ww.sciencedirect.com

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1

Available online at w

ScienceDirect

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

Modelization of hybrid systems with hydrogen andrenewable energy oriented to electric propulsion insailboats

Víctor Alfonsín a,*, Andres Suarez a, Angeles Cancela b, Angel Sanchez b,Rocio Maceiras b

a Defense University Center, Escuela Naval Militar, Plaza de Espa~na 2, 36920 Marín, Spainb Chemical Engineering Department, EEI, University of Vigo, 36310 Vigo, Spain

a r t i c l e i n f o

Article history:

Received 23 March 2014

Received in revised form

9 May 2014

Accepted 16 May 2014

Available online xxx

Keywords:

Sailboat

Electric

Battery

Fuel cell

Hydrogen

Electrolyzer

* Corresponding author. Tel.: þ34 986 804942E-mail addresses: valfonsin@cud.uvigo.es

Please cite this article in press as: Alfonoriented to electric propulsion in sailbj.ijhydene.2014.05.104

http://dx.doi.org/10.1016/j.ijhydene.2014.05.10360-3199/Copyright © 2014, Hydrogen Ener

a b s t r a c t

This paper presents a conceptual model of a hybrid electric sailboat in which energy from

electric grid is stored in batteries and energy from renewable energies (eolic, solar and

hydro) is stored as hydrogen. The main objective of this model is to study the viability of

electrifying traditional sailboats with internal combustion engines into hybrid systems

with batteries and fuel cell. The most important advantage of this design is the possibility

to reduce up to zero emissions of traditional sailboat. Conversion of renewable energy to

hydrogen is performed through an electrolyzer and post conversion to energy is carried out

by a fuel cell. The fuel cell with the batteries forms the hybrid system (batteries-fuel cell)

for propulsion electrical energy supply. In order to model the boat dynamic and energy

systems, modular mathematical models were developed under Matlab®-Simulink®, using a

fixed-step solver for the simulation of global model. A simulated logic controller manages

the global model. In this paper, many models have been used: some of them are based in

literature models and others were developed from experimental data. A control strategy

has also been developed to manage energy flows and then it has been embedded to Mat-

lab® language. The global model permits test the performance of the sailboat.

Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

The increase of energetic and environmental problems has

favored the development of alternative energy conversion

systems [1]. Transport machines with only internal combus-

tion engine (ICE) are being replaced by hybrid systems using

two or more power sources [2,3]. Battery systems are themost

widely used power sources due to their high efficiency and

relatively low cost [4]. On the other hand, fuel cell systems are

., valfonsin@gmail.com (V

sín V, et al., Modelizatioats, International Jou

04gy Publications, LLC. Publ

a new emerging technology that could solve environmental

problems, contribute the accomplishment of Kyoto Protocol

and besides it can help to solve the oil dependence [5e7]. Due

to similarities between batteries and fuel cell systems, their

combined effect promise great results [8].

Furthermore, in recent years the implementation of

renewable sources as photovoltaic and eolic energy in Hybrid

Renewable Energy Systems (HRES) is becoming popular for

power generation [9]. The main disadvantage of these type of

renewable energy sources is their seasonal nature, which

. Alfonsín).

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

ished 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 x x x ( 2 0 1 4 ) 1e1 12

means great variability over time [10,11]. Generation of

hydrogen based on renewable resources using electrolyzers

could become the nexus for the implementation of this tech-

nology in hybrid systems [12,13].

Plus land vehicles [14,15], it is possible to incorporate fuel

cell and battery hybrid systems in boats [16]. Hydrogen pro-

duction is performed with renewable energy provided by

several photovoltaic panels and small eolic generators located

in the own boat. Furthermore, if a sailboat is selected, this

energy production can be increased during the sail navigation

with one or more hydrogenerators.

For this type of systems, where many power sources are

mixed, mathematical simulation has become in object of

study for battery electric vehicles and all kind of hybrid con-

figurations and even specific software has been developed

[17e19]. However, there is a lack of information about battery

electric ship simulations and even less about energy renew-

able sources applied in these kinds of vehicles.

There is not any information about Energy Management

Unit (EMU) strategies in the case of HRES implemented in

sailboats. But, it is possible to find a vast amount of informa-

tion in many application areas as transportation, distributed

generation or portable applications. Some of most important

are: direct integration, single converter based or multiple

converter based [20].

In this paper, a conceptual zero emissions electric sailboat

is studied. Renewable energy sources (photovoltaic, eolic and

hydrogeneration), energy storage with batteries (in the case of

electrical energy provided by the grid) and an electrolyzer-fuel

cell system (for renewable energy) are implemented. Subse-

quently independent modular models with each energy sys-

temhas been developed inMatlab®/Simulink® and embedded.

All of these individual models are managed by a logic

controller implemented in Matlab®.

Fig. 1 e Conceptual design of a zero emission sailboat.

Conceptual design

The idea of zero emissions ships is not new, several kinds of

ships exist with some of the zero-emission characteristics and

even some of them have become commercial. Configurations

with batteries, fuel cells or both (hybrid systems) have been

developed and applied to catamarans, yachts and boats

[21e24]. A conceptual sailboat design with a hybrid battery/

fuel cell energy system has been raised. Battery system is

responsible for storing electrical energy from the grid, and

whole of fuel cell, electrolyzer and hydrogen canisters are

responsible for storing renewable energy.

