See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/264817074

Fuel cell-based distributed generation system with power flow and power
quality control

Article in International Journal of Power and Energy Conversion · January 2013

DOI: 10.1504/IJPEC.2013.050929

CITATIONS READS

3 106

3 authors, including:

Gitanjali Mehta S P Singh

Galgotias University, Greater Noida, India Jawaharlal Nehru Krishi Vishwavidyalaya
46 PUBLICATIONS 306 CITATIONS 82 PUBLICATIONS 1,584 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Gitanjali Mehta on 12 April 2021.

The user has requested enhancement of the downloaded file.

Int. J. Power and Energy Conversion, Vol. 4, No. 1, 2013 73

Fuel cell-based distributed generation system with

power flow and power quality control

Gitanjali Mehta* and S.P. Singh

Electrical Engineering Department,
Indian Institute of Technology,
Roorkee, Uttarakhand-247667, India
E-mail: [email protected]
E-mail: [email protected]
*Corresponding author

R.D. Patidar
Electrical and Electronics Engineering Department,
Mandsaur Institute of Technology,
Mandsaur, MP-458001, India
E-mail: [email protected]

Abstract: This paper presents the modelling and control of fuel cell-based
distributed generation system with embedded active filter functioning. It
includes the modelling of proton electrolyte membrane fuel cell and the power
conditioning unit interfacing the fuel cell to the utility grid with proper power
flow control. The output of the fuel cell stack is given to the DC-DC boost
converter before connecting to the DC side of the voltage source inverter for
interfacing to the grid. In the presented work, the features of active power filter
have been incorporated in the control circuit of the current controlled-voltage
source inverter interfacing the fuel cell to the grid. Thus the inverter is utilised
to inject fuel cell power generated to the grid and also to act as shunt active
power filter to compensate for load current harmonics and reactive power
demand. Simulation and experimentation is carried out to verify the operation
and the control principle.

Keywords: fuel cell; distributed generation; active power filter; harmonics;

active and reactive power.

Reference to this paper should be made as follows: Mehta, G., Singh, S.P. and
Patidar, R.D. (2013) ‘Fuel cell-based distributed generation system with power
flow and power quality control’, Int. J. Power and Energy Conversion, Vol. 4,
No. 1, pp.73–94.

Biographical notes: Gitanjali Mehta received her BTech in Electronics and

Communication Engineering from M.J.P. Rohilkhand University, Bareilly,
India in 2003 and MTech in Power Electronics and ASIC Design from
MNNIT, Allahabad, India in 2005. Currently, she is pursuing her PhD in the
field of distributed generation in the Department of Electrical Engineering,
Indian Institute of Technology, Roorkee, India. Her research interest includes
renewable energy, distributed generation systems and power quality control
using active filters.

Copyright © 2013 Inderscience Enterprises Ltd.

74 G. Mehta et al.

S.P. Singh received his BSc in Electrical Engineering from Aligarh Muslim
University, Aligarh, India in 1978. He received his ME and PhD in
Electrical Engineering from Indian Institute of Technology, Roorkee (formerly
University of Roorkee), India in 1980 and 1994, respectively. Currently, he is a
Professor in the Department of Electrical Engineering, Indian Institute of
Technology, Roorkee, India. His research areas include electric machines and
drives, power converters, power quality, active filters and renewable energy
systems.

R.D. Patidar graduated in Electrical Engineering from Government Engineering

College, Rewa, India in 1995. He completed his ME in Power Electronics from
SGSITS, Indore, India in 2004 and PhD in the field of active filters from
Department of Electrical Engineering, Indian Institute of Technology, Roorkee,
India in 2010. Currently, he is a Professor in the Department of Electrical
Engineering, Mandsaur Institute of Technology, Mandsaur, India. His field of
interest includes power quality, active filters, and renewable energy.

