CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 50 (2014) 306 - 313

The International Conference on Technologies and Materials for Renewable Energy, Environment

and Sustainability, TMREES14

PHOTOVOLTAIC ASSISTED FUEL CELL POWER SYSTEMS

A. Djafoura*, M.S. Aidab, B. Azouic

Université Kasdi Merbah Ouargla, Laboratoire LAGE, Faculté des Sciences Appliquées, Ouargla 30000, Algérie bFaculté des Sciences, Université de Constantine, Constantine 25000, Algérie cFaculté des Sciences de L'Ingénieur, Laboratoire LEB, Université de Batna, Batna 05000, Algérie

Abstract

The appearance of new concepts of decentralized electricity generation and the development of renewable sources are attracting much interest for techniques of energy storage. Renewable sources are the best candidates, but the intermittent of their production needs to find effective means of storing and protecting the environment. A system coupling a photovoltaic array and an electrolyser is used to store electricity by means of a storage form of gas.

This paper presents the results of sizing a system of hydrogen production obtained through an electrolyser, powered by photovoltaic solar modules installed in Ouargla, to meet the needs of hydrogen for a fuel cell of type, PEMFC. Sizing a photovoltaic system for a given site requires knowledge of the solar radiation of the latter, unfortunately, many localities in Algeria do not have such data or the radiations are not sufficiently representative. In this study we developed a calculation program under Matlab for determining the global radiation received by a surface inclined, in a suite we have established a flowchart that helps to size the main components of the installation to produce hydrogen by introducing the necessary technical characteristics of system components and the calculated values of the global radiation for the site of Ouargla, and electric energy needs of the user.

© 2014ElsevierLtd.Thisisanopenaccessarticle under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the Euro-Mediterranean Institute for Sustainable Development (EUMISD) Keywords: Solar energy, photovoltaic system, electrolyser, hydrogen, fuel cell, Sizing.

* Corresponding author. Tel.: +21329712627; fax: +21329712627. E-mail address: djafour.ah@univ-ouargla.dz

1876-6102 © 2014 Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Selection and peer-review under responsibility of the Euro-Mediterranean Institute for Sustainable Development (EUMISD) doi:10.1016/j.egypro.2014.06.037

1. Introduction

Algeria is well endowed with both conventional (non renewable) and non conventional (renewable) sources of energy. The largest non-renewable energy source found in Algeria is fossil (i.e. oil and gas), which is being actively exploited. Renewable sources of energy are also abundant in Algeria, the most important one being the solar. Indeed, the mean yearly sunshine duration varies from a low of 2650 h on the coastal line to 3500 h in the South, the potential of daily solar energy is important. It varies from a low average of 4.66 kWh/m2 in the north to a mean value of 7.26 kWh/m2 in the south. This means that the yearly energy potential on 80% of the territory is of the order of 2650 kWh/m2. The total daily available energy is of the order of 16.56* 1015 Wh [1]. The availability of solar energy is limited only on shiny days. Therefore, it is important to store the solar energy in other form of energy for the usage at night and gloomy weather. Hydrogen has been identified to be an ideal medium for this purpose with an important energy carrier with low harmful emissions, high - efficiency conversions into useful energy forms [2, 3]. The Photovoltaic conversion is one of the most interesting uses of solar energy modes. It provides electricity directly and independently with reliable equipment and duration of relatively high life, allowing reduced maintenance. The combination of this mode of production to the fuel cell through the production of hydrogen to be used by the fuel cell is therefore presented as an ecological mean of energy production. The objective of this work is the study and design of a hybrid photovoltaic system for electrical energy production. Mainly with the production of hydrogen needed to make worked a fuel cell of type, PEMFC, (proton exchange membrane fuel cell). In this work we have developed a mathematical model to design the main elements of the system for producing hydrogen by introducing the technical characteristics necessary as well as climate and solar data of the implantation site of the system.

2. System description

The installation is composed of a photovoltaic generator, a proton exchange membrane electrolyser and electronic interfaces between the power sources and the electrolyser, it produces hydrogen and oxygen those are stored and subsequently consumed by the fuel cell. Fig. 1 shows the system with the option of storing electrical energy in batteries for the cloudy days.

Photovoltaic Generator

^^^ Electricity f?^ Hydrogen Water

Electrolyser

Load Management &

Control Unit

Batteries

Electrical Load

H2 Storage Tank

Fuel Cell

O2 or Air

Figl.The Hybrid system (PV-PEMFC) for energy production

2.1. Photovoltaic Field

Photovoltaic generator (PV) transforms sunlight directly into electricity. It is composed by several photovoltaic modules, which are composed of solar cells. The maximum load current depends on the size and number of series -parallel modules [4].

