DC/DC and DC/AC Converters Control for Hybrid Electric Vehicles Energy Management-Ultracapacitors and Fuel Cell

Transcription

1 686 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013 DC/DC and DC/AC Converters Control for Hybrid Electric Vehicles Energy Management-Ultracapacitors and Fuel Cell Abdallah Tani, Mamadou Baïlo Camara, Member, IEEE, Brayima Dakyo, Member, IEEE, and Yacine Azzouz Abstract This paper presents the ultracapacitors and the fuel cell (FC) connection for hybrid electric vehicles (HEVs) applications. An original method for the embedded energy management is proposed. This method is used to share the energetic request of the HEV between the ultracapacitors and the FC. The ultracapacitors are linked to dc-bus through the buck-boost converter, and the FC is connected to dc-bus via a boost converter. An asynchronous machine is used like traction motor or generator, and it is connected to dc-bus through an inverter. A dc-motor is used to drive the asynchronous machine during the decelerations and the braking operations. The main contribution of this paper is focused on the embedded energy management based on the new European drive cycle (NEDC), using polynomial control technique. The performances of the proposed control method are evaluated through some simulations and the experimental tests dedicated to HEVs applications. Index Terms Asynchronous machine, current control method, dc-bus voltage control, dc/dc converters, energy management, fuel cell (FC), hybrid electric vehicles (HEVs), polynomial control, speed control, ultracapacitors. I. INTRODUCTION T HE energy consumption and an increasing of the oil cost in the world followed by a depletion of the fuels are justifiable reasons to use the hybrid electrical vehicles (HEVs) instead of thermal vehicles [1]. The thermal vehicles have high-exhaust emissions, low-fuel efficiency, and high operating noise, but they have the autonomy in the use. The HEVs are characterized by a low-exhaust emission, a low-operating noise, and the reasonable energy efficiency, but they don t have the autonomy in the use [2]. Considering the energetic autonomy problem, an association of the fuel cell (FC) and the ultracapacitors is proposed to improve the HEVs energetic performances. This com- Manuscript received September 20, 2011; revised January 21, 2012, April 09, 2012, July 08, 2012, and September 24, 2012; accepted October 10, Date of publication October 18, 2012; date of current version January 09, Paper no. TII A. Tani, M. B. Camara, and B. Dakyo are with GREAH Laboratory, University of Havre, Le Havre, France ( camaram@univ-lehavre.fr, brayima.dakyo@univ-lehavre.fr). Y. Azzouz is with IRSEEM Laboratory École d ingénieurs ESIGELEC, Saint-Etienne du Rouvray, France. Color versions of one or more of the figures in this paper are available online at Digital Object Identifier /TII Fig. 1. Electric hybrid system configuration. bination is due to physical characteristics of the FC and the ultracapacitors ( ). Contrary to the batteries, the FCs provide theelectricenergy,rather than storing it, and they continue to deliver the energy, as long as the hydrogen supply is maintained [3], [4]. However, the FCs present some well-known technical limitations: they havealowefficiency during the low power demand, a high cost per watt, and a slow power transfer during the transient operations. For these reasons, the FCs are not generally used in the HEVs to meet the load demand during the start up, and the transient operations [1], [5]. Furthermore, an association of the FC and the ultracapacitors enables to solve these problems as proposed in [6] and [7]. The HEVs equipped with the FC and the ultracapacitors present the following advantages, due to ultracapacitors dynamic behavior [8]: the FC is less solicited during the transient operations. The life time and the size of the FC are improved. The energetic autonomy and the regenerative braking efficiency of the HEVs are improved. The studied system is illustrated in Fig. 1, where the hybrid sources (ultracapacitors/fc) are linked to dc-bus through the dc/dc converters. An inverter is used to drive the asynchronous machine during the traction operations. During the deceleration and the braking operations, the previous asynchronous machine is controlled by adc-motor for the energy recovery by the ultracapacitors. The contribution of this paper is focused on the bidirectional load (motor and generator operations) power sharing between the FC and the ultracapacitors, using the new European drive cycle (NEDC) and the polynomial control technique. The proposed method consists to allocate the average power to the FC and the fluctuating power (due to the acceleration, the deceleration, and the braking operations) to the ultracapacitors. For the hybrid system behavior simulations, the /$ IEEE