Renewable generation is provided through three sources:

wind turbines, photovoltaic panels and hydrogeneration

1. Wind turbines placed to the top of the mast (only one) or

placed in the stern (one or two elements are possible), and

whose operation relies on wind power and it is operative

during the sailboat cruise or evenwhen sailboat remains in

port.

2. Photovoltaic panels, placed on ship hull, whose operation

it is possible during the ship cruise orwhen ship remains in

port, depending on the effective radiation.

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

3. Renewable energy source is hydrogeneration. One or two

marine generators are responsible for electric energy gen-

eration during sail navigation, without electric motor

operation.

Renewable energy flows to the electrolyzer where is con-

verted to hydrogen, and then it is stored in tanks or canisters.

Optimal hydrogen storage devices aremetal hydride canisters

which due to their elevated weight are perfect to be used as

boat ballast.When control decides to supply energy to system,

fuel cell device converts hydrogen to electrical energy. All

implemented devices are part of a conceptual idea for zero

emission sailboats, and one possible distribution is depicted in

Fig. 1.

Model description

Once a conceptual design for a zero emission sailboat has

been established, it is necessary to size the magnitude of each

element and even define the number of them that may be

needed. A design and analysis powerful tool for this procedure

is mathematical simulation with Matlab®-Simulink®

software.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1 3

In Fig. 2 all existing models and a scheme with their

arrangement are shown. All renewable energy sources:

photovoltaic, eolic and hydrogeneration are detailed and each

individual model system is developed through mathematical

equations or experimental data.

Furthermore, in this group the system for renewable en-

ergy storage composed by electrolyzer, fuel cell, compressor

and tank is described. Energy grid storage by batteries is

modeled in order to characterize the battery system, for this, it

is necessary to take into account the dynamic behavior and

the capacity. A propulsion model is explained; in addition to

electric motor model a dynamical model to calculate the

mechanical energy required is developed. Finally, a possible

control to manage and link all systems is shown and subse-

quently it is modeled. The connection of all of them allows a

global simulation thereby helping to scale the hybrid system

with hydrogen and renewable energy.

Renewable energy generation and storage

In this section, energy obtained from wind generator, photo-

voltaic panels and marine generators (hydrogenerators) are

modeled. The surplus of renewable energy sources power is

stored in batteries or H2 tanks based on electrolyzers used to

convert the electrical energy to the H2 energy [25].

Photovoltaic generationModeling of photovoltaic module (PV) is usually based in the

approximation of non-linear IeV curves obtained with ideal

PV circuit model, PV circuit model with series resistance or

both series and parallel resistance [26e28]. Model is based

only on the manufacturer's data and methodology used to

predict the electrical power available from PV sailboat panel is

based on the five-parameter photovoltaic model presented by

De Soto [29], who uses an equivalent circuit with series and

shunt resistances together with a diode in parallel, obtaining

IeV relationship for characteristic curve.

This model has been modified by Hadj Arab et al. [30] to es-

timate themaximumpoweroutput inoperatingconditions [31]:

Pmp ¼ GGref

Pmp;ref

�1þ g

�T� Tref

��(1)

Fig. 2 e Global mode

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

whereG is the incident irradiance, Pmp is the power output, T is

the temperature, Pmp is the maximum power, Gref and Tref are

incident irradiance and temperature refer to standard rated

conditions (SRC). By multiplying the unitary power for each

panel by the number of panels, the power being generated by

the panels at each moment is obtained.

Themain advantage of thismodel, apart from its reliability,

accuracy and simple implementation, is the easiness to obtain

valid parameters in wide operation ranges from the data pro-

vided by the manufacturer, which in many cases are under

standard nominal conditions, usually 1000 W/m2 and 25 �C.

Wind generatorFor this model, wind speed data are put into the model block

that refers to the wind generator in order to calculate the

electrical energy produced by the device from thewind kinetic

energy.

Each wind turbine has its own characteristic curve that

distinguishes it from the others, and defines its behavior for

variouswind speeds, that is, providing the power generated by

the device for each value. The start up speed for the wind

generator is low; this means that there is not energy output at

very low wind speeds.

To determine the output power of wind turbines, a poly-

nomial adjustment is used [32], which relates wind speed to

output power; this information is generally provided by eolic

turbine manufacturers. From manufacturer's data and with

Matlab® adjustment tool (cftool) a polynomial is obtained.

Once this polynomial is obtained, this is implemented as a

Matlab® function that may be subsequently called with

Simulink m-function block during simulation.

The wind speed values can be obtained with an anemom-

eter placed on a boat or even viameteorological data ofmarine

buoys. In the last case, it would be useful to subtract the for-

ward speed of sailboat.

Marine generator (hydrogenerator)This device uses sailing navigation to generate electrical

power from a propeller and a small generator [33]. When

electric motor is running the device or devices are raised up to

avoid the friction of the propeller. Procedure for marine

l with DC BUS.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

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 x x x ( 2 0 1 4 ) 1e1 14

generator model is the same than eolic generator model. The

polynomial adjustment is performed from manufacturer data

which are then implemented to Matlab-Simulink®. Commer-

cial hydrogeneration devices typically incorporate inside AC/

DC converters and the data provided by the manufacturer

usually include converter efficiency [34]. In the case that it is

not possible, efficiency value must be included in model.

