A New Control Approach for Hydrogen Consumption System Handling

A New Control Approach for Hydrogen Consumption System Handling

Sihem Nasri, Sami Ben Slama and Adnane Cherif

Department of Physics, Faculty of Sciences of Tunis El Manar, PB 2092, Belvedere, Tunisia [email protected]; [email protected]

Abstract: The aim of this paper is to evaluate a novel control approach for a Hydrogen Consumption System (HCS). A Proton Exchange Membrane Fuel Cell (PEMFC) is considered as a mainly source that supplies a load. An energy flow management approach is detailed in order to satisfy the load requirements despite the absence of any other power source. Finally, an analysis of the simulation results is conducted using Matlab/Simulink software in order to verify the performance of the proposed system.

Keywords: Proton Exchange Membrane Fuel Cell, Hydrogen consumption, control algorithm, efficiency.

1. Introduction

Non-polluting energy generation and other environmental issues have been driving, during the last few years, an increasing demand for new energy conversion technologies. In addition, the generation of electricity and heat nowadays is mainly done by using fossil fuels [1]. Among the different types of energy sources considered for shifting, from a fossil fuel-based energy system to a renewable-based energy system, are wind power, photovoltaic power, small hydropower plants and fuel cell stacks. The latter has received more attention in the last few years, especially because of their high electrical, overall efficiency (up to 80% for combined heat and power), low aggression to the environment, excellent dynamic response, superior reliability and durability. Fuel cells are expected to play a major role in the economy of this century and for the foreseeable future [2]. It is anticipated that the development and deployment of economical and reliable fuel cells would usher in the sustainable hydrogen age [3]. The PEMFC is a cell of choice for future automotive propulsion applications, in part because of its modest low operation temperature (<100◦C) [4]. The PEMFC system is an energy system that can convert hydrogen and oxygen (or air) to electricity with water as the only byproduct, and hence is also of great interest [5] from an environmental point of view. In fact, Hydrogen may play an important role as an energy carrier of the future and may be converted into useful forms of energy more efficiently than fossil fuels [6]. Thus, the latter may be used as fuel in almost every application where fossil fuels are being used today, but without harmful emissions, with a sole exemption of NOx emissions when hydrogen is combusted [7].

Therefore, hydrogen must be produced. The most abundant source of hydrogen is water, but water splitting requires energy, and because of the laws of thermodynamics, energy required to split water is higher than energies that can be released from producing hydrogen. Because of that, hydrogen is like electricity an energy carrier, a convenient form of energy. From a sustainability point of view, a synergy between hydrogen and electricity and renewable energy sources is particularly interesting.

Some approaches in the literatures have shown interest in analysis, modeling and control of the fuel cell. For example Bernard J. et al. [8] present a model of power train sizing of a fuel- cell hybrid vehicle including an energy storage system. Therefore, they proposed an algorithm for power management that respects a charge sustaining of the energy storage system. Kim J.H et al. [9] develop a model of a durable PEMFC Start-up process by applying a dummy Load.

Their work consists of the investigation of the effect of an application of a dummy load during the startup procedure on the degradations of a membrane electrode assembly (MEA) exposed to the 1200 repetitive start-up–shutdown cycling by using a variety of physicochemical

Received: January 9

, 2014. Accepted: September 20

, 2015

DOI: 10.15676/ijeei.2015.7.3.7

methods such as online CO₂ analysis, field-emission-scanning electron microscopy, etc. Lee F.B [10] was carrying a comparative study of fuels for on-board hydrogen production for fuel- cell-powered automobiles. He has made some combination of partial oxidation and steam- reforming of methanol technically the leading candidate for on-board generation of hydrogen for automotive propulsion.

The paper is organized as follows. Section 2 presents the description of the global system.

Section 3 describes and evaluates the system control approach. Section 4 is intended to discuss the simulation result. Finally, we conclude the paper in section 5.