2.2. Fuel Cell (PEMFC)

A fuel cell is a generator that directly converts the internal energy of a fuel into electrical energy using an electrochemical controlled process. Proton exchange membrane fuel cell (PEMFC) is composed of an assembly of cells which include a cathode chamber and an anode chamber separated by two electrodes and an intermediate electrolyte (proton conducting polymer) [5]. Molar gas flow consumed by the fuel cell is represented by [6]:

F = nc XI __(1)

1 gas, pac „ A _ V,1)

n x F VPpac

Where Fgas, pac : Gas flow consumed (mol / s), nc : Cells number, I : Current (A)

n : Number of moles of electrons exchanged per mole of water (n = 2 for hydrogen, n = 4 for oxygen) F : Faraday constant, (96485 C/mol), rjF pac : Faraday efficiency (%)

2.3. Electrolyser (PEME)

The principle of operation of a Proton Exchange Membrane Electrolyser (PEME) is based on the same concept as a PEM fuel cell. The water electrolysis using polymer electrolyte membrane presents the advantages of highly pure hydrogen at the output with only water and electricity at the input. This process does not require electrolyte recycling or corrosive electrolyte [2]. Molar flow of gas produced by the electrolyser is represented by [6, 7]:

Wl = x *fEI (2)

Where Fgas,El : gas flow consumed (mol / s), rjFei : Faraday efficiency (%)

2.4. Storage system

Energy storage in the autonomous photovoltaic systems is usually provided by batteries, components used in the majority of cases, [8]. The technical characteristics of storage systems can result in significant operational constraints and reduce their field of use. The coupling technologies or hybridizing with complementary properties in certain cases is necessary to circumvent the difficulties associated with the use of a single device. In our case we will consider a system with hybrid storage (storage via hydrogen and storage in the batteries).

3. SOLAR RADIATION

3.1. Site location and measurement

To determine the solar radiation on the surface of the panels, we have used the coordinates of Ouargla site (latitude 31° 52' N, longitude 5° 24' E with an altitude of 141m). Measurement data of insulation are provided from the National Office of Meteorology of Ouargla for a period of ten years of observation (2000-2009), see Table 1,[9].

Table 1. Monthly average of insulation hours for Ouargla [9]

Months Insulation(h)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 249.9 246.5 264.3 283.1 279.8 296.9 335 322.6 257.9 256.8 249.2 202.2

3.2. Model for calculation of solar radiation on a tilted surface

Using the Liu and Jordan model, the monthly average daily global radiation on a tilted surface may be computed from the following equation, [9,10, 11].

Hcp = H o*Kt * R

Where Hcp : Monthly average daily global radiation on a tilted surface, (Wh/m2. day) H o : Monthly average daily extraterrestrial radiation on a horizontal surface. (Wh/m2. day)

R : Ratio of the average beam radiation on tilted surface, Kt : Monthly-average clearness index

Fig.2 Monthly average daily global radiation in function of the tilt angle (P°), (a) January - June, (b) July - December

Fig.2 shows the results of calculating of the monthly average daily global radiation incident on a south facing surface in Ouargla with the different tilt angles with respect to the horizontal surface (-15 ° to 90 °) with step of 1°.

According to Fig. 2 it is clear that the unique optimal angle for each month of the year is the maximum points of each curve. By using El-Kassaby model [12], we have calculated the seasonal optimal angles and by using Gladius model [13], we have calculated the annual optimal angle and by using our chart we have calculated the corresponding global radiation for a south-facing sensor. See Table 2. With an optimum inclination Popt (°) annual of 39.14°, the maximum average daily radiation received in Ouargla is equal to 5.889 kWh/m2.day, which gives an average energy of 2149.6 kWh/m2.year,[9].

Table 2. Monthly average daily global radiations at monthly, seasonal and annual optimum tilt angles

Month Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

Popt(°) 62 61 52 36 17 0 -7 -4 11 29 47 59

Hcp 5271.6 6157.8 6347.8 6092.6 6454.7 6501.6 6953.1 7162.3 6777 6014.2 5994.5 6209.5

(wh/m2.day)

Season Winter Spring Summer Autumn

Popt(°) 57° 15.37° -2.2° 43.5°

Hcp 5255.4 6146.8 6326.8 5800.8 6451.9 6373.3 6889.8 7132.9 6730.3 5881.8 5986.5 6034.2

(wh/m2.day)

Popt(°) 39.14° (Annual optimum tilt angle)

Hcp 4939 5803.5 6229.5 6087.9 6152.9 5670.5 5705 6013.3 6247.3 5949.4 5953.2 5920.3

(wh/m2.day)

4. Sizing of the generator

4.1. The descriptive parameters

The most important parameters needed to know for the calculation of the necessary size of the photovoltaic generator in our system are the electrolyser consumption Ec (kWh / day) and the average daily global radiation incident on the collector or the surface of the modules Ei (kWh / m2. day). To complete the design of our generator we have based on the following assumptions:

- The molar flux produced by the electrolyser is the same as the molar flow consumed by the fuel cell.