Electrolyzer modelIn electrolyzer, electric current passes through a series of

electrolytic cells, where there is a water input current. Due to

the electrolysis provoked by the electric current, water current

is split into two separate currents: hydrogen and oxygen.

The currentevoltage curve chosen for the electrolyzer is

shown in Eq. (4) [35]. The number of active electrolyzers used

depends on the input power, when this exceeds maximum

nominal power for one electrolyzer, both of them are

switched on.

Ve ¼ Ve;0 þ b ln

�IeIe;0

�þ ReIe (2)

where Ve (V) is the voltage of a cell, Ve;0 (V) is the reversible

voltage for a cell, b (V�1) is a characteristic coefficient of the

electrolyzer, Ie;0 (A) is the exchange current, Re (U) is the Ohmic

resistance in the cell and Ie (A) is the current.

Electrolyzer only works for values above 15% of its rated

power although this value can be changed. A parasitic loss of

1% is assumed for each electrolyzer switched on [36].

Compression and storage of hydrogenOnce hydrogen production occurs, high pressure storage will

be necessary. For this reason, a model for a compressor must

be developed and implemented after electrolysis stage. The

compressor buffer is an intermediate tank between the gas

outlet from the electrolyzer, which works continually, and the

compressor input.

The second function has a control system for the tank

level, which, once reached a set level, activates an emptying

signal to the compressor to evacuate the hydrogen towards

the higher capacity tank. The two input variables are the

electrolyzer output flow and the flow to the compressor

creating a hydrogen balance that, by being integrated, repre-

sents the amount of hydrogen:

dNH2 ;buf

dt¼ FH2 ;buf;in � FH2 ;buf;out (3)

where NH2 ;buf is the number of accumulated moles in the

buffer, FH2 ;buf;in (mol s�1) the hydrogen flow entering from the

electrolyzer and FH2 ;buf;out (mol s�1) the output towards the

compressor. Applying a logical comparison system with

memory, a trigger signal can be sent once a particular level is

reached and a deactivation signal can be sent if a lower level is

reached.

When buffer reaches the desired level, the compressor is

triggered and compresses the hydrogen at nominal power, to

achieve a lower volume of gas and thus a smaller final storage

tank. A discontinuous flow model has been chosen, to avoid

large hydrogen flows to the compressor that would need a

much more powerful compressor.

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

Compression takes place in three stages to obtain the

maximum amount of hydrogen compressed, which is ach-

ieved when the ratio between the output and input pressures

for each compressor stage are equal and therefore the energy

consumed at each stage is the same. The molar flow of com-

pressed hydrogen is expressed as follows:

FH2 ;comp ¼ Pet;compnpoly � 1

npolyhcomp

hcomp

RT

264�

P2P1

� hpolyhpoly�1

� 1

375

(4)

where npoly is the polytropic coefficient, hcomp is the efficiency

of the compression, P1 (bar) and P2 (bar) the input and output

pressures, respectively, Pet;comp is the energy used in each

stage.

The tank is the final element in the model. The tank model

is simple, the inflow from the compressor and outflow to the

fuel cell of hydrogen are taken into account and this is inte-

grated in order to have a real value for the amount of hydrogen

stored per year. Thus, using the evolution of the amount of

hydrogen, the tank can be scaled accordingly. The balance for

the tank is the following:

dNH2 ;Tank

dt¼ FH2 ;in;Tank � FH2 ;out;Tank (5)

where: dNH2 ;Tank is the number of accumulatedmoles, FH2 ;in;Tank

(mol s�1) the hydrogen flow entering from the compressor and

FH2 ;out;Tank (mol s�1) the output towards the fuel cell.

Hydrogen consumption modelThe fuel cell works in the opposite way to the electrolyzer,

thus unregulated direct electrical current and water are ob-

tained from the combination of hydrogen and oxygen. The

input current at the anode will be the hydrogen obtained by of

electrolysis, supplied by the pressurized storage tank,

whereas the input current at the cathode will be air with

approximately 21% oxygen, reacting gas, and nitrogen, which

is inert.

In this case, a hydrogen consumption model has been

constructed rather than a fuel cell electric power model [37].

An equation to model the hydrogen intake as a function of

power demanded by the fuel cell (by the motor and the

auxiliary systems) can be used.

Thus, based on the simple reaction stoichiometry, Fara-

day's Constant, fuel cell efficiency and stack potential Vc, the

following equation is obtained [17]:

H2usage ¼Pelectric

2VcF(6)

This equation depends only on the power demanded by the

fuel cell ðPelectricÞ and stack potential Vc is referred to the lower

heating value of hydrogen, as [38]:

hfc ¼Vc

1:25(7)

It yields the hydrogen mass flux ðH2usage Þ, in moles per sec-

ond, whichmakes possible to know the hydrogen tank level at

every moment.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1 5

Grid energy storage. Battery model

Main energy source of zero sailboats is provided by grid and

electrical energy storage made with batteries while boat is in

port. The purpose of battery simulation is to predict the bat-

tery dynamic behavior versus energy demand (battery

discharge) or energy supply (battery charge). Furthermore,

battery capacity is influenced by current that flows through

battery. Magnitude of these effects is influenced by the type of

battery being used. All these factors are considered in this

section about battery model.