2. Description and Modeling of HCS System A. Design of the global system

The system configuration is shown in Figure. 1. It is composed of three compartments which are the PEMFC system, the H₂ gas Tank Storage (HTS) and a control unit. The primary form of energy is stored as an H₂ gas form. The second energy source of the plant is the PEMFC. In any case, for a value of voltage and current desirable to supply the load, a DC-DC converter is installed in series with the PEMFC. For this reason, a control unit is taken into consideration in order to check the state of the energy flow rate in the proposed system and to regulate it. So, the control unit is serving to maintain the load requirements when there is no power source except PEMFC.

Figure 1. Design of a Hydrogen Consumption System B. H₂ gas Tank storage

The storage system considered is the compressed gas tank whose mathematical model of the hydrogen pressure can be determined from the Van der Waals equation of state for real gases [11] (see equation 1).

2. .

P

H HTS

HTS

HTS

Q R T

p  V (1)

Where p_HTS presents the tank storage pressure while the Q^P_H2, R, T_HTS and V_HTS designe respectively the inlet H₂ amount (produced), the universal gas constant (8,31 J/mol. K), the tank storage temperature and the tank volume.

C. Analysis of PEMFC system C.1 PEMFC

The PEMFC has long been known as a converter of hydrogen in energy (electrical + thermal) having a high efficiency, as proven by the comprehensive research done on this technology worldwide. The reasons are well-known: the response to the environmental pressure (clean use), to the problems arising from the centralized production of electricity, the need for having energy alternatives (hydrogen vector) and certain technological requirements

such as the applications space, underwater, portable electronic devices, power supply of isolated sites and Microsystems [12]. One of the most diffused, the PEM fuel cell has a high proton conductivity membrane as an electrolyte [13]. It must be noted that, the choice of the technology of fuel cells with the proton exchange membrane is done due to these interesting performances (weak weight, robust, solid electrolyte, fast starting, broad range of power of 1W to 10MW, etc.). Thus, it is significant to study this technology to be able to control it and extend its application. The variation of the individual cell voltage is found from the maximum cell voltage and the various voltages drops (losses). The PEMFC output voltage can be defined as [14]:

FC rev a ohm con

V E  v v v (2)

The V_FC presents the one cell fuel cell voltage while E_rev, V_a, V_ohm and V_con designee respectively the reversible voltage, the activation overvoltage, the ohmic voltage and the concentration overvoltage.

The PEMFC input current can be deduced from faraday equation (see equation.3)

reac

FC 2

reac

reac 2 F

2

cell

I f ( Q ( H ))

Q ( H )* 2* F * where f ( Q ( H ))

N



 

 

  (3)

Where I_FC and Q^reac(H₂) designe respectively the fuel cell current and the amount of H₂ gas consumed by the PEMFC while F, η_F, N_cell present the faraday constant (96485 mol^-1.K^-1 ), the faraday efficiency (98%) and the PEMFC cell numbers respectively.

The referential hydrogen consumption flux rate can be derived from the value of load current (I_Load). It can be expressed as:

ref Load cell F

2

I .N . Q ( H )

2.F

  (4)

The Q^ref(H2) presents the referential H2 gas amount needed to satisfy the load requirement.

C.2 Boost Converter

The boost circuits produce the output voltage by charging an input inductor with current, from an input voltage source, then discharging the inductor into an output capacitor. In our work, the DC-DC boost plays the role of a power interface that maximizes the power transfer between the PEMFC and the load. For this reason, it can be controlled with Maximum Power Point Tracking (MPPT) whose objective is to make the PEMFC run at an operating point that corresponds to the maximum power (MPP). The following equation describes the relation between the duty cycle αB and respectively the voltage V_Boost and the current I_Boost which present the principal output parameters of DC boost.