2 TANI et al.: DC/DC AND DC/AC CONVERTERS CONTROL FOR HEVS ENERGY MANAGEMENT-ULTRACAPACITORS AND FC 687 TABLE I ULTRACAPACITORS MODULE PARAMETERS Fig. 2. Dynamic model of the ultracapacitors. MATLAB SIMULINK software is used. To validate the proposed method for the energy management, an experimental test bench is carried out in the reduced scale. This system includes a module of the ultracapacitors, a FC emulator, and the electric machines (ac-motor/generator with dc-motor) as a reversible load. The ultracapacitors module is linked to the dc-bus via a buck-boost converter which ensures the energetic exchange between the ultracapacitors and the load. The FC emulator is connected to the dc-bus using a boost converter for the dc-bus voltage management. The control of these converters depends on the energy management method between the hybrid sources (ultracapacitors/fc) and the requested or supplied energy by the asynchronous machine. II. ULTRACAPACITORS AND FC MODELING A. Ultracapacitors Modeling To use the ultracapacitors ( ) as energy storage devices in the hybrid electric vehicles (HEVs), it is necessary to associate several cells in series to obtain a high voltage level due to ultracapacitor cell voltage of 2.7 V. The used model of the is presented in Fig. 2. This model includes an internal resistance, and an equivalent capacitor. This capacitor includes two components, the first component varies linearly with the ultracapacitors module voltage, and the second component is a constant capacitor [9], [10]. The analytical model of the module is presented in (1), where defines the sign of the current. In this equation, if the ultracapacitors are in the discharge operations, and for the charge operations. The estimation method of the used parameters is presented in [10] This model describes the dynamic behavior of the ultracapacitors during the charge and the discharge operations [8]. The estimated parameters for the BOOST-CAP3000F module are presented in Table I. To validate the model of the ultracapacitors, a module of eight cells in series is realized. This module is charged into maximum voltage of 22 V, and discharged with a constant current. The simulation and experimental results obtained for are compared in Fig. 3. (1) Fig. 3. Ultracapacitors model validation. B. FC Modeling The FCs are the electrochemical devices that directly convert the chemical energy of the Fuel into electricity. The energy is released whenever the hydrogen reacts chemically with the oxygen of the air [11]. In the case of the proton exchange membrane FCs (PEMFCs), which are the focus of the most research activities today, the only byproduct is the water and the heat. The PEMFCs technologies are the best candidate among other FC technologies, due to the low operating temperature, the small size, the lightweight, the high-power density, and the relatively short time start-up [12], [13]. In this paper, the PEMFC model is used, and the operations of this one is based on the following electrochemical reaction Electrical Energy+Heat (2) Many models of the FC can be founded in the literature such as in [13], [14], [15], [16], and [17]. These models are generally based on the voltage and current analysis which allows to establish the model of the PEMFC. The PEMFC terminal voltage can be definedasexpressedin(3),where is the thermodynamic potential of the cell, is the voltage drop due to the activation of the anode and the cathode, is the voltage drop due to series resistance, presents the voltage drop due to the concentrations, and is the cell number in series The thermodynamic potential of the cell and the voltage drop due to the activation of the anode, and the cathode are respectively given in (4) and (5). In these equations, is the cell temperature in Kelvin, and present the pressures in (3)

3 688 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013 atmosphere of the hydrogen and the oxygen, respectively. is the FC s current; is the parametric coefficients [18], [14] TABLE II FC PARAMETERS (4) (5) In (5), is the concentration of the oxygen. This concentration can be calculated using the following equation [18]: (6) The and components are given in (7) and (8), where is the current density in, is a constant parameter extracted in [15], [14] In (7), is the electron flow resistance which is approximately constant, and is the proton s flow equivalent resistance which is expressed in (9). In this equation, is the thickness of the polymer membrane in centimeters, is the active cell area in. In this equation, is the specific membrane resistivity for the flow of the hydrated protons in cm. This parameter can be estimated as presented in (10), where presents the membrane s humidity ratio [18]. The used parameters of the FC model are presented in Table II (7) (8) (9) The resulting analytical model is given in (11), where is the sign of the current, and is the equivalent value of the converter duty cycle [10]. In