Battery dynamic behavior. Open circuit voltageIn this case, dynamic behavior model is performed through a

simple battery equivalent circuit (Fig. 3) where it is necessary

to know internal resistance and open circuit electric potential

(OCV) [17,39]. Assuming that current is flowing out the battery,

as is depicted in Fig. 3, basic voltage equation is [40]:

V ¼ E� IZ (8)

where V is terminal voltage, E is open circuit voltage (OCV), I is

current across circuit and Z internal resistance. OCV values

change in function of battery state of charge (SoC), and they

can be introduced in battery model from a Matlab® function.

Internal resistance values and OCV plots are obtained from

experimental data [41] or are provided from literature [42]. All

of them vary depending battery technology used (lead acid,

nickelecadmium, lithium, etc…).

Battery capacity model. Peukert coefficientBattery capacity is affected by the discharge power magni-

tude, generally when discharge is high, the capacity is lower

and vice-versa. The power magnitude is affected by battery

technology used. This effect is important for electric vehicles

due to high and discontinuous discharge currents. To

compensate this effect Peukert's equation is implemented

[43].

Some authors consider this model is non-optimal for

vehicle simulation since temperature effect is not being taken

into account and correction factor is obtained with constant

discharge current [43,44]. But this last factor in a sailboat

simulation is not a problem because current under discharge

conditions is quite constant and the temperature effect can be

considered if necessary [45], moreover the simulation

Fig. 3 e Simple equivalent circuit model composed of four

cells.

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

objective is to obtain a first estimate in order to study viability

for a zero emission sailboat.

Peukert's equation was originally developed for lead acid

batteries, but has since been used for other types [46]. Equa-

tion estimates the available capacity using the following

equation:

Cp ¼ Tdis$Ikdis (9)

where Cp is theoretical Peukert capacity (A h), Idis is discharge

current, Tdis discharge time and k the Peukert's coefficient,

which value varies depending of the kind of battery. Peukert'scoefficient can be estimated with experimental data or data

provided by manufacturer.

Sailboat propulsion

Drive resistance of a vessel over the sea (hydrodynamic

resistance) is a complex task, but Delft Series [47] shows good

results in the case of sailboats. These series have been

developed in Delft Ship Hydrodynamics Laboratory for several

different sailboat hulls. This series let to estimate mechanical

power demand and with motor efficiency can be obtained

electric power to be supplied by batteries and fuel cell system.

Mechanical power. Delft seriesGeometrical coefficients used to define the shape of a boat are:

the prismatic coefficient ðCpÞ, longitudinal center of buoyancyðLCBÞ, ratio between beam of waterline and canoe body draft

ðBwl=TcÞ, ratio of length at the waterline and displaced volume

ðLwl=Vol1=3Þ and relation between beam of waterline and

length at the waterline ðLwl=BwlÞ [47]. If these data are not

available the software (Free!Ship Plus©) can provide this

information.

Frictional resistance (RF) is obtained from sumof resistance

of hull, keel and rudder from next equation [48]:

RF ¼ 12rv2ðScCFh þ SkCFk þ SrCFrÞ (10)

where r is density ofwater, v forward speed of sailboat, S is the

wetted area at zero speed and CF is the friction coefficient for

dipped parts.

Grigson [49] and Katsui et al. [50] proposed methods to

predict friction resistance for model-through ship-scale Rey-

nolds numbers, by solving the momentum integral equation.

The 25th ITTC RC [51] conducted an analytical study on fric-

tion lines, starting to analyze the possible recommendation

for a new formula. This correctional line is shown below:

CF ¼ 0:0066577

ðlog Re� 4:3762Þ0:042612$log Reþ0:56725(11)

And Reynolds number (Re) for dipped parts is determined

separately with:

Rnh ¼ V0:7Lwl

m; Rnk ¼ VCk

my Rnc ¼ VCr

m(12)

where m is the kinematic viscosity (saltwater or freshwater)

and Lwl the waterline length.

To approximate the residual resistance, drive resistance is

included and frictional resistance is excluded, therefore wave

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Fig. 4 e Flow diagram for logic control.

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 x x x ( 2 0 1 4 ) 1e1 16

resistance is included. To obtain mathematic model several

sailboats were submitted to tank testing with different Froude

numbers:

Fn ¼ VffiffiffiffiffiffigL

p (13)

Experimental data were processed and analyzed with sta-

tistical process to obtain a polynomial for Froude number

depend on hull geometry [48]:

Rr

Vcrg¼ a0 þ a1Cp þ a2LCBþ a3

Bwl

Tcþ a4

Lwl

V1=3þ a51wlCp

2 þ a6CpLwl

V1=3

þ a7ðLCBÞ2 þ a8

�Lwl

V1=3

�2

þ a9

�Lwl

V1=3

�3

(14)

Each Delft coefficient (an) is determined for a Froude

number and with Residual resistance and frictional resis-

tance. Total boat resistance can be obtained:

Rt ¼ Rf þ Rr (15)

Effective horsepower (EHP) is calculated with total resis-

tance and the sailboat speed:

EHP ¼ Rt*V (16)

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

To obtain mechanical motor power demanded to the

electric motor, brake horsepower (BHP), it is necessary to es-

timate the propeller open water efficiency (hp):

hp ¼EHPBHP

¼ hhhohmhrr (17)

where: hh is the hull efficiency, h0 is the propeller efficiency, hm

mechanical efficiency, hrr relative rotative efficiency.