 





1 1

1

FC Boost

B

Boost B FC

B

V V

I I

where







 

 

  



  







(5)

C.3 MPPT Controller

The Maximum Power Point Tracking (MPPT) techniques are necessary in Fuel cells applications because the MPP of flow rates of H₂ and O₂ varies. Thus, the use of MPPT control technique strategy is required in order to obtain the maximum power [15]. In order to improve the performance of the HCS system, the converter, known as MPPT, is used to mach continuously the output characteristics of the fuel cell generator. The incremental conductance

method for MPPT is used here. In incremental conductance method the array terminal voltage is always adjusted according to the MPP voltage. So, it is based on the incremental and instantaneous conductance of the PEMFC module [16]. When the generator voltage is operating at voltage V_Ø and current I_Ø, the power P_Ø is equal to (V_Ø* I_Ø). Thus, the MPPT process returns the desired V_Ø for the boost converter and is regulated to desired V_β. The latter is given by the following equation:

FC j

V

 V   V

(6)

The basic equations of this method are given as follows:

dI I

dV V

 

 ^{ }  ^{: At MPPT (7)}

dI I

dV V

 

 ^{ }  : On the left of MPP (8)

dI I

dV V



 

   : On the right of MPP (9)

It must be noted that the increment or the decrease in the Vφ is determined by judging the sign of equation 23.

( * )

1 1

*

dP d V I dI

V dV V dV V dV

   

     

   (10)

Figure 2. Incremental Conductance MPPT Method 3. Control Unit

In this sub-section, we develop and evaluate the HCS control approach. The latter is the main area that ensures a correct operation between the PEMFC and the load. For this reason, we can liberate three scenarios.

A. First scenario

The aim of this scenario is to estimate the amount of the hydrogen used by the PEMFC taking into consideration the amount of H₂ stored in the HTS. The referential amount of H₂ is defined as the needed H₂ quantity that one converted into electricity it satisfies the load requirements. So, this quantity can be a function of the I_Load (see equation.4). This amount can

be compared to the Q^S_{max and} Q^S_min which define respectively the maximum and the minimum amount of H2 that can be stored in the HTS system.

In addition the concerning scenario can be divided in four assumptions.

First assumption

This assumption is intended to control the state of the HTS amount.

HTS is full Q^S (H₂) = Q^S_max

For Q^ref ≥ Q^Smax  Q^reac(H₂)=Q^S_max-Q^S_min; For Q^ref < Q^S_max  Q^diff(H₂)=Q^S_max-Q^ref; Second assumption

For Q^diff< Q^S_min  Q^reac(H₂)=Q^ref(H₂)

For Q^diff > Q^S_min q_sup= Q^diff-Q^S_min Q^reac(H₂)=Q^ref(H₂) - q_sup;

Third assumption

The HTS is not full Q^S (H₂) < Q^S_max For Q^ref ≥ Q^S(H₂)  Q^reac (H₂)=Q^S(H₂)-Q^S_min For Q^ref < Q^S(H₂)  Q^diff (H₂)= Q^S (H₂)- Q^ref Forth assumption

For Q^diff< Q^S_min Q^reac(H₂)=Q^ref(H₂)

For Q^diff> Q^S_min  q_sup= Q^diff- Q^S_min Q^reac(H₂)=Q^ref(H₂)- q_sup

When the HTS is empty  Q^S (H₂)=Q^S_min  we need another power source to supply the load.

Figure 2. Flow chart describing the control of hydrogen amount used by the PEMFC

B. Second scenario

The main objective of this scenario is to estimate the PEMFC current using the relation between it and the amount of hydrogen reacted and to command the MPPT bloc controller as follows:

 Activate it when the IFC does not satisfy the load demand.

 Deactivate it when the PEMFC current is equal to the load one (I_FC= I_Load).

Figure 3. Flow chart describing the MPPT controller activation C. Third scenario

The aim of this scenario to command the boost converter with the recommended duty cycle that ensures the load supply. In that case, we compare the duty cycle obtained from MPPT controller (α_mppt) and the critical one (α_C). The latter is calculated from the relation between the load current and the input boost current (I_FC). The boost converter is commanded with a duty cycle (α_B) as follows:

 α_B=α_mppt for α_mppt= α_C

 αB=αC for α_mppt ≠ αC

Figure 4. Flow chart of recommended boost duty cycle

4. Simulation and results

The expected objective of this work is to reach an effective prediction of the consumed hydrogen amount in order to adjust the operation of the fuel cell in nominal terms by using the static DC-DC converter which must be controlled by an algorithm MPPT.