1 Introduction

The rapidly increasing global energy consumption has become a matter of great concern
for both utilities and the end users. The fossil fuels which are the primary sources of
electric power cause serious environmental pollution and moreover these fossil fuels are
on the verge of extinction. Hence, a transition from conventional energy to more cleaner
and secure energy is necessary to alleviate energy crisis and to address environmental
concerns. Distributed generation (DG) are rapidly increasing across the globe because
they can meet the increasing power demand while complying with the environmental
regulations of low emissions (Blaabjerg et al., 2004; Bose, 2000).
Fuel cell (FC) technology is expected to play an important role in meeting the
growing demand for DG for sustainable power generation. FC is an electrochemical
device that converts the chemical energy of fuel directly into DC form of electrical
energy with heat and water as by-product (Kramer et al., 2008). Interfacing the FC to the
grid presents a quite different and challenging scenario because unlike the conventional
system, the FC cannot be directly connected to the grid. A power conditioning interface
between the FC and grid is required to match the characteristics of FC and the
requirements of the grid connections such as voltage, frequency, active and reactive
power control, harmonic minimisation, etc. (Kramer et al., 2008; Teodorescu et al.,
2006). Real et al. (2007), Correa et al. (2005), Soltani and Bathaee (2008), Wang et al.
(2005), Al-Baghdadi (2007) and Souleman et al. (2009) have presented work on dynamic
modelling of FCs which is required for designing appropriate interfacing circuits and
controllers for practical FC systems and to investigate FC transient responses when
operating conditions change with time. Tanrioven and Alam (2006) have presented an
overall evaluation of a PEMFC power plant in terms of stack modelling, active and
reactive power control and power quality (PQ). Park et al. (2008) have proposed a wide
range active and reactive power flow controller for a solid oxide FC grid interfaced
system.
The increased use of non-linear devices results in many PQ problems in the power
system network. These non-linear devices not only increase the reactive current but also
 Fuel cell-based distributed generation system 75

generate significant current harmonics giving rise to non-sinusoidal current and voltage
waveforms at the point of common coupling (PCC). Many active power filter topologies
(shunt, series, hybrid, unified PQ conditioners) have been proposed in the recent years to
solve the PQ problems. The shunt active power filter (SAPF) presents a dynamic solution
best suited for the compensation necessities. The principle of operation of the SAPF is to
supply the undesired harmonics and reactive power to the load, so that the mains current
is of improved quality (Chaoui et al., 2007; Kale and Ozdemir, 2005). Singh et al. (1998),
Chandra et al. (2000), Jain et al. (2003), Kunjumuhammed and Mishra (2006) and Patidar
and Singh (2010) have presented different control techniques for customer generated
harmonics and reactive power compensation using SAPF. Rahmani et al. (2010) have
presented a non-linear control technique in synchronous reference frame which
compensates for unwanted reactive, unbalanced and harmonic load current components
under sinusoidal supply voltage conditions. Patidar and Singh (2011) have performed a
comparative analysis of different control schemes for reactive and harmonic current
compensation in single phase system using SAPF.
Singh et al. (2011) have proposed a novel control strategy for achieving
maximum benefits from grid interfacing inverters where the grid interfacing inverter is
utilised to inject real power generated from renewable energy source to the grid and/or
operate as SAPF. Wu et al. (2007) have presented a half bridge single-phase two-wire
photovoltaic inverter system that can perform both active power filtering and real power
injection.
With the objective of reducing the cost and increasing the efficiency, a two-stage,
single-phase, grid-interactive FC system has been proposed which includes the
functionality of SAPF. The function of APF is added in the existing inverter of
FC system by making the necessary modification in the control circuit. Thus,
the proposed interface does not require any additional circuit for enabling the existing
inverter to also perform APF function. This concept, thus, reduces the overall design
cost of the system. The proposed system is capable of injecting FC power to electric
grid while compensating load reactive power and harmonics caused by non-linear
loads.
This paper presents the modelling, design and simulation of proton electrolyte
membrane fuel cell (PEMFC)-based power supply system for residential applications.
The system consists of a 6 kW PEMFC, a DC-DC boost converter for power conditioning
and a current controlled-voltage source inverter (CC-VSI) for interfacing to the grid and
providing active filter functions. A step-up transformer is used to increase the voltage
level before the system is connected to the utility grid. Simulation study in
MATLAB/Simulink and experimentation is carried out to verify the proposed system
with its control so as to achieve both the objectives of grid interfacing of FC system as
well as active filtering.

2 Modelling of FC stack

The FC stack is modelled as a controlled voltage source E in series with a constant

resistance Rfc (internal resistance of the FC stack) as shown in Figure 1 (Souleman et al.,
2009). The controlled voltage source is represented as:
76 G. Mehta et al.

⎛ i fc ⎞
E = Eoc − NA ln ⎜ ⎟ (1)
⎝ i0 ⎠
and FC stack output voltage as:
V fc = E − R fc i fc (2)

where Eoc is the open circuit voltage, N is the number of cells in the stack, A is the Tafel
slope, i0 is the exchange current resulting from the continual backward and forward flow
of electrons from and to the electrolyte at no load, ifc is the FC stack current and Vfc is the
FC stack voltage.