Brake horsepower (BHP) corresponds to the mechanical

power supplied by diesel generator, in the case of electric

sailboat is supplied by electric motor, therefore it is necessary

to know electric motor efficiency that lets to calculate the

electric power demand.

Electric demand. Motor efficiencyMotor efficiency can be provided by manufacturer as a single

data or a power curve, but it is possible to model this electric

motor efficiency frommotor losses. Losses are classified in four

types and can be estimated as product of corresponding con-

stant with angular speed (u) or motor torque (T) [17,52]: Copper

conductor losses ðkcT2Þ, iron losses ðkiuÞ, friction and windage

losses ðkuu3Þ, constant losses due to electronic control (C).

Knowing losses value, mechanical efficiency can be

described as:

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Fig. 5 e Sailboat global model modeled under environment Simulink©.

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1 7

hMotor¼outputpower

outputpowerþ losses¼ BHPBHPþkcT2þkiuþkuu3þC

(18)

These constants take typical values depending of kind of

motor selected: brushed DCmotor or brushlessmotor. Adding

the auxiliary elements power demand, the total electric power

demand that the batteries and/or the fuel cell must supply can

be obtained.

Energy and management control

Before to link all modeled systems, it is necessary to adapt the

output conditions of each of the previously systems to DC BUS

voltage conditions, thus a model for electrical conversion can

be developed. Once all systems are operating under the same

conditions, control can be defined and modeled. In this sec-

tion, all of these facts are described.

Electrical conversionShips are normally powered by diesel generators which are

the combination of a diesel engine with electrical generator,

converting liquid fuel (diesel) in electrical power to feed an

electrical motor. When performing a power conversion sys-

tem on a boat by a hybrid systemwith fuel cell and batteries, it

may easily be replaced by an electrical converter which

transform DC battery electrical energy to adequate form to

feed electric motor.

A simple model can be developed with electrical converter

efficiency (DC/DC or inverter DC/AC) and themotor efficiency.

Electrical energy stored is transformed to mechanical power

to move the propeller. Converters take efficiency values be-

tween 90 and 95% [53]. Inverter DC/AC efficiency value is

slightly under DC/DC converter efficiency.

Other elements of hybrid power system include indepen-

dent converter devices as fuel cell, electrolyzers or photovol-

taic panels, and efficiencymust be considered. However, other

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

systems which are modeled with experimental data or curves

provided by manufactures include this efficiency and power

value should not be modified.

Control strategy for power managementOnce all models are defined, it is necessary to determine a

control strategy. Usually, a fuel cell hybrid system works in

parallel, and in this case this structure was selected. Of all

kind of existing architectures for parallel configuration, the

multiple DC/DC converter based topology was implemented

for the power management [54]. This is the most common

energy management strategy for Fuel Cell Hybrid Power

Sources used in Fuel Cell Hybrid Vehicles, and therefore a

good first approximation to implement these systems in ma-

rine applications.

The first step is to develop a flow chart where strategy can

be evaluated. A simple scheme is depicted in Fig. 4. This

strategy is subsequently implemented as a Matlab® function

under m-file extension, which lets to embed logic control in

global Simulink® model.

Control begins with an energy balance parameter, where F

is defined as summation of energy generated with energy

demanded. If parameter F takes negative values, it means that

sailboat operates with energy of fuel cell and/or battery. If

state of charge of battery and hydrogen is higher than the

minimum, both systems operate in hybrid mode, the fuel cell

works with nominal power and the battery supplies the rest of

demand (only when it is necessary). Preference of supplying

fuel cell energy instead of battery energy is due to hydrogen

energy provides from renewable sources whose cost is zero.

If hydrogen level is lower than a minimum, the battery

supplies whole energy. On the other hand if battery state of

charge is lower than minimum, fuel cell system supplies

power demand, and if this power demand is upper than

nominal fuel cell power, sailboat will operate under safemode

(warning) to come back to port.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Fig. 6 e Effective horsepower (EHP) obtained with Freeship

software.

Table 1 e Components of battery and fuel cell hybridsystem.

Component Nominal power (W)

Motor: brushless, asynchronous (48 VDC) 10,000

Eolic generator (24 VDC) 350

Hydrogenerator (24 VDC) 1000

Solar monocrystalline panel (24 VDC) 560

PEM fuel cell (24 VDC) 2400

Electrolyzer (48 VDC) 800

LiFePO4 batteries 40 Ah (48 VDC) 624

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 x x x ( 2 0 1 4 ) 1e1 18

Another significant part of control strategy occurs when

renewable energy is upper than demand power. Two cases are

possible: In thefirst case, theboat is connected to grid inport or

during sail navigation.When it is connected to grid, the battery

will be chargedwith this kind of energy and renewable energy

(photovoltaic and eolic energy) will electrolyze hydrogen.