In the literature there is many researches working in the fuel cell control field that we can mentioned some like that presented in [17], which gives a fuel cell strategy control oriented ti hybrid vehicular application in association with supercapacitor bank or that given by [18], which presents an adaptive MPPT controller dedicated to manipulate the duty cycle of DC-DC converter in fuel cell system and even the work given in [19] which presents a strategy of MPPT controller algorithm dedicated to fuel cell power generation.

Among all this cited researches, we can identify our work as a supplement of the given ones. It is dedicated to control the fuel consumption (H₂ gas) to achieve the proper functioning of fuel cell system device that can be intergraded either with a solar energy for hybrid system implementing oriented to remote area application or with a supercapacitor bank to model a hybrid electric vehicle. For that, to solve the problem related to fuel consumption, we have given a strategy control for MPPT adjustment based on the fluctuate amount H₂ gas consumption.

Thus, we have only treated the part describing the characteristics of the proton membrane exchange fuel cell without highlighting its integration with a whole hybrid system in order to detail the operation mode of fuel cell system and its effects in the variation of the duty cycle dedicated control the boost converter functioning.

Hence, the proposed Hydrogen Consumption System (HCS) is modeled and simulated using Matlab/Simulink environment. Indeed, the presented Figure below (5,6,7,8 and 9) give the simulation results from what we can prove the performance, the reliability and the effectiveness of the proposed HCS handling strategy.

Figure 5. PEMFC Voltage and Efficiency Variation

The figure 5 gives the PEMFC voltage and efficiency in a period of simulation time of 1100 ms. As shown, we can remark that the fuel cell efficiency reaches at maximum 75%.

Additionally, the figure below [6..9] present respectively the variation of the prediction of the H₂ gas amount which can be used by the PEMFC to satisfy the load requirement, the variation of the HTS state of charge, the PEMFC and the boost currents fluctuation as well as the duty cycle variation given by the adopted MPPT controller in the same time period.

Figure 6. Amount of Hydrogen variation

Figure 7. Evolution of H₂ gas State of Charge in the Tank Storage

Figure 8. Current variation of the HCS

Figure 9. Duty Cycle variation and MPPT activation

In the time interval [0 ms-100 ms] (zone A), we can note that the HTS system is full.

However, the referential amount of H₂ (Q^ref) is greater than stored one (Q^S). At this moment, the HTS should keep always its critical minimum value (Q^S_min). For that, the quantity of H₂ which can be reacted will be lower than the referential amout and the stored one.

Consequently, the fuel cell current value is lower than the load one (ILoad). For that, to compensate this problem, the boost converter is intervening to cover the deficit of power. For this reason, the MPPT controller is activated switching the enable input (E_B) to 1. In this case the boost input duty cycle (αB) is the same as the MPPT output one (αMPPT) . We have to note that when the critical functioning point of the boost converter is set when the duty cycle is 0.5.

Hence, when αMPPT≠αC and αC<0.5, the boost converter is operated with the value of αMPPT. In the time interval [320 ms-420 ms](zone B), we remark that the HTS is full. In this case, the Q^ref(H₂) is lower than Q^S. Additionnaly, the HTS keeps its minimum value Q^S_min explained by Q^S>Q^reac. Consequently, the Q^reac quantity of H₂ is equal to the Q^ref one and the current I_FC reaches the value of the load one without boost intervening which verifies the deactivation the enable input of the MPPT controller. The boost current I_Boost, at this case, keeps the same value as the I_FC one.

In the time interval [550 ms-650 ms]( (zone C), we treat the case when the HTS is not full.