Figure 1 FC stack model

The Tafel Slope A, exchange current i0 and the open circuit voltage Eoc are affected by
variations in temperature and partial pressures of hydrogen and oxygen. They can be
represented as:
Eoc = KEn (3)

zFk ( pH 2 + pO2 ) ⎛ −ΔG ⎞

i0 = exp ⎜ ⎟ (4)
Rh ⎝ RT ⎠
RT
A= (5)
zα F
Nernst voltage En, the thermodynamics voltage of the cells depends on the temperature
and partial pressures of reactants inside the stack and is given as:

⎛ −44.43 ⎞ RT ⎛⎜
1 ⎞
En = 0.229 + (T − 298) ⎜ ⎟+ ln ⎝ pH 2 + pO2 2 ⎟⎠ (6)
⎝ zF ⎠ zF
α is the charge transfer coefficient, T is the operating temperature of the cell, pH 2 is the
partial pressure of hydrogen inside the stack, pO2 is the partial pressure of oxygen inside
 Fuel cell-based distributed generation system 77

the stack, k is the Boltzmann’s constant (1.38 × 10–23 J/K), h is the Plank’s constant
(6.626 × 10–34 Js), ΔG is the size of the activation barrier, z is the number of moving
electrons and K is the voltage constant at nominal operating condition. Α and ΔG depends
on the type of electrode and catalyst.
The modelling of PEMFC is based on the equations (1) to (6). Figures 2(a) and 2(b)
shows the V-I and P-I characteristics of the 6 kW 45 V FC stack respectively. Inom, Vnom
and Pnom is the FC stack current, voltage and power at nominal operating point and Imax,
Vmax and Pmax is the FC stack current, voltage and power at maximum operating point.
The FC stack cannot be operated beyond this point as the electrolyte membrane gets
damaged.

Figure 2 The V-I and P-I characteristic of FC stack

(a)

(b)

3 The proposed FCDG system description

The circuit of the proposed single-phase grid-interactive FC system is shown in Figure 3.

A current controlled SAPF with energy storage capacitor Cdc is connected in parallel with
the non-linear load. The SAPF consists of an inductor Lc and a resistor Rc on the AC side
of single-phase IGBT H-bridge CC-VSI. A FCDG unit is connected to the DC-link of
SAPF through a DC-DC converter. The SAPF compensates the current harmonics and
load power factor while the FC supplies power to the grid and the load. The FC stack
output current ifc and the hydrogen flow rate qH 2 are related as:

2 FqH 2
i fc = (7)
N
78 G. Mehta et al.

Thus the hydrogen flow rate needs to be adjusted with change in power demand. A fuel
flow controller is used to control the fuel flow rate to obtain a desired output power
(El-Sharkh, 2004).

Figure 3 Schematic diagram of FCDG grid-interfaced system

3.1 Design of DC-DC boost converter and its control

The FC generates power at variable low DC voltage. A simple DC-DC boost converter is
used to boost the output voltage of the FC stack before being interfaced to the DC-link of
the grid-interfacing inverter. The DC-DC converter thus provides a regulated DC output
voltage under varying load and input voltage conditions. The boost converter has higher
efficiency and requires less components as compared to other DC-DC converter
topologies like push pull, half bride and full bridge, etc., which could possibly be used to
interface the FC system to the grid (Kirubakaran et al., 2009).
Figure 4 shows the boost converter with output voltage control. The output
voltage Vdc is compared with a reference value Vdc,ref and the error signal is processed
through PI controller. The output of the controller is used to control the pulse width
modulated pulses with the right duty ratio so that the output voltage follows the reference
value.
The values of inductor Ld and capacitor Cd are selected according to the rated voltage,
current ripple, voltage ripple and switching frequency of the boost converter. The duty
ratio, D of the boost converter is:
V fc
D = 1− (8)
Vdc

where Vfc is the input voltage and Vdc is the output voltage of the converter. The gain of
the boost converter considering the internal source resistance Rfc of the FC stack can be
given by
 Fuel cell-based distributed generation system 79

Vdc 1
= (9)
V fc R fc
(1 − D) +
(1 − D) R

The parameters derived for DC-DC boost converter and used for simulation is given in
Table 1.

Figure 4 Boost DC-DC converter with control circuit

Table 1 Parameters used for boost DC-DC converter simulation

Parameters Values
Input voltage Vfc = 45 V
Output voltage Vdc = 100 V
Inductance Ld = 0.05 mH
Capacitance Cd = 1,000 μF
Proportional constant Kpd = 0.005
Integral constant Kid = 0.15
Switching frequency fs = 10 kHz