In the second possible case, the boat is moving but if

electric motor is running, renewable energy cannot generate

energy to electrolyze hydrogen because all energy is usedwith

electrical motor and auxiliary elements. But if boat is sailing

without electric motor, hydrogen generation can take place.

Furthermore, hydrogen generator will go into operation

increasing renewable energy generation.

Global model

Once controller is defined, the individual models can be con-

nected to each other. Energy from renewable sources is

Fig. 7 e Power management strategy from a real route sim

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

provided to a DC BUS (usually operating at a voltage of

approximately 48 V). Every one of these systems has an elec-

tric converter, but for eolic generator and hydrogenerator

parameterized curves that include this efficiency have been

used.

Electrolyzer uses excess power from DC BUS to transform

energy and water into hydrogen. Hydrogen serves to store

energy (free source) and when control decides to supply more

energy to DC BUS, fuel cell is responsible for hydrogen

reconversion into electrical energy. Battery system is bidi-

rectional and it can be charged or discharged when control

decides it.

Power demanded by electric motor is the main output

control. When boat is operating with electric motor, controller

decisions are based in this parameter and prioritize this power

demand. An electric conversion DC/DC for DC motors or DC/

AC for induction motors is required.

Finallywhen sailboat remains at port, batteries are charged

with the electricity network and renewable energy sources are

responsible to electrolyze hydrogen.

Fig. 5 shows the sailboat model once implemented under

Simulink© environment. All renewable energy sources and

energy provided by a charging station are interpreted as an

input for logic control and the power demanded by electric

motor as an energy output. Control evaluates this energy flow

and when it is necessary it will supply more energy from

battery and fuel cell.

Results

Once sailboat design is performed and individual models are

developed, some with theoretical equations and others with

experimental data, all of them are linked together with a logic

controller that will manage all systems for a predefined

strategy with the purpose of obtaining positive results.

ulation of one zero sailboat with Matlab®/Simulink®.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Fig. 8 e Batteries and hydrogen consumption for a real route simulation of one zero sailboat with Matlab®/Simulink®

depicted.

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1 9

Subsequently, a conceptual sailboat with zero emissions

can be simulated for real routes and different weather con-

ditions (wind and radiation) in order to test the performance

of the sailboat global model. This global model was tested for

motor sailboats up to 10 m in length or type A, for areas of

ocean navigation and recreational use. The maximum

cruising speed recommended for this type of electrical ships is

6 knots. For these values of speed, the balance between per-

formance and range cannot be guaranteed.

Spanish PER normative requires an autonomy of 12 miles

with a speed constant value of 6 knots [55]. Effective horse-

power (EHP) for a sailboat with 10 m in length and for a cruise

of 12 miles was estimated using Free!Ship software in 2086W.

Fig. 6 shows the EHP obtainedwith this software as function of

the sailboat speed. Consequently, and using a propeller open

water efficiency of hp ¼ 0;46(provided by Free!Ship) the brake

horsepower (BHP) is 4497W. Finally, energy required for cruise

PER normative (105 min) was estimated in 10 kWh.

Nevertheless, in most of the cases, real routes are not

under constant speed, and in sailboats electric motor is

switched off when wind is enough to propel the boat. The

proposedmodel allows to simulate real routes. An example of

one of them is shown in Fig. 7; where a real sailboat simulation

originally with ICE which is adapted to a hybrid energy system

can be seen. The main characteristics of this system are

shown in Table 1. In this simulation, sailboat sails up to 4 h

and it travels a total distance of 24 miles for various speed

ranges, and consequently, with several values for the power

demanded (electric motor and auxiliary energy). Boat sails

without motor supply, only with yachting, and it reaches high

values of total power demanded for the hybrid system. Total

energy consumed was estimated in 11.5 kWh for this travel.

Thus, this cruise characteristic is well above PER normative.

The power management is responsible of satisfying ener-

getic demand with the battery and fuel cell system. The ef-

fects of selecting an adequate control strategy are shown in

Fig. 7. In addition, the consumption of hydrogen and battery

discharge (State of Charge) for this travel can be observed in

Fig. 8. The simulation begins with battery charged with grid

energy. On the other hand, the initial amount of hydrogen

corresponds to the electrolysis of renewable energy (wind and

sun) for a week at port. In this case, when battery state of

charge reaches 50 per cent of state of charge, fuel cells supply

a part of necessary power and it is stored as hydrogen. The last

value of battery state of charge is greater than 20%, a typical

value of security to keep batteries in a good condition and a

safe value to get back at port.

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

Conclusions

In this paper, a conceptual zero emission sailboat is analyzed.

Simulink software was used as a toolbox to convert MCI sail-

boats into hybrid sailboats with batteries and fuel cells. The

hybrid sailboat is based on an electric boat whose energy

source proceed from renewable energies (eolic, solar and

marine) and energy available from recharge point, which is

provided by electricity network when boat berths at the port.

Renewable energy is stored as hydrogen obtained with an

electrolyzer and grid energy is directly stored in batteries.

Hydrogen can be reconverted in electrical energy through a

fuel cell. A logic controller is responsible of managing all

stored energy to supply electric motor demand.

This model can be used as a toolbox in order to study a

possible implementation of zero emission systems in sail-

boats. From somemanufacturer data, the simulation could be

launched, so that, with several simulations the conceptual

design of one Zero Sailboat could be optimized in order to

apply this knowledge in themanufacture of a real boat, which

could verified the normative of different countries.

