ww.sciencedirect.com

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9

Available online at w

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

Fuel cell-battery hybrid powered light electricvehicle (golf cart): Influence of fuel cell on thedriving performance

Ivan Tolj, Mykhaylo V. Lototskyy*, Moegamat Wafeeq Davids,Sivakumar Pasupathi, Gerhard Swart, Bruno G. Pollet

HySA Systems Competence Centre, South African Institute for Advanced Materials Chemistry, Faculty of Natural

Sciences, University of the Western Cape, Private Bag X17, Bellville 7535, South Africa

a r t i c l e i n f o

Article history:

Received 4 May 2013

Received in revised form

11 June 2013

Accepted 17 June 2013

Available online 12 July 2013

Keywords:

Light electric vehicle

Proton exchange membrane fuel cell

Hybrid electric vehicle

Driving performance

* Corresponding author. Tel.: þ27 21 959 931E-mail addresses: mlototskyy@uwc.ac.za

0360-3199/$ e see front matter Copyright ªhttp://dx.doi.org/10.1016/j.ijhydene.2013.06.0

a b s t r a c t

A light electric vehicle (golf cart, 5 kW nominal motor power) was integrated with a

commercial 1.2 kW PEM fuel cell system, and fuelled by compressed hydrogen (two com-

posite cylinders, 6.8 L/300 bar each). Comparative driving tests in the battery and hybrid

(battery þ fuel cell) powering modes were performed. The introduction of the fuel cell was

shown to result in extending the driving range by 63e110%, when the amount of the stored

H2 fuel varied within 55e100% of the maximum capacity. The operation in the hybrid mode

resulted in more stable driving performances, as well as in the increase of the total energy

both withdrawn by the vehicle and returned to the vehicle battery during the driving.

Statistical analysis of the power patterns taken during the driving in the battery and

hybrid-powering modes showed that the latter provided stable operation in a wider power

range, including higher frequency and higher average values of the peak power.

Copyright ª 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

1. Introduction There are many publications dealing with fuel cell-battery

The use of proton exchange membrane fuel cells (PEMFC) in

transportation has recently attracted great interest worldwide

due to the fact that this type of fuel cell has a number of

attractive properties, including positive environmental

impact (the only by-products are pure water and heat), high

efficiency, low operating temperature, high power density,

fast start-up and response to load changes, simplicity, long

life, etc. PEMFCs have been successfully demonstrated in

various applications such as automobiles, scooters, bicycles,

utility vehicles, distributed power generation, backup power,

portable power, airplanes, boats and underwater vehicles [1].

4; fax: þ27 21 959 9312., mvlot@hysasystems.org2013, Hydrogen Energy P72

hybrid powered vehicles and systems. A recent review paper

by Pollet et al. [2] highlighted the current status of hybrid,

battery and fuel cell electric vehicles from electrochemical

and market points of view. The paper analyses the existent

engineering solutions, including architectures of hybrid elec-

tric drive trains. Cost, durability and energy density are shown

to be the main areas where improvements are required to

compete with conventional fossil fuels.

Some details relevant to the scope of the present study are

outlined below.

Hwang and Chang [3] developed a hybrid powered light

electric vehicle (LEV) in which the power system consisted of a

(M.V. Lototskyy).ublications, LLC. Published by Elsevier Ltd. All rights reserved.

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 en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9 10631

PEMFC (the major propulsion power) and a lithium-ion battery

pack (toprovidepeakpowerandtoallowregenerativebreaking).

Barreras et al. [4] designed and developed a multipurpose

utilityAWDelectricvehicle (two2.5kWDCmotors)withahybrid

power train based on a PEMFC and batteries. In this design, the

vehicle reached a maximum speed of 11 km/h on a flat surface

with a maximum power consumption of around 3 kW.

The development of a hybrid PEMFC and its integration

into a mini-train is described by Hsiao et al. [5]. The power

system is composed of a 200 W PEM fuel cell, four metal hy-

dride (MH) hydrogen storage tanks operating at low pressure,

a lead-acid battery pack, and an electronic energy control

system. In this configuration, the fuel cell and lead-acid bat-

teries were connected in parallel.

Jiang et al. [6] presented an experimental study on control

strategies for active power sharing in a hybrid fuel cell-battery

power source. Experimental tests were conducted with three

control objectives: maximum power, maximum fuel cell effi-

ciency, and adaptive strategy aimed to maintain a relatively

constant battery voltage, by switching between the first two

strategies.