3.2 Grid-interfacing/SAPF inverter control

Pulse-width modulated (PWM) H-bridge CC-VSI is used to interconnect the FC system
to the utility grid for power flow and PQ control purposes. A DC-link capacitor on the
DC side of the VSI acts as an energy buffer and maintains a stable DC voltage for the
converter in the steady state condition. This DC-link capacitor decouples the FC from the
grid and also allows independent control of converters on either side of DC-link.
Figure 5 shows the grid-interfacing inverter Vc ∠ δc connected to the PCC Vs ∠ 0
through a coupling impedance X = R + jωL. The AC-side voltage of the inverter is
controlled both in magnitude and phase to control the active and reactive power flow
80 G. Mehta et al.

between the FCDG and the grid (Dai et al., 2008). The active power P and reactive power
Q flow from FCDG to the PCC can be given by
VsVc maV fcVs
P= sin δ c = sin δ c (10)
X X
Vs V
Q= (Vc cos δ c − Vs ) = s ( maV fc cos δ c − Vs ) (11)
X X
where ma is the inverter modulation index.

Figure 5 Single line diagram of grid-interfaced FC system

The grid-interface/SAPF inverter control circuit of proposed grid-interactive FC system is

shown in Figure 6. The inverter consumes a small amount of active power to maintain the
DC-link voltage and to overcome the losses associated with the inverter. The losses in the
inverter are because of the switching loss in the devices, iron and copper losses in the
circuit components, etc. Discrete PI controller is used to maintain a constant DC-link
voltage under varying conditions.

Figure 6 Control circuit of grid-interfacing/SAPF inverter

Thus inverter loss power can as expressed as:

(2
Psl = K p Vdcref )
− Vdc2 + Ki ∫ (V
2
dcref )
− Vdc2 dt (12)
 Fuel cell-based distributed generation system 81

The average value of active power supplied by the grid is:

Ps = PL + Psl − Pfc (13)

where PL is the load active power, Psl is the active loss power of the inverter and Pfc is the
active power supplied by the FC stack. Ps is positive when both grid and FC supply
power to the load. Ps is negative when the FC supply power to the load as per the load
requirement and rest of the power is supplied to the grid. The active power supplied by
the source can be written as,
Vs1 I s1 cos φs
Ps = (14)
2
Vs1 and Is1 are the peak value of fundamental component of PCC voltage and grid current
respectively. The peak value of the desired source currents after compensation can be
obtained as:
2 Ps
I s1 = (15)
Vs1

where cosφs = 1 for unity power factor operation. The multiplication of peak value of the
desired source current with unity grid voltage generates the reference grid current. A low
pass filter is used to filter out the high frequency components in the PCC voltage. The
grid synchronising angle obtained from phase lock loop (PLL) is used to generate unity
grid voltage.

is* (t ) = I s1 sin ω t (16)

The reference current is compared with the load current to get the compensating current
ic(t).

ic (t ) = iL (t ) − is* (t ) (17)

This compensating current is given to hysteresis controller to generate PWM pulses to

switch the gate drives of grid-interfacing inverter.

4 Simulation results and discussions

The entire FCDG circuit is simulated in MATLAB/Simulink using power system block
sets to validate its performance. The simulation results are presented for two different
modes of operation. In the first mode of operation there is no power generation from the
FC stack. Thus in this mode of operation, the grid-interfacing inverter is working as a
SAPF, thus improving the PQ at PCC. In the second mode of operation, apart from
working as a SAPF, the inverter is also injecting active power from FC to the grid. The
grid current waveforms have been analysed to obtain their total harmonic distortions
(THDs) before and after compensation in different modes of operation.
The parameters used for FC/APF inverter simulation is given in Table 2.
82 G. Mehta et al.

Table 2 Parameters used for FC/APF inverter simulation

Parameters Values
Source voltage and vs = 230 V, 50 Hz
frequency
Source resistance Rs = 0.1 Ω
Source inductance Ls = 0.15 mH
Filter inductance Lc = 5 mH
DC-link capacitance Cdc = 1,500 μF
DC-link voltage Vdcref = 100 V
Proportional constant Kpa = 1
Integral constant Kia = 12.5
Switching frequency fsw = 10 kHz
Step-up transformer Vn = 100 V/230 V, Pn = 10 kW
R1 = R2 = 0.002 pu, L1 = L2 = 0.08 pu
Load parameters Mode 1(a) t < 0.4, t > 0.8 Z = 7.5 + j7.85
0.4 < t < 0.8 Z = (7.5 + j7.85) || (7.5 + j0)
Mode 1(b) t < 0.4, t > 0.8 Zdc_side = 7.5 + j7.85
0.4 < t < 0.8 Zdc_side = (7.5 + j7.85) || (7.5 + j0)
Mode 1(c) t < 0.4, t > 0.8 Zdc_side = 7.5 + j7.85
0.4 < t < 0.8 Zdc_side = (7.5 + j7.85) || (7.5 + j0)
Mode 2 t < 0.4, t > 0.8 Zdc_side = 7.5 + j7.85
0.4 < t < 0.8 Zdc_side = (7.5 + j7.85) || (7.5 + j0)

4.1 Mode 1: no supply from FCDG, active power filtering mode

In this mode of operation, there is no power generation from the FC stack. Thus the grid
interfacing inverter is operating as a SAPF. In other words, the grid-interfacing inverter is
utilised as SAPF when there is no power generation from FCDG. In the SAPF mode of
operation, the inverter consumes a small amount of active power from the grid to
maintain the DC-link voltage and to overcome the losses associated with the inverter and
most of the load reactive power need is provided by inverter effectively.