this equation, and present respectively the boost and the buck converters duty cycles. B. Boost Converter Modeling By an analogy to the buck-boost converter, the analytical model of the boost converter is presented in (12), where presents the converter duty cycle [8] (12) C. DC/AC Converter Modeling The dc/ac converter (inverter) presented in Fig. 4 includes six bidirectional semiconductors (IGBT). The six transistors are in the anti-parallel configuration with six diodes. The analytical model of the inverter must be established from the sequences analysis of the converter s operation. These sequences are due to logical signals switching based on the following conventions: (10) is and is is and is (13) III. DC/DC DC/AC CONVERTERS AND ELECTRIC MACHINES MODELING A. Buck-Boost Converter Modeling To establish a model of the buck-boost converter illustrated Fig. 4, it is necessary to analyze the buck and the boost operations. During the boost operations, semiconductor is switched and is in OFF position. In this condition, the ultracapacitors module provides energy to the dc-bus. In buck mode, is switched and become inoff position. So the ultracapacitors receives the energy from the dc-bus (11) The resulting analytical model of the inverter is presented in (14) for the voltage, and the corresponding current is expressed in (15) (14) (15) D. Electric Machines Modeling The analytical model of the asynchronous machine is given in (16), where,,,and are respectively the equivalent currents in the stator and the rotor, and are respectively the stator and the rotor resistances. In this equation, and

4 TANI et al.: DC/DC AND DC/AC CONVERTERS CONTROL FOR HEVS ENERGY MANAGEMENT-ULTRACAPACITORS AND FC 689 Fig. 4. Hybrid system topology for the embedded energy management. present the pulsations of the flux in the stator and the rotor, respectively. and present the dq components of the flux in the stator; is the rotor flux in direction of axis (16) The equations of the flux are given in (17), where and are the inductances in the stator and the rotor, is the mutual inductance (17) The pulsation in the stator is given in (18), where is the number of pair of pole, and is the mechanical speed in rad/s (18) To emulate the HEV behavior during the deceleration and the braking operations, a dc-motor is used to drive the asynchronous machine (MG). The analytical model of the dc-motor is given in (19), where is the internal resistance, is the internal inductance of the motor, and is the dc-motor EMF In this equation, is the buck converter duty cycle, and presents the dc-motor terminal voltage. (19) IV. HYBRID SYSTEM CONTROL METHOD The originality of this paper is focused on the bidirectional load (motor and generator operations) power sharing between the FC and the ultracapacitors, using the New European Drive Cycle (NEDC) and the polynomial control technique. The proposed control method compared to others methods enables to allocate the average power to the FC and the fluctuating power (due to the acceleration, the deceleration, and the braking operations) to the ultracapacitors. In other terms, the proposed method takes into account the dynamic characteristics of the ultracapacitors and the FC, which allows improving the energetic performances and the life time of the sources. The polynomial control technique presents a robust algorithm with good performance in the following situations. When the system has a pure delay or presents a dynamic characteristic which changes during operations. The polynomial control technique is also interesting if the reference should not be exceeded. The polynomial controller is an interesting alternative solution to the conventional PI controller. In other words, the polynomial controller presents the better performances in term of the rapidity and the robustness compared to conventional PI controller. The polynomial controller improves the disturbance rejection. For more information, a comparative study of the PI and the polynomial controller is presented in [10]. More details about the polynomial control technique can be found in [19]. A. Polynomial s Coefficients Estimation Method Thedegreeof,,and polynomials are fixed according to the degree of the discrete system transfer functions. The estimation method of the polynomial s coefficients is based on the closed-loop analysis as illustrated in Fig. 5 for the current, and the dc-bus voltage management. The transfer functions of the discrete systems are presented in (20), where, and, present

