mode) and performing a continuous repetition of NEDC combined cycle. It can be seen that the voltage varies in the urban part of the cycle with slightest variations and in the extra-urban part of the cycle with greater ones. This is caused due to the high power demand, which concludes in a high discharge current. There are some existing recharge notable points produce by regenerative braking (points A and B in the graph). Fig. 9 shows the variation in SoC for a battery mode working operation. It can be clearly seen how the simulation of SoC exceeds nine NEDC cycles. Each cycle NEDC covers 11023 m. b) Performing the test considering the effects of fuel cell range extender. Fig. 10 shows the evolution of SoC when the vehicle is working in an EREV mode. As the battery is drained and the SoC reaches its minimum value (30%), the fuel cell starts working and the battery is charged until the SoC reaches the upper limit (80%). In this working mode, the power supplied by the fuel cell must be lower than the power of the modelled electric engine (80 and 100 kW), because the battery is designed to charge at 80 A maximum, so only a fuel cell power of 32 kW is needed. Most of the batteries in the market will happily charge/discharge at a rate of less than 1C A. This would translate into a 1 h charge/discharge process. In practice, the charging/discharging processes may require/reduce up to twice/half the time. Most batteries can safely be used at rates above 1C, up to the rating specified by the manufacturer. However, a reduction in the battery life is surely expected, although it is difficult to quantify [31]. Fig. 10 then shows the

Fig. 9 e SoC evolution in pure electric mode.

results of the EREV's driving mode when recharging at 80, 40, 20 and 10 A. Fig. 11 shows the all-electric range in each studied case: a two engine powertrain configuration (80 kW and 100 kW nominal power) and different recharging amperage. It can be observed that the Li-Ion battery provides higher range than 100 km design requirement for battery working mode. When driving as an EREV, the fuel cell stack is capable of providing electricity to the battery and recharge it at different amperage levels: from 80 A (high level) to 10 A (low level). The range observed in Fig. 11 shows that the range decreases when increasing the recharging Amps used. This happens when the charging current exceeds 40 A value. This figure also shows the variation in nominal power for the fuel cell stack needed. The fuel cell stack dimensions in terms of power are clearly influenced by the power plant architecture and this fact is relevant, because current FCEVs (summarized in Table 1) power plants must use a 100 kW fuel cell stack as this is the main energy source. Fig. 11 shows (see the point marks) the different sizing for the fuel cell stack in terms of power. The fuel cell stack needed would achieve a maximum power of

Fig. 8 e Battery voltage (volts) versus time (seconds) plot.  et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

9

Fig. 10 e SoC vs time plots (Engine 80 kW configuration).

32 kW, about a third part of the power required in current FCEV designs. Fig. 12 shows the hydrogen tank consumption for two different range extender recharge management strategies (80 A and 40 A). One can see that the hydrogen tank consumption begins at the same time, but the recharge is faster when increasing the recharge current. It is not possible to establish a pattern for linking hydrogen consumption with the

Amps used for recharging the battery, because there exists a strong influence of the driving cycle: the point in the cycle where charging starts can match high power demand or, otherwise, be a drop zone speed. Clearly, if the fuel cell was always supplying the same amperage and also the Li-Ion battery always had the same power demand this pattern could be easily established, but when driving a car this behaviour does not usually happen.

Fig. 11 e Range results and fuel cell stack power for different electric engine power and recharging strategies.  et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

10

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

Fig. 12 e Hydrogen tank consumption.

c) Real-World performing fuel cell range extender Fig. 13 shows a daily driving profile for people in a road trip in Madrid (Spain). These data have been obtained using an OBD II data collector, placed in a vehicle running on the way between a point on the centre of the capital (Madrid) and a border town (Alcala de Henares). It is a 64 km round commute that thousands of drivers do each day. As it can be seen in Fig. 14, the vehicle can perfectly cover the 64 km journey driving in battery mode, but if the driver need to cover this distance two times in one day, a recharge of the vehicle could be necessary with current BEVS to avoid driver's range anxiety. However, with the new Plug-in ERFC-EV powertrain here presented, this problem is avoided. Fig. 14 shows how the range extender starts working when the SoC falls below the limit set for the minimum value. Once the initial charge of the battery is depleted (Point A) and reaches the minimum value allowed (30%