Fisher et al. [7] integrated a lithium-ion battery pack,

together with its management system, into a hydrogen fuel

cell drive train in a lightweight city car. Electronic units were

designed to link the drive train components and to allow

system start-up, motor control and torque monitoring. The

Table 1 e Typical operating characteristics of fuel cell poweredsolutions where PEM fuel cells are used as on-board range ext

Application Peakpower [kW]

Bus voltage[V]

FC stack

Fuel cell powered electric

bicycle

0.378 No DC/DC

Up to 28

Air cooled PEMF

(0.3 kW)

Mini-train 0.560 No DC/DC

12e24

Air cooled PEMF

(0.2 kW)

Hybrid fuel cell wheelchair 0.678 28 Air cooled PEMF

(0.5 kW)

Hydrogen fuel cell hybrid

scooter

2.5 24 Air cooled PEMFC

(0.5 kW)

Lightweight fuel cell

vehicle

4.0 No DC/DC

38e56

Liquid cooled PE

(3.2 kW)

All-wheel-drive (AWD)

vehicle

5.0 48 Air cooled PEMF

(2 � 1.5 kW)

Light electric vehicle (LEV) 8.7 48 Liquid cooled PE

(6.0 kW)

Light electric vehicle (golf cart) 10e12 48 Air cooled PEMFC

(1.2 kW)

Airport electric vehicle 16 80 Air cooled PEMF

(8.5 kW)

Hydrogen-electric hybrid

urban vehicle

25 180 High temperatu

PEMFC

(3.2 kW)

EnerPac� 48.1 fuel cell

power pack for forklift

40 48 Air cooled PEMF

(8.5 kW)

Fuel cell hybrid power

source for electric forklift

50 80 Liquid cooled PE

(2 � 8.0 kW)

Fuel cell hybrid electric city

bus

150 557e670 Liquid cooled PEM

(5.0 kW)

a MH, metal hydride; CGH2, compressed hydrogen gas in cylinders.

authors concluded that the electronic integration was suc-

cessful, but the design needed optimisation and fine tuning.

Thounthong et al. [8] proposed an energy system with a

500W, 50 A PEMFC as amain power source, and battery (68 Ah,

24 V) with supercapacitor (292 F, 30 V, 500 A) as Electro-

chemical Storage Devices. During starts/stops or other sig-

nificant variations in load, the storage elements provided the

balance of energy during the load transition period, and also

absorbed excess energy from regenerative braking.

The development of a fuel cell system and its integration

into a lightweight vehicle was presented by Hwang et al. [9].

The fuel cell system consisted of a 3.2 kW PEMFC, a micro-

controller and other supporting components including com-

pressed hydrogen cylinder, blower, solenoid valve, pressure

regulator, water pump, heat exchanger and sensors. It was

found that the vehicle performed satisfactorily over a 100-km

drive on a plain route with the average speed about 18 km/h.

Measurements also showed that the fuel cell system had an

efficiency of over 30% at the power consumption required for

vehicle cruise, which is higher than that of a typical Internal

Combustion Engine.

The design, fabrication, and testing of an electric bicycle

powered by a PEMFC has been reported by Hwang et al. [10].

The system comprises a 300 W PEMFC stack, MH hydrogen

storage canisters, air pumps, solenoid valves, cooling fans,

pressure and temperature sensors, and amicrocontroller. The

vehicles. The records selected by italic correspond to theenders.

Additionalenergy storage

Hydrogenstoragea

Reference

C Battery (start-up only) MH (6.8 g H2) [10]

C Lead-acid battery (18 V) MH (50 g H2) [5]

C Li-ion battery (24 V, 14.6 Ah) MH (21 g H2) [12]

Lead-acid battery (24 V, 24 Ah) MH (54 g H2) [13]

MFC None CGH2 (125 g H2) [9]

C Lead-acid battery (48 V, 45 Ah) CGH2

(w0.29 kg H2)

[4]

MFC Li-ion battery (48 V, 40 Ah) MH (45.8 g H2) [3]

Lead-acid battery (48 V, 242 Ah) CGH2 (0.28 kg H2) This work

C Lead-acid battery (84 V, 50 Ah) CGH2 (1.34 kg H2) [14]

re Li-ion battery (80 V, 60 Ah) CGH2 (1.8 kg H2) [7]

C Ni-MH battery þ ultracapacitors

(N/A)

CGH2 (0.97 kg H2) [15]

MFC Lead-acid battery (80 V, 300 Ah)

Ultracapacitors (80 F, 97.2 V)

MH (w0.36 kg H2) [16]

FC Na-NiCl2 battery (278 V, 76 Ah) CGH2 (9.6 kg H2) [17]

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 910632

results showed that the system efficiency could reach up to

35%, and the ratio of travel distance to fuel consumption of the

prototype electric bicycle was about 1.35 km g�1 H2.

Nadal and Barbir [11] described the design and perfor-

mance of a prototype Zero Emission Electric Vehicle, powered

primarily by three 21 kW air-breathing PEMFCs using gaseous

hydrogen (w1 kg H2) as fuel. The drive train system also

included air compressor, solenoid valves, pressure regulators,

water pump, water tank, heat exchangers, sensors, pro-

grammable controller, and voltage regulator. The battery

system (10 batteries; 12 V, 1 kWh each) provided power to the

vehicle during periods of peak load, such as vehicle acceler-

ation or travelling at a high constant speed. The batteries also

provided the power to initiate fuel cell start-up. The vehicle

could travel several hundred kilometres solely on fuel cell

power with satisfactory performance.