5 690 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013,,,and are selected as expressed in (26) (26) Fig. 5. Polynomial s coefficients estimation diagrams. the denominators and the numerators of the transfer functions, is the dc-bus voltage smoothing capacitor, is the ultracapacitors current smoothing inductance, is the sampling period The final coefficients obtained from the closed loops analysis, in the case of the FC current and the dc-bus voltage control are respectively expressed in (27) and (28). In these equations is the dc-bus voltage smoothing capacitor, is the FC s current smoothing inductance, is the sampling period, is the current control bandwidth, is the converter control frequency (27) (20) The simplified closed loop transfer function for the ultracapacitor s current control is expressed as follows: (21) The desired polynomial in the closed loop is presented in (22) (22) To reduce the number of the parameters to be identified, the used method consists to choose and identical. These ones are identified, using a simple comparison between the desired polynomial and the denominator of the transfer function in closed loop as presented in (23) (23) (24) The resulting coefficients from this comparison are given in (24), where the and depend of the dynamic response of the system in closed loop. These coefficients can be obtained using the following desired polynomial, where is the system control bandwidth B. DC-Bus Voltage Control Method (25) To manage the dc-bus voltage, two control loops are necessary. The first is the current feedback (inner loop), and the second is the voltage loop (outer loop) as illustrated in Fig. 6. In goal to obtain a minimal static error with disturbance rejection, the following polynomials correctors:,, (28) The dc-bus voltage control law established from (12) is given in (29). To control the dc-bus voltage, the cascaded control loops are necessary: an inner loop for the FC s current control and the outer loop for the dc-bus voltage ones (29) The FC reference current estimated from electric powers balances between the boost converter input and output is presented in (30) (30) In this equation, is the boost converter output current, and present the dc-bus voltage control loop output signal. This last one corresponds to estimate instantaneous dc-bus capacitor s current. C. Ultracapacitors Current Control Method To control the ultracapacitors current, the polynomial control method presented in Fig. 7 is used. In this case, the cascaded control loop is not necessary, because, the control is focused directly on the ultracapacitors current. To obtain a minimal static error with disturbance rejection, the, and are selected as expressed in (31), where and estimation method is same to that presented in (27) (31) To control the ultracapacitors current, the bidirectional converter control laws obtained from the buck-boost converter modeling are used. These control laws are presented in (32) for the buck operations, and in (33) for the boost mode. These

6 TANI et al.: DC/DC AND DC/AC CONVERTERS CONTROL FOR HEVS ENERGY MANAGEMENT-ULTRACAPACITORS AND FC 691 Fig. 6. DC-bus voltage control method. Fig. 7. Ultracapacitors current control loop. Fig. 8. Ultracapacitors reference current estimation method. control laws are compared to triangular waveform to modulate the PWM signals for the buck and the boost operations [8] (32) Fig. 9. Speed control of the asynchronous machine. (33) The reference current of the ultracapacitors is estimated using (34), where is the FC rated current. The corresponding diagram is illustrated in Fig. 8. This method enables to allocate the dynamic power of the load to ultracapacitors, and the average power to the FC. In other words, this method takes into account the conventional dynamic behavior of the ultracapacitors and the FC [20] (34) D. Speed Control of the Electric Machine To control the asynchronous machine speed, the used method is based on indirect rotor flux orientation control. This method is conventionally obtained from the following assumption: and. The pulsation in the rotor is estimated as expressedin(35).toestimatetherotorflux in the direction of axis, (36) is used [21] (35) (36) The used method for the speed control is illustrated in Fig. 9. For this method, two cascaded control loops are necessary. The first is the current feedback (inner loop), and the second is the speed control loop (outer loop). To obtain a minimal static error with disturbance rejection for the speed control, the, and are selected. These polynomials correctors are same to that presented in (26). The final coefficients obtained from the closed loops analysis, in the cases of currents, the speed and the flux control are respectively expressed in (37), (38), and (39). In these equations,