- The demand is constant during the study period (it can change during the day, but the daily average value remains constant).

- Efficiencies of system components are constants.

4.2. Evaluation of Electric Power needs

In the present work we choose the use of an electrolyser with a power of 1.920kW to produce the hydrogen needs for the work of a fuel cell with power of 0.5 kW for 6.5 hours each night. According to the assumptions listed in the previous section we have calculated the electrolyser operating duration by using the equations (1) and (2), further to the results of the calculation, to produce the needs of hydrogen, the electrolyser may be obliged to operate for four hours each day with a consumption of 7.68 kWh/day which must be provided by the photovoltaic generator, see Table 3, for the specifications of the electrolyser and of the fuel cell.

Table 3. Specifications of the PEM electrolyser and the PEMFC Fuel cell [6]. Specifications PEM Electrolyser PEMFC cell bsc 500W

Rated power Rated power

48 V 1920 W

20V 500 W

Maximal power

Number of cells

Operating Pressure Operating current Operating temperature H2 gas production

Voltage by cell

H2 gas pressure O2 gas pressure

2000 W 26

15 bars 40 A 70°C 0.45 m3/h 1.84 V

600 W 32

25 A 60°C

0.2095 atm

4.3. Generator peak power

From the electric power needs calculated in previous section and the average daily solar radiation, the corresponding peak power of the generator (Pc) in kWc is calculated by the following equation [14].

K * Ei

Where:

Ec: Energy consumed per day (kWh / day)

Ei: The average daily global radiation corresponding to optimal modules slops (kWh / m2. day) K: Correction coefficient for systems with battery his value is between 0.55 and 0.75 [14]. The value used in our calculations (system with battery) is equal to 0.68.

4.4. Generator sizing results

The peaks powers of the generator calculated by the equation (4) for each month with an annual optimum tilt panels, 39.14 °, are shown in Fig. 3. To meet the daily energy needs throughout the year, the minimum of peak power needed to be installed is, Pc = 2.1578 kWc. In this work we choose to use the photovoltaic modules TE 1800 Q from Total Energy, see Table 4, for the specifications of these modules.

Table 4. Specifications of the PV module [19]

Model type

TE1800Q from Total Energy

Peak power, Pop Peak power voltage, Vop Peak power current, Iop Open circuit voltage, Voc Short circuit current, Isc The sizes Tolerance (%) Warranty

180 Wc 26.8V 6.8 A

33 V 7.3 A

1462mm*973mm*35mm +/-3 25 year

Fig.3 The peak power required for each month with a 39.14° tilt panels

In order to ensure the system operation over all the year, the required configuration of the photovoltaic generator is calculated as follow, [6]:

The number maximal of modules

NT = ——- = 2159.7 / 180 = 11.99 modules, (Pcmodule : Module peak power (Wc))

Pc mod ule

The number of series modules in each branch

Ns = = 48 / 26.8 = 1.79 modules, ( VL : Load voltage (V) and Vop : Module peak power voltage (V))

The number of branches

Np = = 40 / 6.8 = 5.88 modules, (IL : Load current (A) and lop : Module peak power current (A))

Then we choose to install twelve modules (six branches in parallel with two series modules in each branch), with the total area of 17.07 m2 and a peak power of 2160 Wc.

5. Estimating of the storage capacity

Determining the Battery Park (storage capacity) is made taking into account a number of days of autonomy to ensure with zero production. The number of days varies depending on the application and geographical situation [15, 16, 17].

Cs - Ec X JAut (5)

Rb x Pd xU

Where:

Cs : Storage capacity (Ah)

Ec : Energy consumed per day (kWh / day)

J Aut : Autonomy (day)

Rb : Battery efficiency (%)

Pd : Depth of discharge (%)

U : Working voltage (V)

The capacity in C10 what is necessary to store is evaluated by the next formula [18].

C10 = — (6)

FCS : Is the correction factor (factor used to deduct the capacity in C10 by correcting the calculated capacity), [18]. Fcs = 1.25 for 1 < JAut < 4.

5.1. Batteries sizing results and the choice of a regulator

The number of branches of the batteries can be, Nb > (C10/C10batterie), so for our case with an autonomy of one day we found With batteries of 6 V and 160 Ah, the number of branches, Nb > 2 branches, and for set the voltage of the storage system with the working voltage, we find the number of batteries in series, Ns = 8 batteries. See Table 5, for the specifications of batteries.

For the choice of the regulator, the first parameter to consider is the power of regulator, or the maximum current that can be controlled for a given voltage. For the voltage, the regulator will be able to supporter about the double of its rated voltage, value near the open voltage, Voc, of the panel at a low temperature, [8]. For the characteristics of the chosen regulator, see Table 6, for the regulator specifications.