+ -

1.2

1.1

+

-

Air exaust

+ -

1

Hydrogen suppl

1.3

1.6

AVVehicle powering system

(a)

1.1

1.2

1.4

1.3

1.6

(b)

Fig. 1 e Schematics (a) and general view (b) of the on-board fuel

fuel cell unit; 1.2, Nexa� DC1200 48 V DC/DC converter; 1.3, T70H

converter; 1.5, 48 V/18 Ah battery pack; 1.6, Nexa DC1200 hybrid

2.2, Gas cylinders (6.8 L / 300 bar each); 2.3, Reducer: M1, high-

Inline gas filter (0.5 m); 2.5, Check valve; 2.6, Shut-off valve; 2.7,

A novel hybrid fuel cell power train has been proposed by

Yang et al. [12], for a wheelchair driven by rim motors. The

power train consisted of a fuel cell, a DC/DC converter, and

two secondary battery packs, and its energy management

strategy was made to solve the problems of frequent

charging of the batteries and large power fluctuations of the

fuel cell system. The hybrid fuel cell power system was

simulated, and the simulation results showed not only that

the fuel cell was able to operate efficiently near a constant

power, but also that the charging frequency of secondary

batteries was reduced by about 98% compared to a series-

hybrid fuel cell power train.

Typical specifications and operating performances of the

fuel cell powered vehicles are summarised in Table 1; the

records are sorted in ascending order of the maximum/peak

system power (0.4e150 kW). Due to non-uniform power

2.1 2.2

Vehicle batteries48V; 242Ah

+

-

+-1.5

.4

M2 M1

2.3

y

2.4 2.5

Refuelling2.7

2.6

1.5

2.1 2.2

2.3M2 M1

2.4

2.5

2.6

2.7

cell power system: 1, Fuel cell power pack: 1.1, Nexa� 1200

FL20S02 reverse current diode; 1.4, 24 V step-down DC/DC

extension kit. 2, Hydrogen storage and supply system: 2.1,

pressure manometer; M2, low-pressure manometer; 2.4,

Refuelling connector.

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 en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9 10633

consumption during the operation of a FC-powered vehicle,

the vehicular power systems usually have a hybrid layout

with the PEMFC stack rated to the average power, while the

surplus/peak power consumption by vehicle motors, as well

as quick start-up of the system is provided by the batteries

and/or the supercapacitors.

Recently, special attention in the development of the FC-

powered hybrid vehicles was paid to the solutions where

PEMFCs are used as on-board range extenders. This approach

allows to compromise system costs with its performances,

by allowing the usage of lower power fuel cell stacks and

smaller size vehicle batteries. The corresponding approachwas

Fig. 2 e Vehicle test on the UWC campus

realised by Shang and Pollet [13] who integrated a 500W PEMFC

into commercially available lead-acid battery electric scooter

(2.5 kW in the peak power): the introduction of the fuel cell

resulted in the range extension from 8 to 15 miles, increase of

vehicle top speed and system energy efficiency. Andaloro et al.

[17] developed a hybrid power train for an urban bus (150 kW in

the peak power), by the integration of a 5 kW PEMFC. Cordner

et al. [18] showed technical and economical feasibility of the

integration of 1.2e13 kW PEMFCs as range extenders for racing

electric cars with peak power 40e100 kW. It was also reported

about activities in the development of fuel cell range extenders

for airport EV’s [19] and EV truck [20].

(top) and the driving route (bottom).

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 910634

The most important issue in the development of power

systems for fuel cell hybrid vehicles is in the optimal sizing of

the PEMFC stack (rated power) and batteries/supercapacitors

(energy storage capacity and power density). The optimisation

enables the design (i) to meet the end-user requirements for

peak and average power, (ii) to meet space constraints, and

(iii) to optimise capital and service costs.

Several attempts were undertaken to address the specified

problem,mainly,bymeansofsystemmodelling [6,8,12,16,17,21].

At the same time, quantitative information about the influence

of introductionof aquite low-power fuel cellontheperformance

of battery electric vehicles is lacking andwould be useful for the

further modelling and optimisation activities.

In this work, an existing electric golf cart (5 kW in the

nominal motor power) running on lead-acid batteries was

hybridized with a 1.2 kW PEM fuel cell stack. A comparative

study of the vehicle performances in the absence and pres-

ence of the fuel cell was carried out.

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

0

5

10

15

20

25

30

35

40(a)

Spee

d, d

ista

nce

Time [seconds]

Speed [km/h]Distance [km]

0 1000 2000 3000 4000 5000 6000 7000 8000 9000

0

10

20

30

40(b)

Speed [km/h]Distance [km]

Spee

d, D

ista

nce

Time [seconds]

Fig. 3 e Logs of vehicle speed and driving distance taken

during driving of the vehicle: (a) battery-powered, (b) hybrid

(FCD battery)-powered. The vertical dashed line in (b)

corresponds to themoment of shut-downof the FC stack (out

of H2).

2. Materials and methods

2.1. System description

The hybrid system was built around a Melex Hi-Rise 4 seater

vehicle (Melex Electrovehicles) used for everyday duties on the

campus of the University of the Western Cape (UWC). The

vehicle has a net weight of 760 kg (load up to 450 kg), overall

dimensions of 3765 � 1290 � 2060 mm, and a wheel base of

2400 mm. The propulsion is provided by ZPJ5.0-100L-4 electric

motor (5 kW/33 VAC) powered from eight US125�C lead-acid

batteries connected in series (48 VDC, 242 Ah). The operation

of the vehicle was provided by a Curtis1236 controller. The

voltage (Fig. 1a; V) and current (A) sensors at the input of the

vehicle power system allowed to monitor electric power

withdrawn by the vehicle (>0) and returned to the battery (<0)

during regenerative braking.