7 692 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013 TABLE III PARAMETERS OF THE VEHICLE Fig. 10. DC-motor speed control during the deceleration and the braking operations. is the sampling period, is the current control bandwidth, is the speed control bandwidth, is the flux control bandwidth, is the moment of inertia, is the friction coefficient (37) Fig. 11. Measured speed compared to its reference. (38) (39) for the dc-motor, the fol- To generate the reference speed lowing equation is used: (43) The PWM signals are generated by comparing three signals (obtained from the two to three phases park transformation) to a triangular waveform. The resistive torque which must be compensated by the vehicle to move forward is presented in (40). In this equation, is the grade of the road in degree, is the vehicle s speed in,and is the radius of the wheel in. These parameters are presented in Table III (40) In this paper, the road is assumed flat, i.e.. Using (40), the dynamic equation can be expressed as given in (41), where is the electromagnetic torque E. DC-Motor Speed Control (41) To control the dc-motor speed, the control strategy presented in Fig. 10 is used. To obtain a minimal static error with disturbance rejection, and polynomials are selected as expressed in the following: (42) The conditions for the dc-motor control are focused on the sign of as presented in the following: DC-motor is no controlled DC-motor is controlled (44) During the deceleration and the braking operations of the HEV,, the dc-motor is controlled from the buck converter so that its speed become higher than. During the acceleration and the constant speed operations, i.e., the asynchronous machine is controlled from the inverter as illustrated in Fig. 9. V. SIMULATION AND EXPERIMENTAL VALIDATIONS A. Test Conditions For the hybrid system simulations, the dc-bus voltage reference is fixedto47vandtheusedvalueofthe in the simulation and experimental tests is fixedto5a.the used parameters for the hybrid system simulations are given in Tables I IV. Fig. 11 presents the control result of the electric machine speed using the NEDC. This figure shows that, the measured speed is very close to its reference. The corresponding load s current profile obtained from the HEVs behavior simulation is plotted in Fig. 12. The speed control result obtained from the motor and the generator operations conditions is illustrated in Fig. 13. The acceleration and the constant speed situations correspond

8 TANI et al.: DC/DC AND DC/AC CONVERTERS CONTROL FOR HEVS ENERGY MANAGEMENT-ULTRACAPACITORS AND FC 693 Fig. 14. DC-bus voltage control result. Fig. 12. Load s current profile. TABLE IV USED PARAMETERS FOR THE HYBRID SYSTEM CONTROL Fig. 15. Measured current on the load during the simulation and the experimental tests. for the embedded energy management are implemented in two PIC18F4431 microcontrollers. The used parameters for the hybrid system control are presented in Table IV. Fig. 13. Speed control result obtained from the motor and the generator operations conditions. to motor operations. The deceleration and the braking operations correspond to generator mode. B. Experimental Setup An experimental test bench is carried out to validate the proposed control methods outlined above. The developed experimental test bench includes an ultracapacitors module (18 cells in series with a maximum voltage of 49 V), a programmable dc-source is used as the FC system with a maximum power of 1 kw (25 V/40 A), a buck-boost converter, a boost converter, an inverter, and the asynchronous machine coupled to a dc-motor. These electric machines are used as a bidirectional load (motor and generator operations). The dc-motor is connected to a buck converter. This converter is controlled during the deceleration and the braking operations. The proposed control algorithms C. Simulation and Experimental Results The dc-bus voltage control result is plotted in Fig. 14. This figure shows that, the experimental and the simulation curves are close to reference voltage regardless of the load s demand. In other words, the proposed dc-bus voltage control method is satisfactory. The measured current on the load during the simulation and the experimental tests are plotted in Fig. 15. This current presents four steps. The first step presents the acceleration operations which are characterized by the load s current increasing when the HEVs speed increase. The second step is characterized by the constant speed. During this step, the load s current is constant, and its value is small. The third step corresponds to the deceleration operations. During this step, the load s current presents the negative peaks current (energy recovery operations). The last step is characterized by a null speed which corresponds to zero current of the load. During this last step, the dc-bus voltage presents some variations around its reference (48 V) as illustrated between the 0 and 1000s of the Fig. 14. The contribution of the ultracapacitors in the dc-bus is plotted in Fig. 16. This contribution has the same shape as the load s current, and it presents two steps during the traction operations (asynchronous machine in the motor mode): the first operation corresponds to negative current of the due to low current of the load compared to the FC contribution; the second operation corresponds to positive current of the.