4.1.1 Case 1(a) reactive compensation in linear load under sinusoidal voltage
In this case a linear load is connected at PCC. Thus, both source voltage as well as source
current is sinusoidal but not in phase. The APF is required to compensate the reactive
power only. At t = 0.1, the inverter is switched on. At this instant, the inverter starts
injecting the compensating current so as to compensate the phase difference between the
source voltage and current. Figure 7(a) shows the waveforms of grid voltage Vs load
current IL, grid current Is, APF inverter compensating current Iv and DC-link capacitor
voltage Vdc. The grid voltage Vs (scaled by a factor 0.2) and grid current Is is together
shown in Figure 7(b). At time t = 0.1 s, when the inverter is switched on, there is a
significant improvement in the power factor. Thus the APF compensates the phase
difference between the grid voltage and grid current. The DC-link voltage is maintained
 Fuel cell-based distributed generation system 83

at 150 V. The linear load is changed at time t = 0.4 s and t = 0.8 s. The results confirm the
good dynamic performance of the SAPF for a rapid change in the linear load current. The
corresponding grid, load and inverter active powers (Pgrid, Pload and Pinv) and reactive
powers (Qgrid, Qload and Qinv) is shown in Figure 7(c).

Figure 7 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(a) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) grid, load and inverter active powers (Pgrid, Pload and Pinv) and reactive powers
(Qgrid, Qload and Qinv)

(a)

(b)
84 G. Mehta et al.

Figure 7 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(a) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) grid, load and inverter active powers (Pgrid, Pload and Pinv) and reactive powers
(Qgrid, Qload and Qinv) (continued)

(c)

Figure 8 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(b) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) THDs of grid current before and after compensation

(a)
 Fuel cell-based distributed generation system 85

Figure 8 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(b) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) THDs of grid current before and after compensation (continued)

(b)

(c)

4.1.2 Case 1(b) reactive and harmonic compensation in non-linear load under
sinusoidal voltage
The source voltage is sinusoidal and the load current is non-sinusoidal because of the
non-linear load connected at PCC. The load in this case is a diode bridge rectifier load
followed by an inductor in series with a resistor. Figure 8(a) shows the waveforms of grid
voltage Vs, load current IL, grid current Is, APF inverter compensating current Iv and
DC-link capacitor voltage Vdc. The APF inverter is switched on at t = 0.1 s. The grid
86 G. Mehta et al.

current at t = 0.1 s become pure sinusoidal and in phase with the grid voltage as shown in
Figure 8(b). The fast Fourier transform (FFT) of the grid current before and after
compensation is carried out. The current THD is reduced from 30.58% to 2.70% as
shown in Figure 8(c). The non-linear load is changed at time t = 0.4 s and t = 0.8 s to
verify the dynamic performance of APF.

Figure 9 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(c) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) THDs of grid voltage, grid current before compensation and grid current after
compensation

(a)

(b)
 Fuel cell-based distributed generation system 87

Figure 9 (a) Grid voltage Vs, load current IL, grid current Is, inverter compensating current Iv and
DC-link voltage Vdc for Case 1(c) (b) grid voltage Vs (scaled by factor 0.2) and current
Is (c) THDs of grid voltage, grid current before compensation and grid current after
compensation (continued)

(c)

4.1.3 Case 1(c) reactive and harmonic compensation in non-linear load under
non-sinusoidal voltage
Most of the developed techniques assume sinusoidal supply voltage when compensating
non-linear load currents. Practically, the supply voltage contains harmonics which further
affect the shape of supply current with increased THD level. A non-sinusoidal voltage
signal of 50.93% voltage THD is generated by adding a third harmonic voltage source of
150 Hz in series with the 50 Hz source. This non-sinusoidal voltage is applied to
uncontrolled diode rectifier with series resistive-inductive load to build a non-sinusoidal
load current with 43.98% current THD. Figure 9(a) shows the waveforms of grid voltage
Vs, load current IL, grid current Is, APF inverter compensating current Iv and DC-link
capacitor voltage Vdc. At the instant the inverter is switched on at s = 0.1 s, it starts
injecting the current in a manner so that the grid current profile changes from non-linear
88 G. Mehta et al.

to sinusoidal current. Figure 9(b) reveals a comparison between grid voltage and grid
current before and after compensation. Also the APF maintains its performance during
change in load at 0.4 s and 0.8 s as shown. The FFT of the source current before after
compensation is carried out. The current THD is trimmed down from 43.98% to 7.79% as
shown in Figure 9(c). Thus the SAPF compensates for unwanted reactive and harmonic
load current components under non-sinusoidal supply voltage condition also.