The life of a typical lead-acid battery is reduced by over-

voltage, deep discharge (under-voltage) and high currents

[22]. So the vehicle has its own built-in protection. The battery

discharge indicator (BDI) reports 0 when a battery voltage of

37 V is reached and prevents operation of the vehicle. This

strategy was not changed by the addition of the fuel cell as a

“charger”. The fuel cell control strategy (see below) avoids

over-voltage, and the battery life is actually expected to in-

crease. The expectation is based on the assumption that the

peak and average discharge current from the battery will be

reduced, because part of the driving power comes from the

fuel cell. Once the hydrogen supply to the fuel cell has been

depleted, the battery will be operating without the “charger”,

but this is not worse than the operation within the initial

design.

The on-board fuel cell power system was assembled in the

rear trunk of the vehicle (Fig. 1). It included a standard 1.2 kW

PEMFC power system (1), ªHeliocentris Energiesysteme

GmbH, and hydrogen storage and supply system (2).

The main component of the power system (1) was a Nexa�

1200 fuel cell unit (1.1), with output voltage 20 to 36 VDC and

maximumoutput current of 65 A (1200W rated output power).

The unit included an air cooled FCgen-1020ACS PEMFC stack

(Ballard Power Systems; 36 cells) and the necessary peripheral

components, including a controlled H2 supply valve (input

pressure 1e15 bar; flow rate up to 15 NL/min).

The fuel cell unit (1.1) was connected via a DC/DC converter

(1.2; 48 VDC output) and reverse current diode (1.3) to the 48

VDC bus of the vehicle, in parallel. The step-down DC/DC

converter (1.4) was used to supply the required voltage (24

VDC) for start-up of the fuel cell stack. Four 12 V 18 Ah lead-

acid batteries connected in series (1.5) enabled additional

storage of electric energy and provided power for system

start-up. The system operation providing extended function-

ality was controlled by the Nexa� DC1200 Hybrid Extension

Kit (1.6). The set-points for switching ON and OFF the FC

system during running the vehicle corresponded to bus volt-

ages of 50 and 52 V, respectively, or �1 V as respect to the

voltage (51 V) corresponding to 100% value of BDI set by the

supplier of the vehicle.

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 en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9 10635

The hydrogen storage and supply system (2) included two

carbon-fibre reinforced fully wrapped aluminium-lined com-

posite cylinders (2.1, 2.2; Shanghai Tianhai Dekun Composite

Cylinders Co., Ltd), internal volume 6.8 L, operating pressure

up to 300 bar. The system also included pressure reducer (2.3)

equipped with high-pressure (M1) and low-pressure (M2)

manometers, inline gas filter (2.4), check valve (2.5), shut-off

valve (2.6), and refuelling connector (2.7). Prior to the tests

using the fuel cell power system, it was refuelled with

hydrogen (99.999%) from standard gas cylinders (Air Liquide,

South Africa).

2.2. Testing procedure

The driving tests were performed at the UWC campus, on a

flat route (see Fig. 2). The tests included (a) driving the golf

cart in the battery-powered mode (FC stack was off) and (b)

driving in the hybrid (battery þ FC) powered mode. In both

cases the batteries were first fully charged (BDI ¼ 100%), and

the route was the same. When driven in the hybrid mode,

the hydrogen storage cylinders of the FC fuelling system

0 1000 2000 3000 4000 5000 6000 7000 8000 900035

40

45

50

55(a)

BDI=0

BDI=100%

Batte

ry [V

]

Time [seconds]

0 1000 2000 3000 400

36

38

40

42

44

46

48

50

52

54

BDI=0

BDI=100%

(c)

(a)

(b)

Batte

ry [V

]

Time

Fig. 4 e Logs of vehicle battery voltage taken during driving of

powered, (c) comparison of the trend lines. The horizontal dash

indicator (BDI), and the vertical dashed lines e to the moment o

were filled with H2 to a pressure of 160 bar that corresponds

to 1795 L H2 STP, or 0.16 kg H2 storage capacity. The tests

were carried out at “forced” conditions when during driving

the vehicle was periodically accelerated to 30e40 km/h fol-

lowed by braking to 0e5 km/h. The driving in both battery-

and hybrid-powering modes was carried out until discharge

of the vehicle battery to 37 V that corresponds to the zero

value of BDI set by the supplier of the vehicle. In the hybrid

mode the fuel cell unit was switched off when nearly full

consumption of hydrogen (residual H2 pressure lower than

5 bar) was reached; so the amount of the consumed H2 was

about 155 g.

The details on the calculation of hydrogen consumption by

the fuel cell unit (starting from hydrogen pressure in the

system, or from the measured output power) are presented in

the Supplementary information.

The data logging was carried out using the golf cart on-

board data acquisition system, which uses a 1314 PC Pro-

gramming Station environment (ªCurtis Instruments; model

1314-4402, version 4.3.3). For logging the PEMFC parameters,

the data acquisition software, Nexa� 1200 RC04 Control,

0 1000 2000 3000 4000 5000 6000 7000 8000 900035

40

45

50

55(b)

BDI=0

BDI=100%

Batte

ry [V

]

Time [seconds]

0 5000 6000 7000 8000 9000 [seconds]

the vehicle: (a) battery-powered, (b) hybrid (FC D battery)-

ed lines correspond to the states of battery discharge

f shut-down of the FC stack (out of H2).