9 694 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013 Fig. 16. Contribution of the ultracapacitors in the dc-bus. Fig. 18. Measured current on the ultracapacitors. Fig. 17. Measured current in the dc-bus from the FC. Fig. 19. Ultracapacitors module terminal voltage. During this operation the ultracapacitors module contributes to the traction energy supply via the boost converter. During the energy recovery operations (asynchronous machine in the generator mode), the contribution of the ultracapacitors, and the measured current on the load are negative. So, the ultracapacitors module recovers the energy from the FC and the load. The simulation and the experimental results of the FC contribution in the dc-bus are plotted in Fig. 17. This contribution is always positive, and it presents less variation compared to the load s current variations. In other words, the ultracapacitors ensures the dynamic components of the load, and the FC provides the average components. In consequently, the life time and the size of the FC are improved. Fig. 18 presents the measured current on the ultracapacitors. This current has the same shape as the current, but they are not identical due buck-boost converter conversion ratio. The terminal voltage of the ultracapacitors is plotted in Fig. 19. This voltage presents two situations. The first situation is characterized by the load s demand which corresponds to ultracapacitors module discharge operations. The second situation corresponds to ultracapacitors voltage increasing due to low current of the load compared to the FC contribution or energy recovery process (asynchronous machine in the generator operations). The measured current on the FC is plotted in Fig. 20. This current presents less variation compared to the variations of the load. Fig. 21 shows the FC terminal voltage which is also constant because the measured current on this last one is constant. The performances of the proposed control are illustrated in Figs , which enable to conclude that, the ultracapacitors Fig. 20. Fig. 21. Measured current on the FC. FC terminal voltage. ensures the dynamic components of the load, and the FC provides the average power. To conclude this section, the simulation and the experimental results present some differences in term of the fluctuations and the average values. The differences related to average value are due to used model for the hybrid system simulation which does not takes into account the losses in the semiconductors and the

10 TANI et al.: DC/DC AND DC/AC CONVERTERS CONTROL FOR HEVS ENERGY MANAGEMENT-ULTRACAPACITORS AND FC 695 are proposed. The proposed methods are evaluated through the hybrid system behavior simulations. To validate the theoretical study, an experimental test bench is carried out in reduced scale. The obtained results from the experimental tests are analyzed and compared to that of the simulations ones. The simulation and the experimental results enable to conclude that, the proposed methods for the energy management based on the polynomial s correctors are interesting and effortless to implement in the PIC18F4431 microcontroller. Finally, the energy management based on the load s demand sharing according to the dynamic responses of the FC and the ultracapacitors enables to avoid the accelerated aging of the FC due to transient currents of the load. REFERENCES Fig. 22. (a) Spectrum of the load s current measured in the dc-bus. (b). Spectrum of the ultracapacitors current measuredinthedc-bus.