4.2 Mode 2: FC feeding power, non-linear load at PCC

In this mode of operation, the FC is supplying power to the grid and the load. Thus, the
inverter is also injecting active power from FC to the PCC apart from working as a
SAPF. The AC-side voltage of APF inverter is controlled both in phase and magnitude to
control the active and reactive power, respectively.

Figure 10 (a) FC stack output voltage and current, boost converter output voltage and current
(b) grid voltage Vs, load current IL, grid current Is and inverter output current Iv for
Case 2 (c) grid voltage Vs (scaled by factor 0.2) and current Is (d) grid, load and FC
active powers (Pgrid, Pload and Pfc) and reactive powers (Qgrid, Qload and Qfc)

(a)

(b)
 Fuel cell-based distributed generation system 89

Figure 10 (a) FC stack output voltage and current, boost converter output voltage and current
(b) grid voltage Vs, load current IL, grid current Is and inverter output current Iv for
Case 2 (c) grid voltage Vs (scaled by factor 0.2) and current Is (d) grid, load and FC
active powers (Pgrid, Pload and Pfc) and reactive powers (Qgrid, Qload and Qfc) (continued)

(c)

(d)

The inverter is switched on at t = 0.05 s. Thus at t = 0.05 s the inverter starts injecting
active power generated from FC stack. Since the FC power is more than the load active
power demand Pload, after meeting the load power demand, additional power of the FC
flows to the grid. Figure 10(a) shows the FC stack output voltage Vfc and current Ifc,
and the DC-DC boost converter output voltage Vdc and current Idc. The DC-DC
converter increases the FC stack output voltage of 45 V to 100 V. The PCC voltage Vs,
load current IL, grid current Is and inverter current Iv is shown in Figure 10(b). Thus, the
grid-connected inverter provides the entire load power demand, i.e., active reactive and
harmonic load power and feeds the additional active power to the grid. The grid current
and voltage together is shown in Figure 10(c). The 180-degree phase shift between the
grid voltage and grid current suggests that the additional power is fed to the grid at unity
power factor. The exchange of active and reactive powers between grid, load and FC
90 G. Mehta et al.

stack is shown in Figure 10(d). The active powers of grid, load and FC stack are
represented as Pgrid, Pload and Pfc and the reactive powers of grid, load and FC stack by
Qgrid, Qload and Qfc respectively. Thus in this case the grid interfacing inverter is
simultaneously utilised to inject the power generated from FC to PCC and to improve the
PQ at PCC.

5 Experimental validation

Experimental prototype as shown in Figure 11 is developed and results obtained for

Mode 1 operation. The TMS320F2812 Digital Signal Processor (DSP) is used to realise
the proposed control scheme. The control algorithm is developed in MATLAB using
power system block sets and the TMS320F2812 DSP is connected with MATLAB
through a parallel port connector for transferring the real-time generated codes using the
software Code Composer Studio.
The DC-link voltage is sensed by a DC voltage sensor, load current is sensed by
LA25-NP current sensors and PCC voltage by LV25-P voltage sensor. These signals are
then applied to the inbuilt analogue-to-digital converter of the processor through signal
conditioner. The signal conditioner converts these entire signals into unipolar values in
the range of 0–3 V compatible with DSP. The DSP generated pulses are applied to the
switches through pulse isolation and a driver circuit. The parameters used for the
experimental studies are given in Table 3.

Figure 11 Experimental prototype for Mode 1 operation

 Fuel cell-based distributed generation system 91

Table 3 Parameters used for experimentation

Parameters Values
Source voltage and frequency vs = 30 V, 50 Hz
Source inductance Ls = 0.15 mH
Filter inductance Lc = 2.75 mH
DC-link capacitance Cdc = 1,600 μF
DC-link voltage Vdcref = 75 V
Proportional constant Kpa = 0.08
Integral constant Kia = 0.25

Figure 12 Experimental results for Case 1, (a) Case 1(a) 1 – (vs – 30 V), 2 – (is – 1.67 A),
3 – (ic – 0.75 A), 4 – (Vdc – 70 V) . (time – 10 ms/div) (b) Case 1(b) 1 – (iL – 1.8 A),
2 – (ic – 0.8 A), 3 – (is – 1.60 A), 4 – (Vdc – 70 V) . (time – 10 ms/div) (c) Case 1(c)
1 – (iL – 1.8 A), 2 – (vs – 28.3 V), 3 – (is – 1.72 A), (time – 5 ms/div)

(a)

(b)

(c)
92 G. Mehta et al.