For future work, this global model will allow comparing

different commercial sailboats with several profiles of hulls,

choosing the optimal number of renewable elements and

therefore the battery capacity necessary to obtain theminimal

range of the boat.

r e f e r e n c e s

[1] Sorrentino M, Pianese C, Maiorino M. An integratedmathematical tool aimed at developing highly performingand cost-effective fuel cell hybrid vehicles. J Power Sources2013;221:308e17.

[2] Li Q, Chen W, Li Y, Liu S, Huang J. Energy managementstrategy for fuel cell/battery/ultracapacitor hybrid vehiclebased on fuzzy logic. Int J Electr Power & Energy Syst2012;43:514e25.

[3] Ramos-Paja CA, Romero A, Giral R, Calvente J, Martinez-Salamero L. Mathematical analysis of hybrid topologiesefficiency for PEM fuel cell power systems design. Int J ElectrPower Energy Syst 2010;32:1049e61.

[4] Gerssen-Gondelach SJ, Faaij APC. Performance of batteriesfor electric vehicles on short and longer term. J PowerSources 2012;212:111e29.

[5] Chen P. Configuration of solar-hydrogen mild hybrid fuel cellpower systems for electric vehicles. J Power Sources2012;201:243e52.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

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 x x x ( 2 0 1 4 ) 1e1 110

[6] Anandarajah G, McDowall W, Ekins P. Decarbonising roadtransport with hydrogen and electricity: long term globaltechnology learning scenarios. Int J Hydrogen Energy2013;38:3419e32.

[7] Gonz�alez Y, Rodríguez S. A comparative study on theultrafine particle episodes induced by vehicle exhaust: acrude oil refinery and ship emissions. Atmos Res2013;120e121:43e54.

[8] Pollet BG, Staffell I, Shang JL. Current status of hybrid, batteryand fuel cell electric vehicles: from electrochemistry tomarket prospects. Electrochim Acta 2012;84:235e49.

[9] Bajpai P, Dash V. Hybrid renewable energy systems for powergeneration in stand-alone applications: a review. RenewSustain Energy Rev 2012;16:2926e39.

[10] Ringel M. Fostering the use of renewable energies in theEuropean Union: the race between feed-in tariffs and greencertificates. Renew Energy 2006;31:1e17.

[11] Lloyd B, Subbarao S. Development challenges under theclean development mechanism (CDM)dcan renewableenergy initiatives be put in place before peak oil? EnergyPolicy 2009;37:237e45.

[12] Gibson TL, Kelly NA. Optimization of solar poweredhydrogen production using photovoltaic electrolysis devices.Int J Hydrogen Energy 2008;33:5931e40.

[13] Rodríguez CR, Riso M, Jim�enez Yob G, Ottogalli R, SantaCruz R, Aisa S, et al. Analysis of the potential for hydrogenproduction in the province of C�ordoba, Argentina, from windresources. Int J Hydrogen Energy 2010;35:5952e6.

[14] Brown D, Alexander M, Brunner D, Advani SG, Prasad AK.Drive-train simulator for a fuel cell hybrid vehicle. J PowerSources 2008;183:275e81.

[15] Van Mierlo J, Van den Bossche P, Maggetto G. Models ofenergy sources for EV and HEV: fuel cells, batteries,ultracapacitors, flywheels and engine-generators. J PowerSources 2004;128:76e89.

[16] McConnell VP. Now, voyager? The increasing marine use offuel cells. Fuel Cells Bull 2010;2010:12e7.

[17] Larminie J, Lowry J. Electric vehicle technology explained.Oxford: John Wiley & Sons; 2003.

[18] Markel T, Brooker A, Hendricks T, Johnson V, Kelly K,Kramer B, et al. ADVISOR: a systems analysis tool foradvanced vehicle modeling. J Power Sources2002;110:255e66.

[19] Moore RM, Hauer KH, Friedman D, Cunningham J,Badrinarayanan P, Ramaswamy S, et al. A dynamicsimulation tool for hydrogen fuel cell vehicles. J PowerSources 2005;141:272e85.

[20] Erdinc O, Uzunoglu M. Recent trends in PEM fuel cell-powered hybrid systems: investigation of application areas,design architectures and energy management approaches.Renew Sustain Energy Rev 2010;14:2874e84.

[21] Fonseca N, Farias T, Duarte F, Gonçalves G, Pereira A. TheHIDROCAT project e an all electric ship with photovoltaicpanels and hydrogen fuel cells. World Electr Veh J 2009;3.

[22] Zero CO2 yacht sets off on first stage of mediterraneanmission. Fuel Cells Bull 2010;2010:5.

[23] Nemo H2: Amsterdam's latest canal cruise boat is poweredby hydrogen. Hol Shipbuild 2010;59:29e32.

[24] Siuru B. Sailing the deep blue gets greener. Diesel Prog NorthAm Ed 2010;76:44e5.

[25] Busby RL. Hydrogen and fuel cells: a comprehensive guide.Tulsa: PennWell Corporation; 2005.

[26] García-Valverde R, Espinosa N, Urbina A. Optimized methodfor photovoltaic-water electrolyser direct coupling. Int JHydrogen Energy 2011;36:10574e86.