(c).spectrumofthe FC s current measured in the dc-bus. electrical wiring. The observed fluctuations in Figs. 14, 17, 20, and 21 are due to the performances limitation of the used sensors for data acquisition. In other words, the used filters in the experimental tests are not same to the simulation ones. Fig. 22(a) and (b) present, respectively, the spectrum of the load s current, and the ultracapacitors ones. The FC current spectrum is illustrated in Fig. 22(c). These spectrums show that the FC provides the average power and the ultracapacitors module meets the dynamic fluctuations of the load. In other words, the energy management is focused on the load s demand sharing between the FC and the ultracapacitors with a major allocation of the load s variations to the ultracapacitors during the transient operations. This approach enables to avoid the accelerated aging of the FC due to transient currents of the load. VI. CONCLUSION This paper presents the energy management method for the HEVs applications. The FC and the ultracapacitors models are presented. The dc-bus voltage and the current control methods [1] M. Zandi, A. Payman, J. P. Martin, S. Pierfederici, B. Davat, and F. M. Tabar, Energy management of a fuel cell/supercapacitor/battery power source for electric vehicular applications, IEEE Trans. Veh. Technol., vol. 60, no. 2, pp , Feb [2] M. Bertoluzzo and G. Buja, Development of electric propulsion systems for light electric vehicles,.ieee Trans. Ind. Inf., vol. 7, no. 3, pp , Aug [3] P. Thounthong, S. Pierfederici, J. P. Martin, M. Hinaje, and B. Davat, Modeling and control of fuel cell/supercapacitor hybrid source based on differential flatness control, IEEE Trans. Veh. Technol., vol. 59, no. 6, pp , Jul [4] I. Aharon and A. Kuperman, Topological overview of power trains for battery-powered vehicles with range extenders, IEEE Trans. Power Electron., vol. 26, no. 3, pp , Mar [5] P. J. Grbović, P. Delarue, P. Le Moigne, and P. Bartholomeus, The ultracapacitor-based controlled electric drives with braking and ridethrough capability: Overview and analysis, IEEE Trans. Ind. Electron., vol. 58, no. 3, pp , Mar [6] J. Bauman and M. Kazerani, An analytical optimization method for improved fuel cell-battery Ultracapacitor powertrain, IEEE Trans. Veh. Technol., vol. 58, no. 7, pp , Sep [7] W. S. Liu, J. F. Chen, T. J. Liang, and R. L. Lin, Multicascoded sources for a high-efficiency fuel-cell hybrid power system in highvoltage application, IEEE Trans. Power Electron., vol. 26, no. 3, pp , Mar [8] M. B. Camara, B. Dakyo, and H. Gualous, Supercapacitors and battery energy management based on new european driving cycle, Journal of Energy and Power Engineering, vol. 6, no. 2, pp , Feb. 2012, , Serial no. 51. [9] F. Segura, J. M. Andujar, and E. Duran, Analog current control techniques for power control in PEM fuel-cell hybrid systems: A critical review and a practical application, IEEE Trans. Ind. Electron., vol. 58, no. 4, pp , Apr [10]M.B.Camara,H.Gualous,F.Gustin,A.Berthon,andB.Dakyo, DC/DC converter design for supercapacitor and battery power management in hybrid vehicle applications-polynomial control strategy, IEEE Trans. Ind. Electron., vol. 57, no. 2, Feb [11] W. Na, T. Park, T. Kim, and S. Kwak, Light fuel-cell hybrid electric vehicles based on predictive controllers, IEEE Trans. Veh. Technol., vol. 60, no. 1, pp , Jan [12] H.H.Wu,A.Gilchrist,K.D.Sealy,andD.Bronson, Ahighefficiency 5 kw inductive charger for EVs using dual side control,.ieee Trans. Ind. Inf., vol. 8, no. 3, pp , Aug [13] Q. Li, W. Chen, Y. Wang, S. Liu, and J. Jia, Parameter identification for PEM fuel-cell mechanism model based on effective informed adaptive particle swarm optimization, IEEE Trans. Ind. Electron., vol. 58, no. 6, pp , Jun [14] S. M. Sharifi Asl, S. Rowshanzamir, and M. H. Eikani, Modelling and simulation of the steady-state and dynamic behaviour of a PEM fuel cell, Energy, vol. 35, pp , [15] C. A. Ramos-Paja, R. Giral, L. M. Salamero, J. Romano, A. Romero, and G. Spagnuolo, A PEM fuel-cell model featuring oxygen-excessratio estimation and power-electronics interaction, IEEE Trans. Ind. Electron., vol. 57, no. 6, pp , Jun [16] A. Y. Sendjaja and V. Kariwala, Decentralized control of solid oxide fuel cells,.ieee Trans. Ind. Inf., vol. 7, no. 2, pp , May 2011.