-5

0

5

10

15(a)

Pow

er [k

W]

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 910636

supplied with the FC stack was used. The scan interval was

set to 1 s.

Statistical analysis of the power patterns taken in the

course of the data logging (the positive values correspond to

withdrawn power, and the negative ones to the power

returned for the battery recharge during the driving) was

carried out with the help of Origin� software (OriginLab Cor-

poration, version 6.1); the bin size was set to 300 W. Each

dataset contained merged results (about 15,000 experimental

points) of 2e3 driving tests performed in the same mode

(battery or hybrid powered).

0 1000 2000 3000 4000 5000 6000 7000 8000 9000-15

-10

Time [seconds]

0 1000 2000 3000 4000 5000 6000 7000 8000 9000-15

-10

-5

0

5

10

15 TOTAL

FC

(b)

Pow

er [k

W]

Time [seconds]

Fig. 5 e Typical power patterns taken during driving of the

vehicle: (a) battery-powered, (b) hybrid (FC D battery)-

powered. The vertical dashed line in (b) corresponds to the

moment of shut-down of the FC stack (out of H2).

Table 2 e Summary of the driving tests.

Parameter (a) Batteryonly

(b) Battery þ FC

Average speed [km/h] 16.1 15.8

Maximum distance [km] 23.2 37.9

Distance travelled during

FC operation [km]

e 28.6

Time travelled [h] 1.44 2.40

Withdrawn energy [kWh] 3.77 6.29

Returned energy [kWh] 0.34 0.79

Total energy consumed

[kWh]

3.43 5.50

Hydrogen fuel consumed

[g]

e 133e155a

a The lower value was calculated from the FC power pattern; the

higher one from the change of H2 pressure in the fuel system.

3. Results and discussion

3.1. Driving cycle

Fig. 3 presents the logs of vehicle speed and running distance

for driving the vehicle in battery-powered (a) and hybrid

(battery þ FC)-powered (b) modes. It should be noted that

in the latter mode more stable driving performances were

observed, i.e. the vehicle could be accelerated easier andmore

frequently. More detailed consideration of this effect will be

presented in Section 3.2.

A comparison of logs of vehicle battery voltage, in the

battery-powered (a) and hybrid (battery þ FC)-powered (b)

modes (Fig. 4) shows that in the latter mode the battery

discharge takes placemuch slower, and the significant voltage

drop is observed only after shutting the PEMFC stack down

when hydrogen fuel is no longer available (c).

A comparison of power patterns recorded for powering

modes (a) and (b) (Fig. 5) shows that in the hybridmode (b), the

withdrawn peak power (10e12 kW) was achieved more

frequently than in the case of powering from battery only (a).

The same tendency was observed for the returned peak

power. The contribution of the power generated by PEMFC

stack into the total power varies within 0e1.3 kW reaching the

maximum during acceleration and dropping to 0e100 W

during stops and driving with constant speed, when the ab-

solute value of the total power is low.

The driving in battery-powering (a) and hybrid-powering

(b) modes until the discharge of the battery (BDI ¼ 0) was

characterised with similar loads and driving speeds on the

same route (Fig. 2, Table 2). The operation in mode (b), even

at incomplete H2 fuel capacity (w55% of the maximum one),

resulted in >63% extension of the driving range, as well as

higher values of the energy withdrawn by the vehicle and

the energy returned to the battery. As a result, the total

energy consumed by the vehicle (withdrawn � returned)

was 61% higher for the hybrid mode that corresponded well

to the observed driving range extension. When the H2 tanks

are fully charged (P ¼ 300 bar, 3116 L STP/0.28 kg H2), the

extension of the driving range is expected to increase to

w110%.

Analysis of the data presented in Table 2 allows us to esti-

mate the contribution of the fuel cell system into the energy

balance of the vehicle in the hybrid-powering mode (b). The

introduction of the fuel cell results in the increase of the energy

withdrawnby thevehicleby (6.29� 3.77)¼2.52kWh.Theenergy

returnedto thebatterywasincreasedby (0.79�0.34)¼0.45kWh.

The observed value of the surplus energy consumed by the

vehicle in hybrid mode (Withdrawn � Returned) was equal to

2.07 kWh that, taking into account the amount of the consumed

H2 fuel (0.155 kg) and the lower heating value ofH2 (33.333 kWh/

0

500

1000

1500

2000

4

32

1

(a)

Cou

nts

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 en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9 10637

kg), corresponds to a net fuel efficiency of the FC systemof 40%.

The estimated average H2 consumption during fuel cell opera-

tion (travelling distance 28.6 km) was of 5.42 g/km.1

From the data of Table 2 it can be calculated that the

average power consumed by the vehicle varies from 2.29

(hybrid mode) to 2.38 kW (battery mode). Thus the increase of

the fuel cell power to w2.5 kW, or about 50% of the nominal

vehicle motor power, is expected to result in the possibility to

avoid interruption of the vehicle operation for battery

recharge which in this case could be provided only by the fuel

cell during the driving. It will require the increase of the

amount of the stored H2 which should be enough to provide

the operation of the higher power fuel cell during required

driving period.