[27] Ishaque K, Salam Z. An improved modeling method todetermine the model parameters of photovoltaic (PV)

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

modules using differential evolution (DE). Sol Energy2011;85:2349e59.

[28] Vengatesh RP, Rajan SE. Investigation of cloudless solarradiation with PV module employing MatlabeSimulink. SolEnergy 2011;85:1727e34.

[29] De Soto W. Improvement and validation of a model forphotovoltaic array performance; 2004.

[30] Hadj Arab A, Chenlo F, Benghanem M. Loss-of-loadprobability of photovoltaic water pumping systems. SolEnergy 2004;76:713e23.

[31] Rosell JI, Ib�a~nez M. Modelling power output in photovoltaicmodules for outdoor operating conditions. Energy ConversManag 2006;47:2424e30.

[32] Bilodeau A, Agbossou K. Control analysis of renewableenergy system with hydrogen storage for residentialapplications. J Power Sources 2006;162:757e64.

[33] Queeney T. Electricity from heavy water. Ocean NavigJanuary/February 2012;12.

[34] Watt&Sea. Technical data sheet cruising hydro generator,vol. 2013; 2013.

[35] Santarelli M, Call M, Macagno S. Design and analysis ofstand-alone hydrogen energy systems with differentrenewable sources. Int J Hydrogen Energy 2004;29:1571e86.

[36] Kelly NA, Gibson TL, Ouwerkerk DB. A solar-powered, high-efficiency hydrogen fueling system using high-pressureelectrolysis of water: design and initial results. Int JHydrogen Energy 2008;33:2747e64.

[37] Sanchez A, Cancela A, Urr�ejola S, Maceiras R, Alfonsín V.Simulation framework to evaluate the range of hydrogenfuel cell vehicles. I RE CH E 2010;2:840.

[38] Larminie J, Dicks A. Fuel cell systems explained. Oxford: JohnWiley & Sons Ltd; 2003.

[39] Roscher MA, Sauer DU. Dynamic electric behavior and open-circuit-voltage modeling of LiFePO4-based lithium ionsecondary batteries. J Power Sources 2011;196:331e6.

[40] Bird J. Electrical circuit theory and technology. Oxford: Taylor& Francis; 2007.

[41] Abu-Sharkh S, Doerffel D. Rapid test and non-linear modelcharacterisation of solid-state lithium-ion batteries. J PowerSources 2004;130:266e74.

[42] Conway BE. In: Kiehne HA, editor. Battery technologyhandbook. New York: Marcel Dekker; 1989. p. 517 [JElectroanal Chem 328 (1992) 367e368].

[43] Peukert W. Uber die Abh€anigkeit der Kapazit€at von derEntladestromst€arke bei, vol. 27; 1897. pp. 287e8.

[44] Doerffel D, Sharkh SA. A critical review of using the Peukertequation for determining the remaining capacity of lead-acidand lithium-ion batteries. J Power Sources 2006;155:395e400.

[45] Wu G, Lu R, Zhu C, Chan CC. Apply a piece-wise Peukert'sequation with temperature correction factor to NiMH batterystate of charge estimation. J Asian Electr Veh 2010;8:1419e23.

[46] Omar N, Daowd M, van den Bossche P, Hegazy O, Smekens J,Coosemans T, et al. Rechargeable energy storage systems forplug-in hybrid electric vehicles-assessment of electricalcharacteristics. Energies 2012;5:2952e88.

[47] Larsson L, Eliasson RE. Principles of yatch design. 2nd ed.London: Adlard Coles Nautical; 2000.

[48] Keuning JA, Sonnenberg UB. Approximation of thehydrodynamic forces on a sailing yatch based on the 'DelftSystematic Yatch Hull Series'. In: The international HISWAsimposium on yatch design and yatch construction; 1998.p. 99.

[49] Grigson CWB. A planar friction algorithm and its use inanalysing hull resistance. RINA 1999;76.

[50] Katsui T, Asai H, Himeno Y, Tahara Y. The proposal of a newfriction line, fifth Osaka colloquium on advanced CFDapplications to ship flow and hull form design; 2005.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 1 11

[51] Campana EF, Gorski J, Day AH, Huang D, MacFarlane G,Mikkola T, et al. Final report (The Resistance Committee).ITTC Fukuoka 2008;1:1.

[52] Auinger H. Determination and designation of the efficiencyof electrical machines. Power Eng J 1999;13:15e23.

[53] Choi W, Lee C. Photovoltaic panel integrated powerconditioning system using a high efficiency step-up DCeDCconverter. Renew Energy 2012;41:227e34.

Please cite this article in press as: Alfonsín V, et al., Modelizatioriented to electric propulsion in sailboats, International Jouj.ijhydene.2014.05.104

[54] Zhang X, Mi CC, Masrur A, Daniszewski D. Wavelet-transform-based power management of hybrid vehicles withmultiple on-board energy sources including fuel cell, batteryand ultracapacitor. J Power Sources 2008;185:1533e43.

[55] Government of Spain. Boatmaster's Certif (PER)<br/><br/>2014;2014:1.

on of hybrid systems with hydrogen and renewable energyrnal of Hydrogen Energy (2014), http://dx.doi.org/10.1016/