SoC in this case), the range-extending process will seamlessly activate the on-board electric charge drive system to continue feeding the car for the additional kilometres, streaming power to the battery and recharging it until the end of the journey. See how the different amperage affects the recharging plot graph. With lower Amps (30 A and 40 A) the SoC continues to drop under 30% value due to the power demanded from the electric engine. The driving cycle is in a high acceleration phase and the depletion of the battery is higher than the recharge process at these amperage. This does not occurs when recharging at 80 A. In terms of storage energy, the SoC values are different when the journey is finished: if the recharge is done at 40 A, the final value of SoC grows up to 46.5%, while if the used recharging amperage are lower (30 A) or higher (80 A) the achieved values in SoC achieved are 34% and 77.4% respectively. The hydrogen fuel consumption is also different: the hydrogen tank has a consumption of 12% when recharging at 40 A, 9.58% at 30 A and 20% at 80 A. It is interesting to remark that it is possible to design the system to recharge the battery between other margins (not only 30%e80% SoC), or even to only sustain the battery in the minimum designed 30% SoC value, so that it does not fully discharged during the trip. It is a manufacturer design/driver decision: prioritizing hydrogen consumption is an option if the owner of the vehicle has an electric recharge station at home. It would be necessary to design a switching control method on the range-extender strategy with a power following control method. It is obvious that the mission of the vehicle will dictate the type of control to be employed, but summarizing the results, the vehicle presented here achieves:  105 km range using the electric battery only.  Near 600 km all-electric range NEDC cycle (525 km is the FCEV NEDC current range).  It can be refuelled with hydrogen in less than 10 min.  It could be recharged at home with electricity (from 1 to 8 h).  Fuel cell stack with lower power than current FCEV's. The sizing of the components and the drivetrain architecture and control will be capable of handling real-world situations within the limits of design requirements and will improve the current BEV and FCEV architectures, allowing longer trips and easy refuelling at the same time.

Conclusions

Fig. 13 e Real-World driving profile (64 km).

Most of today's battery electric powered vehicles show a realistic electric range of less than 200 km. It is enough for most daily drives. Nevertheless, experience shows that this approach does not satisfy all customers' expectations. This paper presents the results in energy consumption of a powertrain based in the use of a fuel cell range-extender system. The vehicle drivetrain components have been sized and the suitability of this electric vehicle has been analyse in this paper. The range and fuel consumption with different energy management strategies has been studied. A vehicle homologation cycle and real world city driving cycle has been used for the analysis. As far as energy consumption is concerned, it

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

11

Fig. 14 e Real-World driving SoC evolution plot.

appears that the benefits of this power train configuration are clearly better than other powertrains, allowing higher global range (near 600 km) and also a minimum value of range in battery mode (100 km). The results presented show a powertrain concept that could avoid the commercialization problem associated to current FCEVs and BEVs powertrains and open the door for car design optimization, because the discussed powertrain could be the starting point for a driver oriented management battery system and also for a new research about sizing and managing each fuel storage system. So, in terms of optimization design, it would be possible to: (1) reduce/increase hydrogen tanks and/or reduce battery capacity, understood as the storage hydrogen in litres in the case of the fuel cell and as the cell capacity in Ampere-hours in the case of the battery, (2) recharge the battery at higher/lower C-rates and also, (3) change the lower/upper limits to SoC depletion/recharge. These possibilities will be strategic advantages and innovation for car designers in the future.

Acknowledgements The authors gratefully acknowledge generous assistance of Javier Arboleda, Service Senior Manager at Hyundai Motor Spain and Gema Marı´a Rodado Nieto, Engineer at National Hydrogen Centre (CNH2) Spain.

references

[1] Mayer Helmut. Air pollution in cities. Atmos Environ 1999;33(24):4029e37. [2] Kakosimos Konstantinos, Hertel Ole, Ketzel Matthias, Berkowicz R. Operational Street Pollution Model (OSPM)ea review of performed application and validation studies, and future prospects. Environ Chem 2010;7(6):485e503. [3] Weitnert Jonathan, Ogden Joan, Sperling Dan, Burke Andrew. The future of electric two-wheelers and electric vehicles in China. Energy Policy 2008;36(7):2544e55. [4] Chapman Lee. Transport and climate change: a review. J Transp Geogr 2007;15(5):354e67. [5] Greene David L, Duleep Gopal. Status and prospects of the global automotive fuel cell industry and plans for deployment of fuel cell vehicles and hydrogen refueling infrastructure. Oak Ridge: Oak Ridge National Laboratory; 2013. [6] Wirasingha SG, Schofield N, Emadi A. Plug-in hybrid electric vehicle developments in the US: trends, barriers, and economic feasibility. In: Proc. IEEE-VPPC, Harbin, China; Sep. 2008. p. 1e8. [7] Chan. The state of the art of electric, hybrid, and fuel cell vehicles. Proc IEEE 2007;95(4):704e18. [8] Moriarty Patrick, Honnery Damon. The prospects for global green car mobility. J Clean Prod 2008;16(16):1717e26. [9] Kang, Jee Eun, Brown Tim, Recker Will W, Samuelsen G Scott. Refueling hydrogen fuel cell vehicles with 68 proposed refueling stations in California: measuring deviations from daily travel patterns. Int J Hydrogen Energy 2014;39(7):3444e9. [10] European Commission. Proposal for a directive on the deployment of alternative fuels infrastructure. DOI http://