-15000 -10000 -5000 0 5000 10000 15000

0

20

40

60

80

100

(b)

(a)

% o

f cou

nts

-15000 -10000 -5000 0 5000 10000 15000

Power [W]

-15000 -10000 -5000 0 5000 10000 150000

500

1000

1500

2000

4

32

1

(b)

Cou

nts

Power [W]

Fig. 6 e Power distribution histograms taken during driving

of the vehicle: (a) battery-powered, (b) hybrid (FCD battery)-

powered. The parts of the histograms correspond to

regenerative braking (1), standby (2), cruise (3) and

acceleration (4).

3.2. Statistic analysis of the power patterns

Fig. 6 presents the power distribution histograms taken

from the power patterns when operating in battery-powered

(a) and hybrid powered (b) modes. As it can be observed, the

power distribution has the following features: (i) quite large

number of counts at about zero power (standby mode of the

vehicle), and (ii) rather smooth power distribution, which

corresponds to the vehicle driving. The latter is charac-

terised by two peaks where the first one corresponds to

cruising and smooth acceleration (power > 0) and braking

(power < 0). The second peak corresponds to the forced

acceleration.

Driving in the hybrid-powering mode (Fig. 6b) is charac-

terised by the wider first peak exhibiting higher percentage of

the counts in bothwithdrawal (>0) of the power, and its return

(<0) to the vehicle battery. Note that in this case the higher

power return is caused not only by the regenerative braking,

but also by the recharge of the battery from the fuel cell that

corresponds well with the results of the driving tests (Table 2).

The integrated histograms (Fig. 7) also show higher values of

both withdrawn and returned power when operating in the

hybrid (b) mode, as compared to the battery-powered (a) one.

Also, the operation in the hybrid mode provides better forced

acceleration as seen from the second peak (Fig. 6b), which is

characterised by higher maximum and is narrower than that

of the corresponding peak for the battery-powered mode

(Fig. 6a).

Results of processing the power distribution data

(excluding standby mode) are presented in Fig. 8 and Table

3. As it can be seen, in both cases the power patterns can

be quite well fitted by two-peak Gaussian distribution

function. From Table 3, it can be noted that the hybrid mode

(b) can provide operation at the higher power, including

both “soft driving” (average power corresponding to the

centroid of the first peak about 1.8 kW, or w1.22 times

higher than for the battery-powered mode (a)), and, also,

“forced driving” (centroid at 7.8 kW that is 2.17 times higher

than the operation in mode (a)). The latter feature explains

better driving performances of the vehicle in the “forced”

Power [W]

Fig. 7 e Integrated power distribution histograms taken

during driving of the vehicle: (a) battery-powered, (b)

hybrid (FC D battery)-powered.

1 The calculations based on the FC power pattern yielded thevalue of 133 g, or 14% lower (see Supplementary information). Itcorresponds to the net fuel efficiency about 47% and the averageH2 consumption of 4.65 g/km

-15000 -10000 -5000 0 5000 10000 150000

1

2

3

4

5

6

7

8(a)

% o

f cou

nts

Power [W]

-15000 -10000 -5000 0 5000 10000 15000

0

1

2

3

4

5

6

7(b)

% o

f cou

nts

Power [W]

Fig. 8 e Gaussian fitting of the power distribution data

(Fig. 6).

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 910638

driving mode (including higher frequency and higher

average values of the peak power) when using hybrid

(FC þ battery) powering. In addition, the higher standard

deviation of the first peak for the hybrid mode allows for the

stable operation of the vehicle in a wider power range.

Table 3 e Fitting parameters of the power distributiondata (Fig. 8).

Parameter (a) Battery only (b) Battery þ FC

Value Error Value Error

Centroid (1) [W] 1470 40 1800 70

Deviation (1) [W] 1440 60 1970 70

Amplitude (1) [%] 5.7 0.2 4.8 0.1

Centroid (2) [W] 3600 500 7800 200

Deviation (2) [W] 3600 300 1700 200

Amplitude (2) [%] 1.2 0.2 1.5 0.1

Pearson correlation

coefficient, R2

0.98288 0.95383

4. Conclusions

Integration of 1.2 kW fuel cell system into light electric vehicle (5 kW

nominal motor power) results in the extension of the vehicle driving

range by 63e110%,when the amount of the storedH2 fuel varieswithin

55e100% of the maximum capacity (0.28 kg H2). The operation in the

hybrid (batteryþ fuel cell) poweringmode results inmore stable driving

performance, as well as in the increase of the total energy both with-

drawn from and returned to the vehicle battery during the driving.

On the basis of the test results, it is expected that the in-

crease of the fuel cell power tow2.5 kW, or 50% of the nominal

motor power, will enable to avoid the interruption of the

vehicle operation for battery rechargewhich in this case could

be provided only by the fuel cell during the driving.

Statistical analysis of thepowerpatterns takenduringdriving

inthebatteryandhybrid-poweringmodesshowed that the latter

provides stable operation in a wider power range, including

higher frequency and higher average values of the peak power.