So, we can see that in this period of time the Q^ref quantity is lower than the Q^S one. Hence, after checking the critical state of the tank storage (Q^ref>Q^S_min), the tank is able to deliver the required quantity necessary to supply the load in the normal condition. For that, the Q^reac is equal to Q^ref and system didn’t need the intervening of the boost converter to regulate the current state which explains the deactivation of the MPPT controller (E_B=0).

Finally, we can deduce that thanks to the adpted control approach , our Hydrogen Consumption System (HCS) will be sustainable, proper and effective to be integrated with other energy sources to model a hybrid power system oriented to diverse applications like transportation, buildings and remote area.

5. Conclusion

In this paper, we have analyzed and evaluated a model of HCS system which is mainly composed of a PEMFC. We have proposed a new control approach for energy flow management which is based on the current values and the amount of hydrogen. The applied strategy guarantees the load supply when the PEMFC is taken as the only power source of the proposed system. Finally, we have evaluated our contribution through simulations and we have shown the effectiveness and the toughness of the proposed approach.

Our future work includes the incorporation of the studied system with a renewable energy source as well as the elaboration of a new supervisory approach.

Nomenclature

V_FC: PEMFC Voltage (V) I_FC: PEMFC Current (A)

V_Boost: Output boost converter Voltage (V) I_Boost: Output boost converter Current (A) E_rev: Reversible Voltage (V) I_Load: Load Current (A)

V_act: Activation overpotential (V) Q^reac: H₂ amount consumed by PEMFC (mol) V_con: Concentration overpotential (V) Q^ref: H₂ referential amount (mol)

V_ohm: Ohmic overpotential (V) Q^S: H₂ stored amount (mol)

αB: Input Boost duty cycle Q^S_min: H₂ minimum storage capacity (mol) αMPPT: Output MPPT controller duty cycle Q^S_max: H₂ maximum storage capacity (mol) αC: Critical duty cycle Q^diff: H₂ differential amount (mol)

E_B: Input MPPT controller activation Q^P_H2: H₂ produced amount (mol)

F: Faraday constant R: Real gas constant

N_cell: Number of PEMFC cells p_HTS: HTS pressure (bar) THTS: HTS input temperature (K) VHTS: HTS input volume (Liter) ηF: Faraday efficiency q_sup: H₂ surplus quantity (mol) 6. References

[1]. B.S. Sami, B.C. Abderrahmen and C. Adnane, “A MIMO State Space Non Linear Modelling of a PEMFuel Cell with a DC/DC Boost Converter”, International Review on Modelling & Simulations, Vol. 5 Issue 2, p1009, 2012.

[2]. M.A.J. Cropper, S. Geiger, M.J. David,” Fuel cells: a survey of current developments”, Journal of Power Sources Vol. 131 pp. 57–61, 2004.

[3]. DS. Scott, “Until Something Better Comes Along”, International Journal of Hydrogen Energy, Vol. 29 pp. 1439–1442, 2004.

[4]. S. Gottesfeld, T. Zawodzinki, “Polymer electrolytes fuels cells”, Adv Electrochem Sci Eng vol.5, pp.195–301, 1997.

[5]. B.C.H. Steele and A. Heinzel, Materials for fuel cell technologies, Nature Vol. 414, pp.

345:352, 2001.

[6]. T.N. Veziroglu and F. Barbir, Hydrogen Energy Technologies, UNIDO-Emerging Technologies Series, United Nations Industrial Development Organisation, Vienna, Austria, 1998.

[7]. F. Barbir and T.N. Veziroglu, “Environmental benefits of the solar hydrogen energy system”, Environmental Issues and Management of Waste in Energy and Mineral Production, pp. 1209–1218, 1992.

[8]. J. Bernard , S. Delprat, F.N. Buchi and T.M. Guerra, “Fuel-Cell Hybrid Powertrain:

Toward Minimization of Hydrogen Consumption”, Vehicular Technology, IEEE Transactions on Vol.58, Issue. 7, pp. 3168 – 3176, 2009.