Figure 12 shows the experimental results for active power filtering mode of operation
when there is no power generation from the FC stack. Figure 12(a) shows the
experimental results for reactive compensation in steady-state under sinusoidal voltage
and linear load. It is observed that APF compensate the reactive component but current
THD in this case after compensation is increased to some extent but within the
recommended limits. Figure 12(b) shows the experimental results for reactive and
harmonic compensation in steady-state under sinusoidal voltage and non-linear load. The
source current THD in this case is reduced from 35.87% to 2.62% along with reactive
compensation. Figure 12(c) shows the experimental results for reactive and harmonic
compensation in steady-state under non-sinusoidal voltage (THD 6.4%) and non-linear
load (THD 38.87%). Source current is found in phase with source voltage and sinusoidal
which ensures both harmonic and reactive compensation capabilities. The source current
THD in this case is reduced from 37.88% to 2.92%.

6 Conclusions

This paper presents the simulation model and control of a grid-interactive FC system with
embedded active filter functions under conditions of varying load demand. The features
of SAPF have been incorporated in the inverter interfacing the FC to the grid without any
additional hardware cost. Thus a common inverter with the proposed control is utilised to
inject active power generated from FC to the PCC and also to operate as SAPF. The
results show that the voltage source inverter not only injects active power at the PCC but
also compensates load reactive power and maintains a sinusoidal current from/to grid
thus proving the effectiveness of the proposed control in harmonic reduction and power
factor correction. After compensation, the grid current is sinusoidal and in phase with
grid voltage. The THD in the grid current is well within the IEEE 519-1992
recommended limits.

References
Al-Baghdadi, M.A.R.S. (2007) ‘A simple mathematical model of performance for proton exchange
membrane fuel cells’, International Journal of Sustainable Energy, Vol. 26, No. 2, pp.79–90.
Blaabjerg, F., Chen, Z. and Kjaer, S.B. (2004) ‘Power electronics as efficient interface in dispersed
power generation systems’, IEEE Transactions on Power Electronics, Vol. 19, No. 5,
pp.1184–1194.
Bose, B.K. (2000) ‘Energy, environment, and advances in power electronics’, IEEE Transactions
on Power Electronics, Vol. 15, No. 4, pp.688–701.
Chandra, A., Singh, B., Singh, B.N. and Al-Haddad, K. (2000) ‘An improved control algorithm of
shunt active filter for voltage regulation, harmonic elimination, power-factor correction, and
balancing of nonlinear loads’, IEEE Transactions on Power Electronics, Vol. 15, No. 3,
pp.495–507.
Chaoui, A., Gaubert, J.P., Krim, F. and Champenois, G. (2007) ‘PI controlled three-phase shunt
active power filter for power quality improvement’, International Journal of Electric Power
Components and Systems, Vol. 35, No. 12, pp.1331–1344.
Correa, J.M., Farret, F.A., Popov, V.A. and Simoes, M.G. (2005) ‘Sensitivity analysis of the
modeling parameters used in simulation of proton exchange membrane fuel cells’, IEEE
Transactions on Energy Conversion, Vol. 20, No. 1, pp.211–218.
 Fuel cell-based distributed generation system 93