Acknowledgements

The work was funded by the Department of Science and

Technology of South Africa via HySA Program, projects KP3-

S02 and KP3-S03, and supported by Impala Platinum Hold-

ings Limited (Implats), South Africa. The authors are also

grateful to Melex Electrovehicles, South Africa, for providing

the vehicle used in this study, as well as for the consultations

regarding the features of the vehicle and its control system.

Appendix A. Supplementary information

Supplementary information related to this article can be

found at http://dx.doi.org/10.1016/j.ijhydene.2013.06.072.

r e f e r e n c e s

[1] Barbir F. PEM fuel cells: theory and practice. Burlington:Elsevier/Academic Press; 2005.

[2] Pollet BG, Staffell I, Shang JL. Current status of hybrid, batteryand fuel cell electric vehicles: from electrochemistry tomarket prospects. Electrochim Acta 2012;84:235e9.

[3] Hwang JJ, ChangWR. Characteristic study on fuel cell/batteryhybrid power system on a light electric vehicle. J PowerSources 2012;207:111e9.

[4] Barreras F, et al. Design and development of a multipurposeutility AWD electric vehicle with a hybrid powertrain basedon PEM fuel cells and batteries. Int J Hydrogen Energy2012;37:15367e79.

[5] Hsiao D-R, Huang B-W, Shih N-C. Development and dynamiccharacteristics of hybrid fuel cell-powered mini-trainsystem. Int J Hydrogen Energy 2012;37:1058e66.

[6] Jiang Z, Gao L, Blackwelder MJ, Dougal RA. Design andexperimental tests of control strategies for active hybrid fuelcell/battery power sources. J Power Sources 2004;130:163e71.

[7] FisherP, Jostins J,HilmansenS,KendallK.Electronic integrationof fuel cell and battery system in novel hybrid vehicle. J PowerSources 2012;220:114e21.

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 en e r g y 3 8 ( 2 0 1 3 ) 1 0 6 3 0e1 0 6 3 9 10639

[8] Thounthong P, Rael S, Davat B. Energy management of fuelcell/battery/supercapacitor hybrid power source for vehicleapplications. J Power Sources 2009;193:376e85.

[9] Hwang JJ, Wang DY, Shih NC. Development of alightweight fuel cell vehicle. J Power Sources2005;141:108e15.

[10] Hwang JJ, Wang DY, Shih NC, Lai DY, Chen CK. Developmentof fuel-cell-powered electric bicycle. J Power Sources2004;133:223e8.

[11] Nadal M, Barbir F. Development of a hybrid fuel/cell batterypowered electric vehicle. Int J Hydrogen Energy1996;21:497e505.

[12] Yang Y-P, Guan R-M, Huang Y- M. Hybrid fuel cell powertrainfor a powered wheelchair driven by rim motors. J PowerSources 2012;212:192e204.

[13] Shang JL, Pollet BG. Hydrogen fuel cell hybrid scooter(HFCHS) with plug-in features on Birmingham campus. Int JHydrogen Energy 2010;35:12709e15.

[14] Fontela P, et al. Airport electric vehicle powered by fuel cell. JPower Sources 2007;169:184e93.

[15] Infintium Fuel Cell Systems, Inc. EnerPac� systems.Available from: http://ifcsglobal.com/products.html.

[16] Karanen TM, et al. Development of integrated fuel cell hybridpower source for electric forklift. J Power Sources2011;196:9058e68.

[17] AndaloroL,NapoliG,SergiF,DispenzaG,AntonucciV.Designofa hybrid electric fuel cell power train for an urban bus. Int JHydrogen Energy 2013;38:7725e32.

[18] Cordner M, Matian M, Offer GJ, et al. Designing, building,testing and racing a low-cost fuel cell range extender for amotorsport application. J Power Sources 2010;195:7838e48.

[19] InnovaTek funding for fuel cell range extender in airport EV.Fuel Cells Bulletin June 2011:3e4.

[20] Integration of proton motor fuel cell range extender in EVtruck. Fuel Cells Bulletin April 2012:2e3.

[21] Wishart J, Dong Z, Secanell M. Optimization of a PEM fuel cellsystem based on empirical data and a generalizedelectrochemical semi-empirical model. J Power Sources2006;161:1041e55.

[22] Power Electronics and Photovoltaic Power SystemsLaboratory, ECEN 4517/5517Lecture: lead-acid batteries.Department of Electrical and Computer Engineering,University of Colorado, http://ecee.colorado.edu/wecen4517/materials/Battery.pdf; Spring 2013.

Supplementary information to the paper FUEL CELL-BATTERY HYBRID POWERED LIGHT ELECTRIC VEHICLE (GOLF

CART): INFLUENCE OF FUEL CELL ON THE DRIVING PERFORMANCE, by Ivan Tolj, Mykhaylo V. Lototskyy, Moegamat Wafeeq Davids, Sivakumar Pasupathi,

Gerhard Swart, Bruno G. Pollet

Details on the calculation of hydrogen consumption by the fuel cell unit Hydrogen consumption by the fuel cell unit during driving the vehicle in hybrid mode was determined by two ways.