[9]. J.H. Kim, E.A. Cho, J.H. Jang, H.J. Kim, T.H. Lim, I.H. Oh, J.J. Ko, and I.J. Son,

“Development of a Durable PEMFCStart-Up Process by Applying a Dummy Load”, J.

Electrochem. Soc, Vol.157: pp 118-124, 2010.

[10]. F.B. Lee, “A comparative study of fuels for on-board hydrogen production for fuel-cell- powered automobiles”, International Journal of Hydrogen Energy Vol.26 pp.381–397, 2001.

[11]. P. Pfeifer , C. Wall , O. Jensen, H. Hahn and M. Fichtner , “Thermal coupling of a high temperature PEM fuel cell with a complex hydride tank”, International Journal of Hydrogen Energy, Vol. 4, Issue 8, pp. 3457–3466, 2009.

[12]. B.S. Sami, B.C. Abderrahmen and C. Adnane, “Efficient Design of a Hybrid (PV-FC) Water Pumping System with Separate MPPT Control Algorithm”, International Journal of Computer Science and Network Security (JCSNS), Vol.12 No.1, January 2012.

[13]. M. Khemariya and A. Mittal, “Modeling and Simulation of different components of a stand-alone Photovoltaic and PEM Fuel Cell Hybrid System”, International Journal on Emerging Technologies pp. 60-67, 2010.

[14]. R. Seyezhai and B.L. Mathur, ‘‘Mathematical Modeling of Proton Exchange Membrane Fuel Cell’’, International Journal of Computer Applications, Vol.20, No.5, pp. 0975 – 8887, April 2011.

[15]. A. Saadi et A. Moussi, ‘Etude de l’Effet des Fluctuations de Température sur les Techniques d’Optimisation des Systèmes de Pompage Photovoltaïque’, CEE’2002, Batna, Algeria, pp. 232 - 237, Octobre 2002.

[16]. M. Lokanadham and K.Vijaya Bhaskar, ‘‘Incremental Conductance Based Maximum Power Point Tracking (MPPT) for Photovoltaic System’’, International Journal of Engineering Research and Applications (IJERA), Vol. 2, Issue 2, pp.1420-1424, 2012.

[17]. A. Mansour , B. Faouzi, G. Jamel and E. Ismahen, ‘‘Design and analysis of a high frequency DC-DC converters for fuel cell and super-capacitor used in electrical vehicle’’, International Journal of Hydrogen Energy Vol.39 pp.1580–1592, 2014.

[18]. J. Junsheng and C. Xueying, ‘‘Adaptive Control of MPPT for Fuel Cell Power System’’, Journal of Convergence Information Technology(JCIT), Vol. 8, No. 4, 2013.

[19]. J. Junsheng and C. Xueying, ‘‘A Real-time Tracking Control of Fuel Cell Power Systems for Maximum Power Point’’, Journal of Computational Information Systems Vol. 9, No.

5, pp.1933–194, 2013.

Sihem Nasri was born in Tunis, Tunisia, in 1986. She received the Master’s degree in Electronics from the Faculty of Sciences of Tunis (FST) in 2011.

Currently, she is pursuing the Ph.D. degree in Electronics with the Faculty of Sciences of Tunis, in the laboratory of Innovation of communicant and cooperative mobiles (Innov’Com), the Higher School of Communication of Tunis (SUPCOM). Her research interests include electrical power systems integrating energy storage devices and power system management.

Sami Ben Slama received the engineer, master and doctorate degrees, in electronics from Faculty of sciences of Tunis (FST), respectively in 2005, 2009 and 2014. He is assistant professor in King Abdul-Aziz University, Jeddah Saudi Arabia. He field of interest concerns the photovoltaic power, energy system, and Matlab modeling.

Adnane Cherif received the engineer, master and doctorate degrees from National Engineering School of Tunis (ENIT), in Tunisia. He is a university teacher in Electronics at Faculty of Sciences of Tunis (FST). He is responsible of the master specialty communication systems and networks.

His field of interest concerns photovoltaic power system, digital signal processing, energy system designing and modeling.