Dai, M., Marwali, M.N., Jung, J.W. and Keyhani, A. (2008) ‘Power flow control of a
single distributed generation unit’, IEEE Transactions on Power Electronics, Vol. 23, No. 1,
pp.343–352.
El-Sharkh, M.Y., Rahman, A., Alam, M.S., Sakla, A.A., Byrne, P.C. and Thomas, T. (2004)
‘Analysis of active and reactive power control of a stand-alone PEM fuel cell power plant’,
IEEE Transactions on Power Systems, Vol. 19, No. 4, pp.2022–2028.
Jain, S.K., Agarwal, P. and Gupta, H.O. (2003) ‘Design simulation and experimental investigations
on a shunt active power filter for harmonics and reactive power compensation’, International
Journal of Electric Power Components and Systems, Vol. 31, No. 7, pp.671–692.
Kale, M. and Ozdemir, E. (2005) ‘Harmonic and reactive power compensation with shunt active
power filter under non-ideal mains voltage condition’, Journal of Electric Power Systems
Research, Vol. 74, No. 3, pp.363–370.
Kirubakaran, A., Jain, S. and Nema, R.K. (2009) ‘A review on fuel cell technologies and
power electronic interface’, Renewable and Sustainable Energy Reviews, Vol. 13, No. 9,
pp.2430–2440.
Kramer, W., Chakraborty, S., Kroposki, B. and Thomas, H. (2008) ‘Advanced power electronic
interfaces for distributed energy systems, part-1, systems and topologies’, Technical Report
NREL/TP-581-42672.
Kunjumuhammed, L.P. and Mishra, M.K. (2006) ‘A control algorithm for single-phase active
power filter under non-stiff voltage source’, IEEE Transactions on Power Electronics, Vol.
21, No. 3, pp.822–825.
Park, S., Chen, C. and Lai, J. (2008) ‘Wide range active and reactive power flow controller for a
solid oxide fuel cell power conditioning system’, in Proceedings Applied Power Electronics
Conference and Exposition, pp.952–958.
Patidar, R.D. and Singh, S.P. (2010) ‘Digital signal processor based shunt active filter controller for
customer-generated harmonics and reactive power compensation’, Journal of Electric Power
Components and System, Vol. 38, No. 8, pp.937–959.
Patidar, R.D. and Singh, S.P. (2011) ‘Performance evaluation of different line current shaping
controllers of shunt active power filters’, International Journal of Power Electronics, Vol. 3,
No. 4, pp.417–435.
Rahmani, S., Mendalek, N. and Al-Haddad, K. (2010) ‘Experimental design of a nonlinear control
technique for three-phase shunt active power filter’, IEEE Transactions on Industrial
Electronics, Vol. 57, No. 10, pp.3364–3375.
Real, A.J., Arce, A. and Bordons, C. (2007) ‘Development and experimental validation of a PEM
fuel cell dynamic model’, Journal of Power Sources, Vol. 173, No. 1, pp.310–324.
Singh, B., Al-Haddad, K. and Chandra, A. (1998) ‘A new control approach to three-phase active
filter for harmonics and reactive power compensation’, IEEE Transactions on Power Systems,
Vol. 13, No. 1, pp.133–138.
Singh, M., Khadkikar, V., Chandra, A. and Varma, R.K. (2011) ‘Grid interconnection of
renewable energy sources at the distribution level with power-quality improvement features’,
IEEE Transactions on Power Delivery, Vol. 26, No. 01, pp.307–315.
Soltani, M. and Bathaee, S.M.T. (2008) ‘A new dynamic model considering effects of temperature,
pressure and internal resistance for PEM fuel cell power modules’, in Proceedings Electric
Utility Deregulation and Restructuring and Power Technologies Conference, pp.2757–2762.
Souleman, N.M., Tremblay, O. and Dessaint, L.A. (2009) ‘A generic fuel cell model for the
simulation of fuel cell power systems’, in Proceedings of IEEE Vehicle Power and Propulsion
Conference, pp 1722–1729.
Tanrioven, M. and Alam, M.S. (2006) ‘Modeling, control, and power quality evaluation of PEM
fuel cell based power supply system for residential use’, IEEE Transactions on Industrial
Applications, Vol. 42, No. 6, pp.1582–1589.
 94 G. Mehta et al.

Teodorescu, R., Blaabjerg, F., Liserre, M. and Timbus, A.V. (2006) ‘Overview of control and grid
synchronization for distributed power generation systems’, IEEE Transactions on Industrial
Electronics, Vol. 53, No. 5, pp.1398–1409.
Wang, C., Nehrir, M.H. and Shaw, S.R. (2005) ‘Dynamic models and model validation for PEM
fuel cells using electrical circuits’, IEEE Transactions on Energy Conversion, Vol. 20, No. 2,
pp.442–451.
Wu, T., Nien, H., Shen, C. And Chen, T. (2005) ‘A single-phase inverter system for PV power
injection and active power filtering with nonlinear inductor consideration’, IEEE Transactions
on Industry Applications, Vol. 41, No. 4, pp.1075–1083.

Appendix

The value of DC-link capacitor Cdc and filter inductor Lc of the APF is estimated using
the following relations:
Vs I cT
Cdc =
Vdcref − Vdc2
2

maVdcref
Lc >
Δlsw, p − p f sw

where Vs is the rms value of grid voltage, Ic is the maximum APF current, T is grid
voltage time period, Vdc is the DC-link capacitor voltage, Vdcref is the reference DC-link
voltage, ma is the modulation index, ΔIsw,p–p is the peak to peak value of switching ripple
of the APF compensating current and fsw is the switching frequency.

View publication stats