Table 5. Datasheet of batteries [6]

Table 6. Datasheet of the regulator [6]

Excessive level of discharge Maximal depth of discharge (PD)

Capacity in C10 Efficiency (^b) Overcharge level

160 Ah

Maximal current of the generator Maximal power of the generator Maximal operating current Maximal operating power

Efficiency (^r)

2880 W

2160 W

6. Conclusion

Solar hydrogen system as an Energy supply is a good solution to solve fuel logistic problem for remote, desert areas in Algeria that can at least be capable of providing necessary light and energy. In this study we have showed an example of sizing of a solar generator that can make a stand-alone hybrid power generation system, mainly with the production of hydrogen needed to run a PEMFC fuel cell, for the purpose of proper management of the electrical energy produced by the PV system, we can say that the results of this design are perfectly theoretical (in the absence of an experiment). The supposition of some assumptions makes these results approximate.

As a perspective, a design based on technical criteria for a realistic case can lead to an acceptable result on the one hand the sizing of the PV array, the storage tank of gas, the storage batteries, and on the other hand, the global efficiency of the system.

[1] R. Boudries, R. Dizene. Potentialities of hydrogen production in Algeria. Int J Hydrogen Energy 2008; 33 (17): 4476 - 4487

[2] F. Barbir. PEM electrolysis for production of hydrogen from renewable energy sources. Solar Energy 78, pp.661-669, 2005.

[3] R.E. Clarke, S Giddey, F.T. Ciachi, S.P.S. Badwal, B. Paul, J. Andrews. Direct coupling of an electrolyser to a solar PV system for generating hydrogen. International Journal of Hydrogen Energy 34, pp. 2531-2542, 2009.

[4] B. Omar, A. Guen-Bouazza, B. Bouazza, T. Boussoukaia, N. E Chabane-Sari. Conception et dimensionnement d'une installation solaire autonome alimentant un poste radio téléphonique mobile de type Emetteur/Récepteur (E/R). 6ème séminaire international sur la physique énergétique, Bechar, 21- 23 Octobre, 2002.

[5] P. Stevens, F. Novel-Cattin, A. Hammou, C. Lamy M. Cassir. Pile à combustible. Technique de l'ingénieur, Doc. D3340, 10 Août 2000.

[6] H. Abdi, N. Ait Messaoudene, M. Omari, Y. Bekhta. Etude et dimensionnement d'une installation solaire de production d'hydrogène. 2ème

WIH2 Ghardaïa, 2007.

[7] Q. Ulleberg. Modeling of advanced alkaline electrolysers a system simulation approach. Int J Hydrogen Energy 28, 2003, pp. 21 - 33.

[8] A. Labouret, M. Villoz. Energie solaire photovoltaïque, Manuel du professionnel. Dunod, Paris, 2003.

[9] A. Djafour. Etude d'un système de production d'hydrogène solaire. Thèse de Doctorat, Université d'Ouargla 2013.

[10] Beckman W.A, Duffie J.A. Solar engineering of thermal processes. A wiley- interscience, publication, New York. 1980.

[11] K. Messaitfa. Evaluation de l'apport quantitatif des inclinaisons optimales d'un système de pompage photovoltaïque. Enersole 01, Adrar, Algérie, 30-31 Octobre 2001, pp255-272.

[12] M.M. Kassaby. Monthly and daily optimum tilt angle for south facing solar collectors, theoretical model, experimental and empirical correlations. Solar and Wind Technology, Vol. 5, N°6, pp. 589 -596, 1988.

[13] G. Lewis. Optimum sizing on a Collector for a Domestic Water Heating System. Solar and Wind Technology, Vol. 4, N°3, pp. 411 - 414, 1987.

[14] L. Chancelier et E. Laurent. L'Electricité Photovoltaïque. Editeur Gret, France, 1996.

[15] M. Benbitour, A. Gamma, M. belhamel. Logiciel de dimensionnement des systèmes PV. Bulletin des énergies renouvelables N°10, Décembre 2006.

[16] J. Royer et al. Le Pompage Photovoltaïque. Manuel de cours, Institut de l'Énergie des Pays ayant en commun l'usage du Français, Québec, Canada, 1998.

[17] H. Belmili, M. Ayad, E. M. Berkouk, M. Haddadi. Optimisation de dimensionnement des installations photovoltaïques autonomes -Exemples d'applications, éclairage et pompage au fil du soleil. Revue des Energies Renouvelables CICME'08 Sousse, 2008, pp. 27 - 39.

[18] O. Kebour . Etude et Analyse d'un Habitat Solaire Photovoltaïque. Mémoire de Magister, (Blida) 2004.

[19] Web site, http://www.kelvin-solartechnik.de/te.htm. (Site visited in Mai 2006).

References