According to the first method, before starting the fuel cell, hydrogen pressure in the fuelling system was measured with the help of high-pressure manometer (M1; see Figure 1 in the main text); the ambient temperature was measured as well. The amount of H2 stored in the system was then calculated using real gas equation with five virial coefficients:

⎟⎠⎞

⎜⎝⎛ +++++=

=

44

33

221

01

;

TA

TA

TA

TAAPZ

TRZVPnH

; (1)

where nH is number of moles H2 in the system, P is hydrogen pressure [Pa], V=1.36⋅10–2 m3 is the system internal volume, R=8.3143 J/(mol K) is a universal gas constant, T [K] is the temperature and Z is H2 compressibility factor calculated from the pressure and temperature using virial coefficients A0…A4; A0=4.93482⋅10–11 Pa–1, A1=2.04036⋅10–6 Pa–1 K, A2=8.15334⋅10–5 Pa–1 K2, A3=–0.065561 Pa–1 K3, and A4=4.56516 Pa–1 K4. Accordingly, the H2 amount in the system was determined as 22.4⋅nH [L] (STP), or as 2.016⋅10–3⋅nH [kg].

Similar measurements and calculations were done at the end of the driving, after switching the fuel cell off, and the amount of the consumed H2 was calculated as the difference of H2 amounts in the system before and after the driving.

The second method was based on the calculation of flow rate of H2 consumed during the operation of the fuel cell depending on its output power. The dependence of the H2 flow rate on the stack power was determined in the course of off-board calibration of the fuel cell system whose output was connected to ZS5612 electronic load, Höcherl & Hackl GmbH. Hydrogen flow rate at the input of the system was directly measured by BRONKHORST flow meter, type EL-FLOW F-201AV-50k-AGD-22-V, range 1–50 L/min STP, accuracy ±0.5%.

Figure S1 presents time dependence (resolution of 50 milliseconds) of the measured H2 flow rate at the stack power of 1 kW. It is seen that hydrogen consumption is characterised by steady-state flow

periodically interrupted by the reproducible short purging impulses whose frequency gradually increases with the increase of the stack power. To take the purging into account, in addition to the measurements of steady-state hydrogen flow, the measurements of the total gas amount at the constant power were done as well, when the flow meter was switched to the counter mode. For each value of the output power, the average H2 flow rate was then calculated by dividing the measured total gas amount by the measurement time (10 minutes).

0 100 200 300 400 500 600 7000

10

20

30

40

50

60

FR(H

2) [L

/min

STP

]

Time [seconds]

295 296 297 298 299 300 301 302 303 304 305

10

20

30

40

50

60

Figure S1. Time dependence of H2 consumption by the fuel cell unit during its operation at the output power of 1 kW.

Both steady-state and average (with purging) hydrogen flow rates were found to be fitted well by second-order polynomial dependencies (see Figure S2). The H2 consumption calculated at the nominal stack power (1200 W) taking purging into account (Figure S2(b)) was of 13.96 L/min STP that is about 7% lower than the value specified by the FC unit supplier (15 L/min STP). The underestimation is caused by the “loss” of part of the counted hydrogen amount during the purging, when the flow rate during a short period (0.5 second) is higher than the upper measurement limit of the flow meter (see inset in Figure S1). Nevertheless, the measurement error was not too high, and the usage of a flow meter with the higher upper measurement limit, on our opinion, is inexpedient, since in this case the measurement error, especially at lower study-state flow, will significantly increase.

0 200 400 600 800 1000 12000

2

4

6

8

10

12

14

16

(b)

Y =0.269+0.00808 X+2.77232E-6 X2

R2=0.99948

(a)Y =0.069+0.00787 X+2.09375E-6 X2

R2=0.99994Y: F

R(H

2) [L

/min

STP

]

X: Power [W]

Figure S2. Polynomial fits of dependencies of H2 consumption by the fuel cell unit on the output power: (a) – steady-state H2 flow rate; (b) – average (for 10 minutes) H2 flow rate including purging impulses.

Figure S3 presents time dependencies of the flow rate calculated from the measured fuel cell power (a), total amount of the consumed H2 (b), and the distance (c) driven by the vehicle in hybrid mode; the corresponding driving logs are presented in the main text, see Figures 3(b), 4(b), and 5(b). The total amount of the consumed H2 during fuel cell operation (6531 seconds; 28.6 km travelled) was found to be equal to 1495 L STP, or 133 g that corresponds to average H2 consumption of 4.65 g/km.

The calculations based on the measurements of the hydrogen pressure of the system (P0=160 bar, T0=25oC; P1=5 bar, T1=25oC) using Eqn. (1) yielded the amounts of H2 in the system as 1795 L STP / 160 g and 60 L STP / 5 g, respectively. The difference accounts to 1735 L STP / 155 g that corresponds to the average H2 consumption of 5.42 g/km.

0 1000 2000 3000 4000 5000 60000

5

10

15

20

25

30

0

200

400

600

800

1000

1200

1400

1600FR

(H2)

[L/m

in],

Dis

tanc

e [k

m]

Time [seconds]

(c)

(b)

(a)

H2 c

onsu

med

[L]

Figure S3. Differential (a) and integral (b) consumption of H2 fuel, as well as distance travelled (c) in the hybrid driving mode.