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035

12

[11] [12]

[13]

[14]

[15]

[16]

[17] [18]

[19] [20]

[21]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e1 2

eur-lex.europa.eu/legal-content/EN/TXT/?uri¼celex: 52013SC0005 (accessed 06.09.15). The Scandinavian Hydrogen Highway.http://www. scandinavianhydrogen.org/ (accessed 13.10.15). Melaina Marc, Bremson Joel. Refueling availability for alternative fuel vehicle markets: sufficient urban station coverage. Energy Policy 2008;36(8):3233e41. Rousseau A, Badin F, Redelbach M, Kim A, Da Costa A, Santini D, et al. Comparison of energy consumption and costs of different HEVs and PHEVs in European and American context. In: Proceedings of the European Electric Vehicle ConferenceeEEVC, Bru¨ssel; 2012. ~ o, Silva Carla, Brito Francisco P, Martins Jorge. Ribau Joa Analysis of four-stroke, Wankel, and microturbine based range extenders for electric vehicles. Energy Convers Manag 2012;58:120e33. Tate ED, Harpster Michael O, Savagian Peter J. The electrification of the automobile: from conventional hybrid, to plug-in hybrids, to extended-range electric vehicles. In: SAE international journal of passenger cars-electronic and electrical systems 1.2008-01-0458; 2008. p. 156e66.  elektrische auto. Opgeroepen op 04 12, Opel. (sd). Ampera. De 2014, van opel.nl: http://www.opel.nl/showroom/ampera/ default.aspx (accessed 14.09.15). Scott, P. The New BMW i3. Available online: http://www. asymcar.com/graphics/14/i3/bmwi3b.pdf (accessed 14.09.15). Mintz M, Gillette J, Mertes C, Burr Stephanie. Economic impacts associated with commercializing fuel cell electric vehicles in California: an analysis of the California Road Map using the JOBS H2 Model. No. ANL/ESD-15/1. Argonne National Laboratory (ANL); 2014. Melaina M, Eichman J. Hydrogen energy storage: Grid and transportation services. 2015. Alternative Fuels Data Center. U.S. Department of Energy. http://www.afdc.energy.gov/fuels/stations_counts.html (accessed 23.09.15). International Energy Agency. Technology Roadmap hydrogen and fuel cells. 2015. DOI, https://www.iea.org/ publications/freepublications/publication/

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29] [30] [31] [32]

TechnologyRoadmapHydrogenandFuelCells.pdf (accessed 24.09.15). Boulanger Albert G, Chu Andrew C, Maxx Suzanne, Waltz David L. Vehicle electrification: status and issues. Proc IEEE 2011;99(6):1116e38.  pez Alberto, De la Torre Nieves. Alvarez Roberto, Lo Evaluating the effect of a driver's behaviour on the range of a battery electric vehicle. Proc Institution Mech Eng Part D J Automob Eng September 2015;229(10):1379e91.   n, Alvarez Roberto, Zubelzu Sergio, Dı´az Guzma  pez Alberto. Analysis of low carbon super credit policy Lo efficiency in European Union greenhouse gas emissions. Energy 2015;82:996e1010. de la Torre N, Lopez A, Alvarez R. Battery model for lifepreserving conditions. In: Electric Vehicle Symposium and Exhibition (EVS27), 2013 World. IEEE; 2013. Aneiros Eva, Lobo D, Lopez A, Alvarez R. A proposed mathematical model for discharge curves of Li-Ion batteries. In: New Concepts in Smart Cities: Fostering Public and Private Alliances (SmartMILE), 2013 International Conference on. IEEE; 2013. Pukrushpan Jay T, Stefanopoulou Anna G, Peng Huei. Modeling and control for PEM fuel cell stack system. In: American Control Conference, 2002. Proceedings of the 2002, vol. 4. IEEE; 2002. Di Dio V, La Cascia Diego, Liga R, Miceli R. Integrated mathematical model of proton exchange membrane fuel cell stack (PEMFC) with automotive synchronous electrical power drive. In: Electrical Machines, 2008. ICEM 2008. 18th International Conference on. IEEE; 2008. Wang Chao-Yang. Fundamental models for fuel cell engineering. Chem Rev 2004;104(10):4727e66. O'Hayre R, Cha S, Colella W, Prinz FB. Fuel cell fundamentals. 2nd ed. John Wiley and Sons, Inc.; 2009. Larminie J, Lowry J. Electric vehicle technology explained. John Wiley & Sons; 2004. ix35 Fuel Cell. Realizing the dream. http://worldwide. hyundai.com/WW/Showroom/Eco/ix35-Fuel-Cell/PIP/index. html (accessed 11.11.15).

 et al., A new approach to battery powered electric vehicles: A hydrogen fuel-cell-based  ndez RA, Please cite this article in press as: Ferna range extender system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.035