11 696 IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, VOL. 9, NO. 2, MAY 2013 [17] A.M.Dhirde,N.V.Dale,H.Salehfar,M.D.Mann,andT.H.Han, Equivalent electric circuit modeling and performance analysis of a PEM fuel cell stack using impedance spectroscopy, IEEE Trans. Energy Convers., vol. 25, no. 3, pp , Sep [18] J.-H. Jung, S. Ahmed, and P. Enjeti, PEM fuel cell stack model development for real-time simulation applications, IEEE Trans. Ind. Electron., vol. 58, no. 9, pp , Sep [19] M. B. Camara, B. Dakyo, and H. Gualous, Polynomial control method of DC/DC converters for DC-bus voltage and currents management Battery and supercapacitors, IEEE Trans. Power Electron., vol. 27, no. 3, Mar [20] S. Chae, Y. Song, S. Park, and H. Jeong, Digital current sharing method for parallel interleaved DC-DC converters using input ripple voltage,.ieee Trans. Ind. Inf., vol. 8, no. 3, pp , Aug [21] D. G. Holmes, B. P. McGrath, and S. G. Parker, Current regulation strategies for vector-controlled induction motor drives, IEEE Trans. Ind. Electron., vol. 59, no. 10, pp , Oct Mamadou Baïlo Camara (M 12) was born in Mamou, Guinea, in He received the B.S. degree in electrical engineering from the Polytechnic Institute of Conakry (IPC), Conakry, Guinea, in 2003, and the M.S. and Ph.D. degrees from the University of Franche Comté, Belfort, France, in 2004 and 2007, respectively. He is currently an Associate Professor with the Groupe de Recherche en Electrotechnique et Automatique du Havre Laboratory (GREAH), University of Le Havre, Le Havre, France. Since 2004, he has been working in power electronics and electric vehicle research projects, involving static converter topologies, ultracapacitors, batteries, and electrical energy management for hybrid vehicle applications. Brayima Dakyo (M 06) received the B.S. and Dr.Eng. degrees from Dakar University, Dakar, Senegal, in 1984 and 1987, respectively, and the Ph.D. and Habilitation degrees from the University of Le Havre, France, in 1988 and 1997, respectively. He is a full Professor of Electrical Engineering, and Head of the GREAH Laboratory, University of Le Havre. His current interests include power electronics, converter fed electrical machines, analytical model, wind and solar energy systems, energy management, and system design with storage. Abdallah Tani was born in Algeria in July He received the B.S. degree in electromechanic engineering from the University of Bejaia, Bejaia, Algeria, in 2008, and the M.S. degree from University of Le Havre, Le Havre, France, in Since 2009, he has been working in power electronics, dc distribution system and hybrid electric vehicle research projects, involving converter topologies, FC ultracapacitors, and batteries dedicated to transport applications. Yacine Azzouz wasborninferkane,algeria.hereceived the B.S. degree from the Institute of Annaba, Annaba, Algeria, in 1990, the M.S. degree from the National Polytechnic Institute of Toulouse, Toulouse, France, in 1995, and the Ph.D. degree in electrical engineering from the University of Aix Marseille III, Marseille, France, in He is currently an Associate Professor with the IRSEEM Laboratory, École d ingénieurs ESIG- ELEC, Saint-Etienne du Rouvray, France.